Gecko [ www.thearyasamaj.org ]

Name	Size	Permission
domain.h	57 B	-rw-r--r--
help	151.48 KB	-rw-r--r--
help.dir	15.94 KB	-rw-r--r--
help.pag	152.22 KB	-rw-r--r--
pmda_linux.so	270.27 KB	-rwxr-xr-x
pmdalinux	270.5 KB	-rwxr-xr-x

Code Editor : help.pag

PcPh22

@ 60.0 set of all processors��
@ 60.1 set of all disks��
@ 60.2 load averages for 1, 5, and 15 minutes��
@ 60.3 set of network interfaces��
@ 60.4 set of interrupt lines��
@ 60.5 mounted block-device-backed filesystem��
@ 60.6 swap devices��
@ 60.7 network filesystem (NFS) v2 operations��
@ 60.8 network filesystem (NFS) v3 operations��
@ 60.10 set of all disk partitions��
@ 60.11 SCSI devices��
@ 60.12 kernel memory slabs��
@ 60.14 network filesystem (NFS) v4 client operations��
@ 60.15 network filesystem (NFS) v4 server operations��
@ 60.17 network interface addresses (inet and ipv6)��
@ 60.18 temporary filesystem (tempfs) mounts��
@ 60.19 non-uniform memory access (NUMA) nodes��
@ 60.23 internet control message protocol (ICMP) message types��
@ 60.24 device mapper driver devices��
@ 60.25 multi-device driver devices��
@ 60.28 IPC shm_stat shared memory IDs��
@ 60.29 IPC msg_stat message queue IDs��
@ 60.30 IPC sem_stat semaphore IDs��
@ 60.31 buddyinfo memory fragmentation sets, per-NUMA-node��
@ 60.32 zoneinfo memory types, per-NUMA-node��
@ 60.33 low memory regions for each zoneinfo memory type, per-NUMA-node��
@ 60.34 scsi tape devices, per scsi tape device�The scsi tape instance domain includes st[0-9]+ devices, but not any of the
derived devices such as nst0, nst0a, st0l, st0m and so forth. The derived
devices all share the same statistics in the kernel as the st devices.�
@ 60.35 serial devices (aka ttys)��
@ 60.36 set of soft interrupts��
@ 60.37 pressure time averages for 10 seconds, 1 minute and 5 minutes��
@ 60.38 set of compressed memory devices��
@ 60.39 fibre channel host bus adapters��
@ 60.40 per-processor IRQs��
@ 60.41 per-processor soft IRQs��
@ 60.42 scsi devices identified by unique WWID��
@ kernel.uname.release release level of the running kernel�Release level of the running kernel as reported via the release[]
value returned from uname(2) or uname -r.

See also pmda.uname.�
@ kernel.uname.version version level (build number) and build date of the running kernel�Version level of the running kernel as reported by the version[]
value returned from uname(2) or uname -v.  Usually a build number
followed by a build date.

See also pmda.uname.�
@ kernel.uname.sysname name of the implementation of the operating system�Name of the implementation of the running operating system as reported
by the sysname[] value returned from uname(2) or uname -s.  Usually
"Linux".

See also pmda.uname.�
@ kernel.uname.machine name of the hardware type the system is running on�Name of the hardware type the system is running on as reported by the machine[]
value returned from uname(2) or uname -m, e.g. "i686".

See also pmda.uname.�
@ kernel.uname.nodename host name of this node on the network�Name of this node on the network as reported by the nodename[]
value returned from uname(2) or uname -n.  Usually a synonym for
the host name.

See also pmda.uname.�
@ kernel.uname.distro Linux distribution name�The Linux distribution name, as determined by a number of heuristics.
For example:
+ on Fedora, the contents of /etc/fedora-release
+ on RedHat, the contents of /etc/redhat-release�
@ kernel.percpu.cpu.user percpu user CPU time metric from /proc/stat, including guest CPU time��
@ kernel.percpu.cpu.vuser percpu user CPU time metric from /proc/stat, excluding guest CPU time��
@ kernel.percpu.cpu.nice percpu nice user CPU time metric from /proc/stat, including guest CPU time��
@ kernel.percpu.cpu.vnice percpu nice user CPU time metric from /proc/stat, excluding guest CPU time��
@ kernel.percpu.cpu.sys percpu sys CPU time metric from /proc/stat��
@ kernel.percpu.cpu.idle percpu idle CPU time metric from /proc/stat��
@ kernel.percpu.cpu.wait.total percpu wait CPU time�Per-CPU I/O wait CPU time - time spent with outstanding I/O requests.�
@ kernel.percpu.cpu.intr percpu interrupt CPU time�Total time spent processing interrupts on each CPU (this includes
both soft and hard interrupt processing time).�
@ kernel.percpu.cpu.irq.soft percpu soft interrupt CPU time�Per-CPU soft interrupt CPU time (deferred interrupt handling code,
not run in the initial interrupt handler).�
@ kernel.percpu.cpu.irq.hard percpu hard interrupt CPU time�Per-CPU hard interrupt CPU time ("hard" interrupt handling code
is the code run directly on receipt of the initial hardware
interrupt, and does not include "soft" interrupt handling code
which is deferred until later).�
@ kernel.percpu.cpu.steal percpu CPU steal time�Per-CPU time when the CPU had a runnable process, but the hypervisor
(virtualisation layer) chose to run something else instead.�
@ kernel.percpu.cpu.guest percpu guest CPU time�Per-CPU time spent running (virtual) guest operating systems.�
@ kernel.percpu.cpu.guest_nice percpu nice guest CPU time�Per-CPU nice time spent running (virtual) guest operating systems.�
@ kernel.percpu.softirqs per CPU per soft interrupt counts from /proc/softirqs��
@ kernel.percpu.sirq per CPU aggregate soft interrupt counts from /proc/softirqs��
@ kernel.all.softirqs.total total interrupts counts from /proc/softirqs�Aggregated interrupt counts for each type of interrupt reported
by the kernel in /proc/softirqs.�
@ kernel.all.interrupts.total total interrupts counts from /proc/interrupts�Aggregated interrupt counts for each type of interrupt reported
by the kernel in /proc/interrupts.�
@ kernel.all.interrupts.missed MIS count from /proc/interrupts��
@ kernel.all.interrupts.errors interrupt error count from /proc/interrupts�This is a global counter (normally converted to a count/second)
for any and all errors that occur while handling interrupts.�
@ kernel.all.entropy.avail entropy available to random number generators��
@ kernel.all.entropy.poolsize maximum size of the entropy pool��
@ disk.dev.read per-disk read operations�Cumulative number of disk read operations since system boot time (subject
to counter wrap).�
@ disk.dev.write per-disk write operations�Cumulative number of disk write operations since system boot time (subject
to counter wrap).�
@ disk.dev.total per-disk total (read+write) operations�Cumulative number of disk read and write operations since system boot
time (subject to counter wrap).�
@ disk.dev.blkread per-disk block read operations�Cumulative number of disk block read operations since system boot time
(subject to counter wrap).�
@ disk.dev.blkwrite per-disk block write operations�Cumulative number of disk block write operations since system boot time
(subject to counter wrap).�
@ disk.dev.blktotal per-disk total (read+write) block operations�Cumulative number of disk block read and write operations since system
boot time (subject to counter wrap).�
@ disk.dev.read_bytes per-disk count of bytes read��
@ disk.dev.write_bytes per-disk count of bytes written��
@ disk.dev.total_bytes per-disk count of total bytes read and written��
@ disk.dev.scheduler per-disk I/O scheduler�The name of the I/O scheduler in use for each device.  The scheduler
is part of the block layer in the kernel, and attempts to optimise the
I/O submission patterns using various techniques (typically, sorting
and merging adjacent requests into larger ones to reduce seek activity,
but certainly not limited to that).�
@ hinv.map.scsi_id scsi disk physical device unique identifier�The unique identifier (e.g World Wide Id) of a scsi device. SCSI path
aliases to the same physical device share the same ID string, e.g.
for multipath scsi path names. This can be used to aggregate traffic
statistics to each physical device and to determine the proportion
of traffic over different paths. See also multipathd(1), where the
same ID strings (WWID) are used to identify different paths to each
physical device.�
@ disk.dev.avactive per-disk count of active time�Counts the number of milliseconds for which at least one I/O is in
progress for each device.

When converted to a rate, this metric represents the average utilization of
the disk during the sampling interval.  A value of 0.5 (or 50%) means the
disk was active (i.e. busy) half the time.�
@ disk.dev.read_rawactive per-disk raw count of read response time�For each completed read on each disk the response time (queue time plus
service time) in milliseconds is added to the associated instance of
this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding reads for a disk.  When divided by the number
of completed reads for a disk (disk.dev.read), the value represents the
stochastic average of the read response (or wait) time for that disk.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dev.r_await = delta(disk.dev.read_rawactive) / delta(disk.dev.read)�
@ disk.dev.write_rawactive per-disk raw count of write response time�For each completed write on each disk the response time (queue time plus
service time) in milliseconds is added to the associated instance of
this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding writes for a disk.  When divided by
the number of completed writes for a disk (disk.dev.write), the value
represents the stochastic average of the write response (or wait)
time for that disk.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dev.w_await = delta(disk.dev.write_rawactive) / delta(disk.dev.write)�
@ disk.dev.total_rawactive per-disk raw count of I/O response time�For each completed I/O on each disk the response time (queue time plus
service time) in milliseconds is added to the associated instance of
this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding I/Os for a disk.  When divided by the number
of completed I/Os for a disk (disk.dev.total), the value represents the
stochastic average of the I/O response (or wait) time for that disk.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dev.await = delta(disk.dev.total_rawactive) / delta(disk.dev.total)�
@ disk.dev.aveq per-disk time averaged count of request queue length�When converted to a rate, this metric represents the time averaged disk
request queue length during the sampling interval.  A value of 2.5 (or 250%)
represents a time averaged queue length of 2.5 requests during the sampling
interval.�
@ disk.dev.read_merge per-disk count of merged read requests�Count of read requests that were merged with an already queued read request.�
@ disk.dev.write_merge per-disk count of merged write requests�Count of write requests that were merged with an already queued write request.�
@ disk.dev.capacity per-disk physical device capacity�Total space presented by each block device, from /proc/partitions.�
@ disk.dev.discard per-disk discard operations�Cumulative number of disk discard operations since system boot time.�
@ disk.dev.blkdiscard per-disk block discard operations�Cumulative number of disk block discard operations since system boot time.�
@ disk.dev.discard_bytes per-disk count of bytes discard'ed��
@ disk.dev.discard_merge per-disk count of merged discard requests�Count of discard requests that were merged with an already queued discard
request.�
@ disk.dev.discard_rawactive per-disk raw count of discard response time�For each completed discard on each disk the response time (queue time plus
service time) in milliseconds is added to the associated instance of
this metric.
It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dev.d_await = delta(disk.dev.discard_rawactive) / delta(disk.dev.discard)�
@ disk.dev.flush per-disk flush operations��
@ disk.dev.flush_rawactive per-disk raw count of flush response time�For each completed flush on each disk the response time (queue time plus
service time) in milliseconds is added to the associated instance of
this metric.
It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dev.f_await = delta(disk.dev.flush_rawactive) / delta(disk.dev.flush)�
@ disk.dm.read per-device-mapper device read operations��
@ disk.dm.write per-device-mapper device write operations��
@ disk.dm.total per-device-mapper device total (read+write) operations��
@ disk.dm.blkread per-device-mapper device block read operations��
@ disk.dm.blkwrite per-device-mapper device block write operations��
@ disk.dm.blktotal per-device-mapper device total (read+write) block operations��
@ disk.dm.read_bytes per-device-mapper device count of bytes read��
@ disk.dm.write_bytes per-device-mapper device count of bytes written��
@ disk.dm.total_bytes per-device-mapper device count of total bytes read and written��
@ disk.dm.read_merge per-device-mapper device count of merged read requests��
@ disk.dm.write_merge per-device-mapper device count of merged write requests��
@ disk.dm.avactive per-device-mapper device count of active time�Counts the number of milliseconds for which at least one I/O is in
progress for each device-mapper device.

When converted to a rate, this metric represents the average utilization
of the device during the sampling interval.  A value of 0.5 (or 50%)
means the device was active (i.e. busy) half the time.�
@ disk.dm.aveq per-device-mapper device time averaged count of request queue length��
@ disk.dm.read_rawactive per-device-mapper raw count of read response time�For each completed read on each device-mapper device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding reads for a device-mapper device.
When divided by the number of completed reads for a device-mapper device
(disk.dm.read), the value represents the stochastic average of the read
response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dm.r_await = delta(disk.dm.read_rawactive) / delta(disk.dm.read)�
@ disk.dm.write_rawactive per-device-mapper raw count of write response time�For each completed write on each device-mapper device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding writes for a device-mapper device.
When divided by the number of completed writes for a device-mapper device
(disk.dm.write), the value represents the stochastic average of the
write response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dm.w_await = delta(disk.dm.write_rawactive) / delta(disk.dm.write)�
@ disk.dm.total_rawactive per-device-mapper raw count of I/O response time�For each completed I/O on each device-mapper device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding I/Os for a device-mapper device.
When divided by the number of completed I/Os for a device-mapper device
(disk.dm.total), the value represents the stochastic average of the I/O
response (or wait) time for that device-mapper device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dm.await = delta(disk.dm.total_rawactive) / delta(disk.dm.total)�
@ disk.dm.capacity per-device-mapper physical device capacity�Total space presented by each device mapper device, from /proc/partitions.�
@ disk.dm.discard per-device-mapper device discard operations��
@ disk.dm.blkdiscard per-device-mapper device block discard operations��
@ disk.dm.discard_bytes per-device-mapper device count of discard bytes��
@ disk.dm.discard_merge per-device-mapper device count of merged discard requests��
@ disk.dm.discard_rawactive per-device-mapper raw count of discard response time�For each completed discard on each device-mapper device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding discards for a device-mapper device.
When divided by the number of completed discards for a device-mapper device
(disk.dm.discards), the value represents the stochastic average of the
discard response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dm.d_await = delta(disk.dm.discard_rawactive) / delta(disk.dm.discard)�
@ disk.dm.flush per-device-mapper device flush operations��
@ disk.dm.flush_rawactive per-device-mapper raw count of flush response time�For each completed flush on each device-mapper device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding flushes for a device-mapper device.
When divided by the number of completed flushes for a device-mapper device
(disk.dm.flush), the value represents the stochastic average of the flush
response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.dm.f_await = delta(disk.dm.flush_rawactive) / delta(disk.dm.flush)�
@ hinv.map.dmname per-device-mapper device persistent name mapping to dm-[0-9]*��
@ disk.md.read per-multi-device device read operations��
@ disk.md.write per-multi-device device write operations��
@ disk.md.total per-multi-device device total (read+write) operations��
@ disk.md.blkread per-multi-device device block read operations��
@ disk.md.blkwrite per-multi-device device block write operations��
@ disk.md.blktotal per-multi-device device total (read+write) block operations��
@ disk.md.read_bytes per-multi-device device count of bytes read��
@ disk.md.write_bytes per-multi-device device count of bytes written��
@ disk.md.total_bytes per-multi-device device count of total bytes read and written��
@ disk.md.read_merge per-multi-device device count of merged read requests��
@ disk.md.write_merge per-multi-device device count of merged write requests��
@ disk.md.avactive per-multi-device device count of active time�Counts the number of milliseconds for which at least one I/O is in
progress for each multi-device device.

When converted to a rate, this metric represents the average utilization
of the device during the sampling interval.  A value of 0.5 (or 50%)
means the device was active (i.e. busy) half the time.�
@ disk.md.aveq per-multi-device device time averaged count of request queue length��
@ disk.md.read_rawactive per-multi-device raw count of read response time�For each completed read on each multi-device device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding reads for a multi-device device.
When divided by the number of completed reads for a multi-device device
(disk.md.read), the value represents the stochastic average of the read
response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.md.r_await = delta(disk.md.read_rawactive) / delta(disk.md.read)�
@ disk.md.write_rawactive per-multi-device raw count of write response time�For each completed write on each multi-device device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding writes for a multi-device device.
When divided by the number of completed writes for a multi-device device
(disk.md.write), the value represents the stochastic average of the
write response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.md.w_await = delta(disk.md.write_rawactive) / delta(disk.md.write)�
@ disk.md.total_rawactive per-multi-device raw count of I/O response time�For each completed I/O on each multi-device device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding I/Os for a multi-device device.
When divided by the number of completed I/Os for a multi-device device
(disk.md.total), the value represents the stochastic average of the I/O
response (or wait) time for that multi-device device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.md.await = delta(disk.md.total_rawactive) / delta(disk.md.total)�
@ disk.md.status per-multi-device "mdadm --test --detail <device>" return code��
@ disk.md.capacity per-multi-device device capacity�Total space presented by each multi-device device, from /proc/partitions.�
@ disk.md.discard per-multi-device device discard operations��
@ disk.md.blkdiscard per-multi-device device block discard operations��
@ disk.md.discard_bytes per-multi-device device count of discard bytes��
@ disk.md.discard_merge per-multi-device device count of merged discard requests��
@ disk.md.discard_rawactive per-multi-device raw count of discard response time�For each completed discard on each multi-device device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding discards for a multi-device device.
When divided by the number of completed discards for a multi-device device
(disk.md.discards), the value represents the stochastic average of the
discard response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.md.d_await = delta(disk.md.discard_rawactive) / delta(disk.md.discard)�
@ disk.md.flush per-multi-device device flush operations��
@ disk.md.flush_rawactive per-multi-device raw count of flush response time�For each completed flush on each multi-device device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding flushes for a multi-device device.
When divided by the number of completed flushes for a multi-device device
(disk.md.flush), the value represents the stochastic average of the flush
response (or wait) time for that device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.md.f_await = delta(disk.md.flush_rawactive) / delta(disk.md.flush)�
@ hinv.map.mdname per-multi-device device persistent name mapping to md[0-9]*��
@ disk.all.read_merge total count of merged read requests, summed for all disks�Total count of read requests that were merged with an already queued read request.�
@ disk.all.write_merge total count of merged write requests, summed for all disks�Total count of write requests that were merged with an already queued write request.�
@ disk.all.avactive total count of active time, summed for all disks�Counts the number of milliseconds for which at least one I/O is in
progress on each disk, summed across all disks.

When converted to a rate and divided by the number of disks (hinv.ndisk),
this metric represents the average utilization of all disks during the
sampling interval.  A value of 0.25 (or 25%) means that on average every
disk was active (i.e. busy) one quarter of the time.�
@ disk.all.read_rawactive raw count of read response time, summed for all disks�For each completed read on every disk the response time (queue time plus
service time) in milliseconds is added to this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding reads across all disks.  When divided
by the number of completed reads for all disks (disk.all.read), value
represents the stochastic average of the read response (or wait) time
across all disks.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.all.r_await = delta(disk.all.read_rawactive) / delta(disk.all.read)�
@ disk.all.write_rawactive raw count of write response time, summed for all disks�For each completed write on every disk the response time (queue time
plus service time) in milliseconds is added to this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding writes across all disks.  When divided by
the number of completed writes for all disks (disk.all.write), value
represents the stochastic average of the write response (or wait) time
across all disks.�
@ disk.all.total_rawactive raw count of I/O response time, summed for all disks�For each completed I/O on every disk the response time (queue time
plus service time) in milliseconds is added to this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding I/Os across all disks.  When divided by
the number of completed I/Os for all disks (disk.all.total), value
represents the stochastic average of the I/O response (or wait) time
across all disks.�
@ disk.all.aveq total time averaged count of request queue length, summed for all disks�When converted to a rate, this metric represents the average across all disks
of the time averaged request queue length during the sampling interval.
A value of 1.5 (or 150%) suggests that (on average) over all disks there was
a time averaged queue length of 1.5 requests during the sampling interval.�
@ disk.all.read total read operations, summed for all disks�Cumulative number of disk read operations since system boot time
(subject to counter wrap), summed over all disk devices.�
@ disk.all.write total write operations, summed for all disks�Cumulative number of disk read operations since system boot time
(subject to counter wrap), summed over all disk devices.�
@ disk.all.total total (read+write) operations, summed for all disks�Cumulative number of disk read and write operations since system boot
time (subject to counter wrap), summed over all disk devices.�
@ disk.all.blkread block read operations, summed for all disks�Cumulative number of disk block read operations since system boot time
(subject to counter wrap), summed over all disk devices.�
@ disk.all.blkwrite block write operations, summed for all disks�Cumulative number of disk block write operations since system boot time
(subject to counter wrap), summed over all disk devices.�
@ disk.all.blktotal total (read+write) block operations, summed for all disks�Cumulative number of disk block read and write operations since system
boot time (subject to counter wrap), summed over all disk devices.�
@ disk.all.read_bytes count of bytes read for all disk devices��
@ disk.all.write_bytes count of bytes written for all disk devices��
@ disk.all.total_bytes total count of bytes read and written for all disk devices��
@ disk.all.discard total discard operations, summed for all disks��
@ disk.all.blkdiscard block discard operations, summed for all disks��
@ disk.all.discard_bytes count of discard bytes for all disk devices��
@ disk.all.discard_merge total count of merged discard requests, summed for all disks�Total count of discard requests that were merged with an already queued
discard request.�
@ disk.all.discard_rawactive raw count of discard response time, summed for all disks�For each discard on every disk the response time (queue time plus
service time) in milliseconds is added to this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding discards across all disks.  When divided
by the number of completed discards for all disks (disk.all.discard), the
value represents the stochastic average of the discard response (or wait)
time across all disks.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.all.d_await = delta(disk.all.discard_rawactive) / delta(disk.all.discard)�
@ disk.all.flush total flush operations, summed for all disks��
@ disk.all.flush_rawactive raw count of flush response time, summed for all disks�For each flush on every disk the response time (queue time plus
service time) in milliseconds is added to this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding flushes across all disks.  When divided
by the number of completed flushes for all disks (disk.all.flush), the
value represents the stochastic average of the flush response (or wait)
time across all disks.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.all.f_await = delta(disk.all.flush_rawactive) / delta(disk.all.flush)�
@ disk.partitions.read read operations metric for storage partitions�Cumulative number of disk read operations since system boot time
(subject to counter wrap) for individual disk partitions or logical
volumes.�
@ disk.partitions.write write operations metric for storage partitions�Cumulative number of disk write operations since system boot time
(subject to counter wrap) for individual disk partitions or logical
volumes.�
@ disk.partitions.total total (read+write) I/O operations metric for storage partitions�Cumulative number of disk read and write operations since system boot
time (subject to counter wrap) for individual disk partitions or
logical volumes.�
@ disk.partitions.blkread block read operations metric for storage partitions�Cumulative number of disk block read operations since system boot time
(subject to counter wrap) for individual disk partitions or logical
volumes.�
@ disk.partitions.blkwrite block write operations metric for storage partitions�Cumulative number of disk block write operations since system boot time
(subject to counter wrap) for individual disk partitions or logical
volumes.�
@ disk.partitions.blktotal total (read+write) block operations metric for storage partitions�Cumulative number of disk block read and write operations since system
boot time (subject to counter wrap) for individual disk partitions or
logical volumes.�
@ disk.partitions.read_bytes number of bytes read for storage partitions�Cumulative number of bytes read since system boot time (subject to
counter wrap) for individual disk partitions or logical volumes.�
@ disk.partitions.write_bytes number of bytes written for storage partitions�Cumulative number of bytes written since system boot time (subject to
counter wrap) for individual disk partitions or logical volumes.�
@ disk.partitions.total_bytes total number of bytes read and written for storage partitions�Cumulative number of bytes read and written since system boot time
(subject to counter wrap) for individual disk partitions or logical
volumes.�
@ disk.partitions.read_merge per-disk-partition count of merged read requests��
@ disk.partitions.write_merge per-disk-partition count of merged write requests��
@ disk.partitions.avactive per-disk-partition device count of active time�Counts the number of milliseconds for which at least one I/O is in
progress for each disk partition.

When converted to a rate, this metric represents the average utilization
of the disk partition during the sampling interval.  A value of 0.5
(or 50%) means the disk partition was active (i.e. busy) half the time.�
@ disk.partitions.aveq per-disk-partition device time averaged count of request queue length��
@ disk.partitions.read_rawactive per-disk-partition raw count of read response time�For each completed read on each disk partition the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding reads for a disk partition.  When divided by
the number of completed reads for a disk partition (disk.partitions.read),
the value represents the stochastic average of the read response (or wait)
time for that disk partition.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.partitions.r_await = delta(disk.partitions.read_rawactive) /
                             delta(disk.partitions.read)�
@ disk.partitions.write_rawactive per-disk-partition raw count of write response time�For each completed write on each disk partition the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding writes for a disk partition.
When divided by the number of completed writes for a disk partition
(disk.partitions.write), the value represents the stochastic average of
the write response (or wait) time for that disk partition.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.partitions.w_await = delta(disk.partitions.write_rawactive) /
                             delta(disk.partitions.write)�
@ disk.partitions.total_rawactive per-disk-partition raw count of I/O response time�For each completed I/O on each disk partition the response time (queue
time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding I/Os for a disk partition.
When divided by the number of completed I/Os for a disk partition
(disk.partitions.total), the value represents the stochastic average of
the I/O response (or wait) time for that disk partition.�
@ disk.partitions.capacity per-disk-partition capacity�Total space presented by each disk partition, from /proc/partitions.�
@ disk.partitions.discard discard operations count for storage partitions��
@ disk.partitions.blkdiscard block discard operations count for storage partitions��
@ disk.partitions.discard_bytes number of discard bytes for storage partitions��
@ disk.partitions.discard_merge per-disk-partition count of merged discard requests��
@ disk.partitions.discard_rawactive per-disk-partition raw count of discard response time�For each completed discard on each disk partition the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding discards for a disk partition.  When divided by
the number of completed discards for a disk partition (disk.partitions.discard),
the value represents the stochastic average of the discard response (or wait)
time for that disk partition.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.partitions.d_await = delta(disk.partitions.discard_rawactive) /
                             delta(disk.partitions.discard)�
@ disk.partitions.flush flush operations metric for storage partitions��
@ disk.partitions.flush_rawactive per-disk-partition raw count of flush response time�For each completed flush on each disk partition the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding flushes for a disk partition.  When divided by
the number of completed flushes for a disk partition (disk.partitions.flush),
the value represents the stochastic average of the flush response (or wait)
time for that disk partition.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.partitions.f_await = delta(disk.partitions.flush_rawactive) /
                             delta(disk.partitions.flush)�
@ swap.pagesin pages read from swap devices due to demand for physical memory��
@ swap.pagesout pages written to swap devices due to demand for physical memory��
@ swap.in number of swap in operations��
@ swap.out number of swap out operations��
@ kernel.all.pswitch context switches metric from /proc/stat��
@ kernel.all.sysfork fork rate metric from /proc/stat��
@ kernel.all.running number of currently running processes from /proc/stat��
@ kernel.all.blocked number of currently blocked processes from /proc/stat��
@ kernel.all.boottime boot time from /proc/stat��
@ kernel.all.intr interrupt count metric from /proc/stat�The value is the first value from the intr field in /proc/stat,
which is a counter of the total number of interrupts processed.
The value is normally converted to a rate (count/second).
This counter usually increases by at least HZ/second,
i.e. the clock interrupt rate, wehich is usually 100/second.

See also kernel.percpu.intr and kernel.percpu.interrupts to get
the breakdown of interrupt count by interrupt type and which CPU
processed each one.�
@ kernel.percpu.interrupts interrupt counters from /proc/interrupts�Counts of each interrupt type, for each CPU.�
@ kernel.percpu.intr interrupt count metric from /proc/interrupts�Aggregate count of each CPUs interrupt processing count, calculated
as the sum of all interrupt types in /proc/interrupts for each CPU.�
@ mem.physmem total system memory metric reported by /proc/meminfo�The value of this metric corresponds to the "MemTotal" field
reported by /proc/meminfo. Note that this does not necessarily
correspond to actual installed physical memory - there may
be areas of the physical address space mapped as ROM in
various peripheral devices and the bios may be mirroring
certain ROMs in RAM.�
@ mem.freemem free system memory metric from /proc/meminfo��
@ mem.buddyinfo.pages fragmented page count from /proc/buddyinfo��
@ mem.buddyinfo.total page fragmentation size from /proc/buddyinfo��
@ mem.util.used used memory metric from /proc/meminfo�Used memory is the difference between mem.physmem and mem.freemem.�
@ mem.util.free free memory metric from /proc/meminfo�Alias for mem.freemem.�
@ mem.util.available available memory from /proc/meminfo�The amount of memory that is available for a new workload,
without pushing the system into swap. Estimated from MemFree,
Active(file), Inactive(file), and SReclaimable, as well as the "low"
watermarks from /proc/zoneinfo.�
@ mem.util.shared shared memory metric from /proc/meminfo�Shared memory metric. Currently always zero on Linux 2.4 kernels
and has been removed from 2.6 kernels.�
@ mem.util.bufmem I/O buffers metric from /proc/meminfo�Memory allocated for buffer_heads.�
@ mem.util.cached page cache metric from /proc/meminfo�Memory used by the page cache, including buffered file data.
This is in-memory cache for files read from the disk (the pagecache)
but doesn't include SwapCached.�
@ mem.util.other unaccounted memory�Memory that is not free (i.e. has been referenced) and is not cached.
mem.physmem - mem.util.free - mem.util.cached - mem.util.buffers�
@ mem.util.active Kbytes on the active page list (recently referenced pages)�Memory that has been used more recently and usually not reclaimed unless
absolutely necessary.�
@ mem.util.inactive Kbytes on the inactive page list (candidates for discarding)�Memory which has been less recently used.  It is more eligible to be
reclaimed for other purposes�
@ mem.util.swapCached Kbytes in swap cache, from /proc/meminfo�Memory that once was swapped out, is swapped back in but still also
is in the swapfile (if memory is needed it doesn't need to be swapped
out AGAIN because it is already in the swapfile. This saves I/O)�
@ mem.util.highTotal Kbytes in high memory, from /proc/meminfo�This is apparently an i386 specific metric, and seems to be always zero
on ia64 architecture (and possibly others). On i386 arch (at least),
highmem is all memory above ~860MB of physical memory. Highmem areas
are for use by userspace programs, or for the pagecache. The kernel
must use tricks to access this memory, making it slower to access
than lowmem.�
@ mem.util.highFree Kbytes free high memory, from /proc/meminfo�See mem.util.highTotal. Not used on ia64 arch (and possibly others).�
@ mem.util.lowTotal Kbytes in low memory total, from /proc/meminfo�Lowmem is memory which can be used for everything that highmem can be
used for, but it is also availble for the kernel's use for its own
data structures. Among many other things, it is where everything
from the Slab is allocated.  Bad things happen when you're out of lowmem.
(this may only be true on i386 architectures).�
@ mem.util.lowFree Kbytes free low memory, from /proc/meminfo�See mem.util.lowTotal�
@ mem.util.swapTotal Kbytes swap, from /proc/meminfo�total amount of swap space available�
@ mem.util.swapFree Kbytes free swap, from /proc/meminfo�Memory which has been evicted from RAM, and is temporarily on the disk�
@ mem.util.dirty Kbytes in dirty pages, from /proc/meminfo�Memory which is waiting to get written back to the disk�
@ mem.util.writeback Kbytes in writeback pages, from /proc/meminfo�Memory which is actively being written back to the disk�
@ mem.util.mapped Kbytes in mapped pages, from /proc/meminfo�files which have been mmaped, such as libraries�
@ mem.util.slab Kbytes in slab memory, from /proc/meminfo�in-kernel data structures cache�
@ mem.util.commitLimit Kbytes limit for address space commit, from /proc/meminfo�The static total, in Kbytes, available for commitment to address
spaces. Thus, mem.util.committed_AS may range up to this total. Normally
the kernel overcommits memory, so this value may exceed mem.physmem�
@ mem.util.committed_AS Kbytes committed to address spaces, from /proc/meminfo�An estimate of how much RAM you would need to make a 99.99% guarantee
that there never is OOM (out of memory) for this workload. Normally
the kernel will overcommit memory. That means, say you do a 1GB malloc,
nothing happens, really. Only when you start USING that malloc memory
you will get real memory on demand, and just as much as you use.�
@ mem.util.pageTables Kbytes in kernel page tables, from /proc/meminfo��
@ mem.util.reverseMaps Kbytes in reverse mapped pages, from /proc/meminfo��
@ mem.util.cache_clean Kbytes cached and not dirty or writeback, derived from /proc/meminfo��
@ mem.util.anonpages Kbytes in user pages not backed by files, from /proc/meminfo��
@ mem.util.bounce Kbytes in bounce buffers, from /proc/meminfo��
@ mem.util.NFS_Unstable Kbytes in NFS unstable memory, from /proc/meminfo��
@ mem.util.slabReclaimable Kbytes in reclaimable slab pages, from /proc/meminfo��
@ mem.util.slabUnreclaimable Kbytes in unreclaimable slab pages, from /proc/meminfo��
@ mem.util.active_anon anonymous Active list LRU memory��
@ mem.util.inactive_anon anonymous Inactive list LRU memory��
@ mem.util.active_file file-backed Active list LRU memory��
@ mem.util.inactive_file file-backed Inactive list LRU memory��
@ mem.util.unevictable kbytes of memory that is unevictable��
@ mem.util.mlocked kbytes of memory that is pinned via mlock()��
@ mem.util.shmem kbytes of shmem��
@ mem.util.kernelStack kbytes of memory used for kernel stacks��
@ mem.util.hugepagesTotal a count of total hugepages��
@ mem.util.hugepagesFree a count of free hugepages��
@ mem.util.hugepagesRsvd a count of reserved hugepages��
@ mem.util.hugepagesSurp a count of surplus hugepages��
@ mem.util.directMap4k amount of memory that is directly mapped in 4kB pages��
@ mem.util.directMap2M amount of memory that is directly mapped in 2MB pages��
@ mem.util.directMap1G amount of memory that is directly mapped in 1GB pages��
@ mem.util.vmallocTotal amount of kernel memory allocated via vmalloc��
@ mem.util.vmallocUsed amount of used vmalloc memory��
@ mem.util.vmallocChunk amount of vmalloc chunk memory��
@ mem.util.mmap_copy amount of mmap_copy space (non-MMU kernels only)��
@ mem.util.quicklists amount of memory in the per-CPU quicklists��
@ mem.util.corrupthardware amount of memory in hardware corrupted pages��
@ mem.util.anonhugepages amount of memory in anonymous huge pages��
@ mem.util.hugepagesTotalBytes amount of memory in total hugepages��
@ mem.util.hugepagesFreeBytes amount of memory in free hugepages��
@ mem.util.hugepagesRsvdBytes amount of memory in reserved hugepages��
@ mem.util.hugepagesSurpBytes amount of memory in surplus hugepages�
User memory (Kbytes) in pages not backed by files, e.g. from malloc()�
@ mem.numa.util.total per-node total memory��
@ mem.numa.util.free per-node free memory��
@ mem.numa.util.used per-node used memory��
@ mem.numa.util.active per-node Active list LRU memory��
@ mem.numa.util.inactive per-node Inactive list LRU memory��
@ mem.numa.util.active_anon per-node anonymous Active list LRU memory��
@ mem.numa.util.inactive_anon per-node anonymous Inactive list LRU memory��
@ mem.numa.util.active_file per-node file-backed Active list LRU memory��
@ mem.numa.util.inactive_file per-node file-backed Inactive list LRU memory��
@ mem.numa.util.highTotal per-node highmem total��
@ mem.numa.util.highFree per-node highmem free��
@ mem.numa.util.lowTotal per-node lowmem total��
@ mem.numa.util.lowFree per-node lowmem free��
@ mem.numa.util.unevictable per-node Unevictable memory��
@ mem.numa.util.mlocked per-node count of Mlocked memory��
@ mem.numa.util.dirty per-node dirty memory��
@ mem.numa.util.writeback per-node count of memory locked for writeback to stable storage��
@ mem.numa.util.filePages per-node count of memory backed by files��
@ mem.numa.util.mapped per-node mapped memory��
@ mem.numa.util.anonpages per-node anonymous memory��
@ mem.numa.util.shmem per-node amount of shared memory��
@ mem.numa.util.kernelStack per-node memory used as kernel stacks��
@ mem.numa.util.pageTables per-node memory used for pagetables��
@ mem.numa.util.NFS_Unstable per-node memory holding NFS data that needs writeback��
@ mem.numa.util.bounce per-node memory used for bounce buffers��
@ mem.numa.util.writebackTmp per-node temporary memory used for writeback��
@ mem.numa.util.slab per-node memory used for slab objects��
@ mem.numa.util.slabReclaimable per-node memory used for slab objects that can be reclaimed��
@ mem.numa.util.slabUnreclaimable per-node memory used for slab objects that is unreclaimable��
@ mem.numa.util.hugepagesTotal per-node total count of hugepages��
@ mem.numa.util.hugepagesFree per-node count of free hugepages��
@ mem.numa.util.hugepagesSurp per-node count of surplus hugepages��
@ mem.numa.alloc.hit per-node count of times a task wanted alloc on local node and succeeded��
@ mem.numa.alloc.miss per-node count of times a task wanted alloc on local node but got another node��
@ mem.numa.alloc.foreign count of times a task on another node alloced on that node, but got this node��
@ mem.numa.alloc.interleave_hit count of times interleaving wanted to allocate on this node and succeeded��
@ mem.numa.alloc.local_node count of times a process ran on this node and got memory on this node��
@ mem.numa.alloc.other_node count of times a process ran on this node and got memory from another node��
@ mem.numa.max_bandwidth maximum memory bandwidth supported on each numa node�Maximum memory bandwidth supported on each numa node. It makes use of a
bandwith.conf file which has the bandwidth information for each node :
node_num:bandwidth
The node_num must match with any node in sysfs/devices/system/node directory.
And, the bandwidth is expressed in terms of MBps. This config file should be
filled up manually after running some bandwidth saturation benchmark tools.�
@ mem.numa.util.hugepagesTotalBytes per-node total amount of hugepages memory��
@ mem.numa.util.hugepagesFreeBytes per-node amount of free hugepages memory��
@ mem.numa.util.hugepagesSurpBytes per-node amount of surplus hugepages memory��
@ mem.vmstat.nr_dirty number of pages in dirty state�Instantaneous number of pages in dirty state, from /proc/vmstat�
@ mem.vmstat.nr_dirty_background_threshold background writeback threshold��
@ mem.vmstat.nr_dirty_threshold dirty throttling threshold��
@ mem.vmstat.nr_dirtied count of pages dirtied�Count of pages entering dirty state, from /proc/vmstat�
@ mem.vmstat.nr_file_hugepages count of hugepages used for files��
@ mem.vmstat.nr_file_pages count of pages used for files��
@ mem.vmstat.nr_file_pmdmapped count of PMD mappings used for files��
@ mem.vmstat.nr_foll_pin_acquired count of pinned user memory acquisitions��
@ mem.vmstat.nr_foll_pin_released count of pinned user memory releases��
@ mem.vmstat.nr_writeback number of pages in writeback state�Instantaneous number of pages in writeback state, from /proc/vmstat�
@ mem.vmstat.nr_unstable number of pages in unstable state�Instantaneous number of pages in unstable state, from /proc/vmstat�
@ mem.vmstat.nr_page_table_pages number of page table pages�Instantaneous number of page table pages, from /proc/vmstat�
@ mem.vmstat.nr_mapped number of mapped pagecache pages�Instantaneous number of mapped pagecache pages, from /proc/vmstat
See also mem.vmstat.nr_anon for anonymous mapped pages.�
@ mem.vmstat.nr_slab number of slab pages�Instantaneous number of slab pages, from /proc/vmstat
This counter was retired in 2.6.18 kernels, and is now the sum of
mem.vmstat.nr_slab_reclaimable and mem.vmstat.nr_slab_unreclaimable.�
@ mem.vmstat.nr_written count of pages written out�Count of pages written out, from /proc/vmstat�
@ mem.vmstat.numa_foreign count of foreign NUMA zone allocations��
@ mem.vmstat.numa_hit count of successful allocations from preferred NUMA zone��
@ mem.vmstat.numa_interleave count of interleaved NUMA allocations��
@ mem.vmstat.numa_local count of successful allocations from local NUMA zone��
@ mem.vmstat.numa_miss count of unsuccessful allocations from preferred NUMA zona��
@ mem.vmstat.numa_other count of unsuccessful allocations from local NUMA zone��
@ mem.vmstat.pgpgin page in operations�Count of page in operations since boot, from /proc/vmstat�
@ mem.vmstat.pgpgout page out operations�Count of page out operations since boot, from /proc/vmstat�
@ mem.vmstat.pswpin pages swapped in�Count of pages swapped in since boot, from /proc/vmstat�
@ mem.vmstat.pswpout pages swapped out�Count of pages swapped out since boot, from /proc/vmstat�
@ mem.vmstat.pgalloc_high high mem page allocations�Count of high mem page allocations since boot, from /proc/vmstat�
@ mem.vmstat.pgalloc_normal normal mem page allocations�Count of normal mem page allocations since boot, from /proc/vmstat�
@ mem.vmstat.pgalloc_dma dma mem page allocations�Count of dma mem page allocations since boot, from /proc/vmstat�
@ mem.vmstat.pgalloc_dma32 dma32 mem page allocations�Count of dma32 mem page allocations since boot, from /proc/vmstat�
@ mem.vmstat.pgalloc_movable movable mem page allocations�Count of movable mem page allocations since boot, from /proc/vmstat�
@ mem.vmstat.pgfree page free operations�Count of page free operations since boot, from /proc/vmstat�
@ mem.vmstat.pgactivate pages moved from inactive to active�Count of pages moved from inactive to active since boot, from /proc/vmstat�
@ mem.vmstat.pgdeactivate pages moved from active to inactive�Count of pages moved from active to inactive since boot, from /proc/vmstat�
@ mem.vmstat.pgfault page major and minor fault operations�Count of page major and minor fault operations since boot, from /proc/vmstat�
@ mem.vmstat.pgmajfault major page fault operations�Count of major page fault operations since boot, from /proc/vmstat�
@ mem.vmstat.pgrefill_high high mem pages inspected in refill_inactive_zone�Count of high mem pages inspected in refill_inactive_zone since boot,
from /proc/vmstat�
@ mem.vmstat.pgrefill_normal normal mem pages inspected in refill_inactive_zone�Count of normal mem pages inspected in refill_inactive_zone since boot,
from /proc/vmstat�
@ mem.vmstat.pgrefill_dma dma mem pages inspected in refill_inactive_zone�Count of dma mem pages inspected in refill_inactive_zone since boot,
from /proc/vmstat�
@ mem.vmstat.pgrefill_dma32 dma32 mem pages inspected in refill_inactive_zone�Count of dma32 mem pages inspected in refill_inactive_zone since boot,
from /proc/vmstat�
@ mem.vmstat.pgrefill_movable movable mem pages inspected in refill_inactive_zone�Count of movable mem pages inspected in refill_inactive_zone since boot,
from /proc/vmstat�
@ mem.vmstat.pgsteal_direct mem pages directly reclaimed�Count of mem pages directly reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_kswapd mem pages reclaimed by kswapd�Count of mem pages reclaimed by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_file file memory pages reclaimed�Count of file memory pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_high high mem pages reclaimed�Count of high mem pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_normal normal mem pages reclaimed�Count of normal mem pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_dma dma mem pages reclaimed�Count of dma mem pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_dma32 dma32 mem pages reclaimed�Count of dma32 mem pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_movable movable mem pages reclaimed�Count of movable mem pages reclaimed since boot, from /proc/vmstat�
@ mem.vmstat.pgsteal_kswapd_dma dma mem pages reclaimed by kswapd�Count of dma mem pages reclaimed by kswapd since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_kswapd_dma32 dma32 mem pages reclaimed by kswapd�Count of dma32 mem pages reclaimed by kswapd since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_kswapd_normal normal mem pages reclaimed by kswapd�Count of normal mem pages reclaimed by kswapd since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_kswapd_movable movable mem pages reclaimed by kswapd�Count of movable mem pages reclaimed by kswapd since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_direct_dma dma mem pages reclaimed�Count of dma mem pages reclaimed since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_direct_dma32 dma32 mem pages reclaimed�Count of dma32 mem pages reclaimed since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_direct_normal normal mem pages reclaimed�Count of normal mem pages reclaimed since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgsteal_direct_movable movable mem pages reclaimed�Count of movable mem pages reclaimed since boot, from /proc/vmstat
since Linux 2.6.23�
@ mem.vmstat.pgsteal_total cumulative total mem pages reclaimed��
@ mem.vmstat.pgscan_direct directly scanned mem pages�Count of mem pages scanned directly since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgscan_direct_throttle throttled direct scanned mem pages�Count of throttled mem pages scanned directly since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgscan_direct_total cumulative total of directly scanned mem pages��
@ mem.vmstat.pgscan_kswapd mem pages scanned by kswapd�Count of mem pages scanned by kswapd since boot, from /proc/vmstat
since Linux 3.4�
@ mem.vmstat.pgscan_kswapd_high high mem pages scanned by kswapd�Count of high mem pages scanned by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_kswapd_normal normal mem pages scanned by kswapd�Count of normal mem pages scanned by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_kswapd_dma dma mem pages scanned by kswapd�Count of dma mem pages scanned by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_kswapd_dma32 dma32 mem pages scanned by kswapd�Count of dma32 mem pages scanned by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_kswapd_movable movable mem pages scanned by kswapd�Count of movable mem pages scanned by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_kswapd_total cumulative total mem pages scanned by kswapd��
@ mem.vmstat.pgscan_direct_high high mem pages scanned�Count of high mem pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_direct_normal normal mem pages scanned�Count of normal mem pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_direct_dma dma mem pages scanned�Count of dma mem pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_direct_dma32 dma32 mem pages scanned�Count of dma32 mem pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.pgscan_direct_movable movable mem pages scanned�Count of movable mem pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.pginodesteal pages reclaimed via inode freeing�Count of pages reclaimed via inode freeing since boot, from /proc/vmstat�
@ mem.vmstat.slabs_scanned slab pages scanned�Count of slab pages scanned since boot, from /proc/vmstat�
@ mem.vmstat.swap_ra count of swap pages readahead��
@ mem.vmstat.swap_ra_hit count of swap pages readahead that were used��
@ mem.vmstat.kswapd_steal pages reclaimed by kswapd�Count of pages reclaimed by kswapd since boot, from /proc/vmstat�
@ mem.vmstat.kswapd_low_wmark_hit_quickly count of times low watermark reached quickly�Count of times kswapd reached low watermark quickly, from /proc/vmstat�
@ mem.vmstat.kswapd_high_wmark_hit_quickly count of times high watermark reached quickly�Count of times kswapd reached high watermark quickly, from /proc/vmstat�
@ mem.vmstat.kswapd_skip_congestion_wait count of times kswapd skipped waiting on device congestion�Count of times kswapd skipped waiting due to device congestion as a
result of being under the low watermark, from /proc/vmstat�
@ mem.vmstat.kswapd_inodesteal pages reclaimed via kswapd inode freeing�Count of pages reclaimed via kswapd inode freeing since boot, from
/proc/vmstat�
@ mem.vmstat.pageoutrun kswapd calls to page reclaim�Count of kswapd calls to page reclaim since boot, from /proc/vmstat�
@ mem.vmstat.allocstall direct reclaim calls�Count of direct reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.allocstall_dma direct dma memory reclaim calls�Count of direct dma reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.allocstall_dma32 direct dma32 memory reclaim calls�Count of direct dma32 reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.allocstall_high direct high memory reclaim calls�Count of direct high mem reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.allocstall_movable direct movable memory reclaim calls�Count of direct movable mem reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.allocstall_normal direct normal memory reclaim calls�Count of direct normal mem reclaim calls since boot, from /proc/vmstat�
@ mem.vmstat.pgrotated pages rotated to tail of the LRU�Count of pages rotated to tail of the LRU since boot, from /proc/vmstat�
@ mem.vmstat.nr_anon_pages number of anonymous mapped pagecache pages�Instantaneous number of anonymous mapped pagecache pages, from /proc/vmstat
See also mem.vmstat.mapped for other mapped pages.�
@ mem.vmstat.nr_anon_transparent_hugepages number of anonymous transparent huge pages�Instantaneous number of anonymous transparent huge pages, from /proc/vmstat�
@ mem.vmstat.nr_bounce number of bounce buffer pages�Instantaneous number of bounce buffer pages, from /proc/vmstat�
@ mem.vmstat.nr_slab_reclaimable reclaimable slab pages�Instantaneous number of reclaimable slab pages, from /proc/vmstat.�
@ mem.vmstat.nr_slab_unreclaimable unreclaimable slab pages�Instantaneous number of unreclaimable slab pages, from /proc/vmstat.�
@ mem.vmstat.nr_vmscan_write pages written by VM scanner from LRU�Count of pages written from the LRU by the VM scanner, from /proc/vmstat.
The VM is supposed to minimise the number of pages which get written
from the LRU (for IO scheduling efficiency, and for high reclaim-success
rates).�
@ mem.vmstat.htlb_buddy_alloc_fail huge TLB page buddy allocation failures�Count of huge TLB page buddy allocation failures, from /proc/vmstat�
@ mem.vmstat.htlb_buddy_alloc_success huge TLB page buddy allocation successes�Count of huge TLB page buddy allocation successes, from /proc/vmstat�
@ mem.vmstat.nr_active_anon number of active anonymous memory pages��
@ mem.vmstat.nr_active_file number of active file memory memory pages��
@ mem.vmstat.nr_file_pages number of file pages��
@ mem.vmstat.nr_free_pages number of free pages��
@ mem.vmstat.nr_inactive_anon number of inactive anonymous memory pages��
@ mem.vmstat.nr_inactive_file number of inactive file memory pages��
@ mem.vmstat.nr_isolated_anon number of isolated anonymous memory pages��
@ mem.vmstat.nr_isolated_file number of isolated file memory pages��
@ mem.vmstat.nr_kernel_misc_reclaimable number of misc reclaimable kernel pages��
@ mem.vmstat.nr_kernel_stack number of pages of kernel stack��
@ mem.vmstat.nr_mlock number of pages under mlock��
@ mem.vmstat.nr_shmem number of shared memory pages��
@ mem.vmstat.nr_unevictable number of unevictable pages��
@ mem.vmstat.nr_writeback_temp number of temporary writeback pages��
@ mem.vmstat.compact_blocks_moved count of compact blocks moved��
@ mem.vmstat.compact_fail count of unsuccessful compactions for high order allocations��
@ mem.vmstat.compact_pagemigrate_failed count of pages unsuccessfully compacted��
@ mem.vmstat.compact_pages_moved count of pages successfully moved for compaction��
@ mem.vmstat.compact_stall count of failures to even start compacting��
@ mem.vmstat.compact_success count of successful compactions for high order allocations��
@ mem.vmstat.thp_fault_alloc transparent huge page fault allocations��
@ mem.vmstat.thp_fault_fallback transparent huge page fault fallbacks��
@ mem.vmstat.thp_fault_fallback_charge transparent huge page fault fallback charges��
@ mem.vmstat.thp_file_alloc transparent huge page file allocations��
@ mem.vmstat.thp_file_mapped transparent huge page file mappings��
@ mem.vmstat.thp_collapse_alloc transparent huge page collapse allocations��
@ mem.vmstat.thp_collapse_alloc_failed transparent huge page collapse failures��
@ mem.vmstat.thp_split count of transparent huge page splits��
@ mem.vmstat.thp_zero_page_alloc count of transparent huge page zeroed page allocations��
@ mem.vmstat.thp_zero_page_alloc_failed count of transparent huge page zeroed page allocation failures��
@ mem.vmstat.unevictable_pgs_cleared count of unevictable pages cleared��
@ mem.vmstat.unevictable_pgs_culled count of unevictable pages culled��
@ mem.vmstat.unevictable_pgs_mlocked count of mlocked unevictable pages��
@ mem.vmstat.unevictable_pgs_mlockfreed count of unevictable pages mlock freed��
@ mem.vmstat.unevictable_pgs_munlocked count of unevictable pages munlocked��
@ mem.vmstat.unevictable_pgs_rescued count of unevictable pages rescued��
@ mem.vmstat.unevictable_pgs_scanned count of unevictable pages scanned��
@ mem.vmstat.unevictable_pgs_stranded count of unevictable pages stranded��
@ mem.vmstat.zone_reclaim_failed number of zone reclaim failures��
@ mem.vmstat.balloon_inflate count of virt guest balloon page inflations��
@ mem.vmstat.balloon_deflate number of virt guest balloon page deflations��
@ mem.vmstat.balloon_migrate number of virt guest balloon page migrations��
@ mem.vmstat.compact_daemon_wake number of times the memory compaction daemon was woken��
@ mem.vmstat.compact_daemon_free_scanned count of pages scanned for migration by compaction daemon��
@ mem.vmstat.compact_daemon_migrate_scanned count of pages scanned for migration by compaction daemon��
@ mem.vmstat.compact_free_scanned count of pages scanned for freeing��
@ mem.vmstat.compact_isolated count of isolated compaction pages��
@ mem.vmstat.compact_migrate_scanned count of pages scanned for migration��
@ mem.vmstat.drop_pagecache count of calls to drop page cache pages��
@ mem.vmstat.drop_slab count of calls to drop slab cache pages��
@ mem.vmstat.nr_free_cma count of free Contiguous Memory Allocator pages��
@ mem.vmstat.nr_pages_scanned count of pages scanned during page reclaim��
@ mem.vmstat.nr_vmscan_immediate_reclaim prioritise for reclaim when writeback ends��
@ mem.vmstat.numa_pages_migrated count of NUMA page migrations��
@ mem.vmstat.numa_pte_updates count of NUMA page table entry updates��
@ mem.vmstat.numa_huge_pte_updates count of NUMA huge page table entry updates��
@ mem.vmstat.numa_hint_faults count of page migrations from NUMA PTE fault hints��
@ mem.vmstat.numa_hint_faults_local count of NUMA PTE fault hints satisfied locally��
@ mem.vmstat.oom_kill count of out-of-memory kills��
@ mem.vmstat.pglazyfreed count of pages lazily freed��
@ mem.vmstat.pgmigrate_fail count of unsuccessful NUMA page migrations��
@ mem.vmstat.pgmigrate_success count of successful NUMA page migrations��
@ mem.vmstat.thp_deferred_split_page count of huge page enqueues for splitting��
@ mem.vmstat.thp_split_page count of huge page splits into base pages��
@ mem.vmstat.thp_split_page_failed count of failures to split a huge page��
@ mem.vmstat.thp_split_pmd count of times a PMD was split into table of PTEs�This can happen, for instance, when an application calls mprotect() or
munmap() on part of huge page. It doesn't split the huge page, only the
page table entry.�
@ mem.vmstat.thp_split_pud count of times a huge page PUD was split��
@ mem.vmstat.thp_swpout count of transparent huge page swapouts��
@ mem.vmstat.thp_swpout_fallback count of transparent huge page swapout fallbacks��
@ mem.vmstat.workingset_activate count of page activations to form the working set��
@ mem.vmstat.workingset_nodes count of NUMA nodes holding the working set��
@ mem.vmstat.workingset_nodereclaim count of NUMA node working set page reclaims��
@ mem.vmstat.workingset_refault count of refaults of previously evicted pages��
@ mem.vmstat.workingset_restore count of restored of previously evicted pages��
@ mem.vmstat.compact_isolated count of page isolations for memory compaction��
@ mem.vmstat.nr_shmem_hugepages number of huge pages used for shared memory��
@ mem.vmstat.nr_shmem_pmdmapped number of PMD mappings used for shared memory��
@ mem.vmstat.nr_zone_inactive_anon number of inactive anonymous memory pages in zones��
@ mem.vmstat.nr_zone_active_anon number of inactive file memory pages in zones��
@ mem.vmstat.nr_zone_inactive_file number of isolated anonymous memory pages in zones��
@ mem.vmstat.nr_zone_active_file number of isolated file memory pages in zones��
@ mem.vmstat.nr_zone_unevictable number of unevictable memory pages in zones��
@ mem.vmstat.nr_zone_write_pending count of dirty, writeback and unstable pages��
@ mem.vmstat.nr_zspages number of compressed pages��
@ mem.vmstat.thp_file_alloc count of times huge pages were allocated and put in page cache��
@ mem.vmstat.thp_file_mapped count of times huge pages were used for file mappings��
@ mem.zoneinfo.free free space in each zone for each NUMA node��
@ mem.zoneinfo.min min space in each zone for each NUMA node��
@ mem.zoneinfo.low low space in each zone for each NUMA node��
@ mem.zoneinfo.high high space in each zone for each NUMA node��
@ mem.zoneinfo.scanned scanned space in each zone for each NUMA node��
@ mem.zoneinfo.spanned spanned space in each zone for each NUMA node��
@ mem.zoneinfo.present present space in each zone for each NUMA node��
@ mem.zoneinfo.managed managed space in each zone for each NUMA node��
@ mem.zoneinfo.nr_free_pages number of free pages in each zone for each NUMA node.��
@ mem.zoneinfo.nr_alloc_batch number of pages allocated to other zones due to insufficient memory�Number of pages allocated to other zones due to insufficient memory, for
each zone for each NUMA node.�
@ mem.zoneinfo.nr_inactive_anon number of inactive anonymous memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_active_anon number of active anonymous memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_inactive_file number of inactive file memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_active_file number of active file memory memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_unevictable number of unevictable pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_mlock number of pages under mlock in each zone for each NUMA node��
@ mem.zoneinfo.nr_anon_pages number of anonymous mapped pagecache pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_mapped number of mapped pagecache pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_file_pages number of file pagecache pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_dirty number of pages dirty state in each zone for each NUMA node��
@ mem.zoneinfo.nr_writeback number of pages writeback state in each zone for each NUMA node��
@ mem.zoneinfo.nr_slab_reclaimable number of reclaimable slab pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_slab_unreclaimable number of unreclaimable slab pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_page_table_pages number of page table pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_kernel_stack number of pages of kernel stack in each zone for each NUMA node��
@ mem.zoneinfo.nr_unstable number of pages unstable state in each zone for each NUMA node��
@ mem.zoneinfo.nr_bounce number of bounce buffer pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_vmscan_write pages written from the LRU by the VM scanner�Count of pages written from the LRU by the VM scanner in each zone
for each NUMA node.The VM is supposed to minimise the number of
pages which get written from the LRU (for IO scheduling efficiency,
and for high reclaim-success rates).�
@ mem.zoneinfo.nr_vmscan_immediate_reclaim prioritise for reclaim when writeback ends in each zone for each NUMA node��
@ mem.zoneinfo.nr_writeback_temp number of temporary writeback pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_isolated_anon number of isolated anonymous memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_isolated_file number of isolated file memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_shmem number of shared memory pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_dirtied count of pages entering dirty state in each zone for each NUMA node��
@ mem.zoneinfo.nr_written count of pages written out in each zone for each NUMA node��
@ mem.zoneinfo.numa_hit successful allocations from preferred NUMA zone�Count of successful allocations from preferred NUMA zone in each zone
for each NUMA node.�
@ mem.zoneinfo.numa_miss unsuccessful allocations from preferred NUMA zone�Count of unsuccessful allocations from preferred NUMA zone in each zone
for each NUMA node.�
@ mem.zoneinfo.numa_foreign foreign NUMA zone allocations�Count of foreign NUMA zone allocations in each zone for each NUMA node.�
@ mem.zoneinfo.numa_interleave count of interleaved NUMA allocations in each zone for each NUMA node��
@ mem.zoneinfo.numa_local successful allocations from local NUMA zone�Count of successful allocations from local NUMA zone in each zone for
each NUMA node.�
@ mem.zoneinfo.numa_other unsuccessful allocations from local NUMA zone�Count of unsuccessful allocations from local NUMA zone in each zone for
each NUMA node.�
@ mem.zoneinfo.workingset_refault count of refaults of previously evicted pages in each zone for each NUMA node��
@ mem.zoneinfo.workingset_activate count of page activations to form the working set in each zone for each NUMA node��
@ mem.zoneinfo.workingset_nodereclaim count of NUMA node working set page reclaims in each zone for each NUMA node��
@ mem.zoneinfo.nr_anon_transparent_hugepages number of anonymous transparent huge pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_free_cma count of free Contiguous Memory Allocator pages in each zone for each NUMA node��
@ mem.zoneinfo.cma count of Contiguous Memory Allocator pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_swapcached swap cache pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_shmem_hugepages shared memory huge pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_shmem_pmdmapped shared memory PMD mappings in each zone for each NUMA node��
@ mem.zoneinfo.nr_file_hugepages file-backed huge pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_file_pmdmapped file-backed PMD mappings in each zone for each NUMA node��
@ mem.zoneinfo.nr_kernel_misc_reclaimable miscellaneous reclaimable kernel pages in each zone for each NUMA node��
@ mem.zoneinfo.nr_foll_pin_acquired foll pin acquired pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_foll_pin_released count of foll pin released pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_refault_anon anonymous refault working set pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_refault_file file-backed refault working set pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_active_anon active anonymous working set pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_active_file active file-backed working set pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_restore_anon restored anonymous working set pages for each zone for each NUMA node��
@ mem.zoneinfo.workingset_restore_file file-backed restored anonymous working set pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zspages count of zsmalloc memory allocator pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_inactive_file zone inactive file-backed pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_active_file zone active file-backed pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_inactive_anon zone inactive anonymous pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_active_anon zone active anonymous pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_unevictable zone unevictable pages for each zone for each NUMA node��
@ mem.zoneinfo.nr_zone_write_pending zone write-pending pages for each zone for each NUMA node��
@ mem.zoneinfo.protection protection space in each zone for each NUMA node��
@ mem.ksm.full_scans Number of times that KSM has scanned for duplicated content��
@ mem.ksm.merge_across_nodes Kernel allows merging across NUMA nodes��
@ mem.ksm.pages_shared The number of nodes in the stable tree��
@ mem.ksm.pages_sharing The number of virtual pages that are sharing a single page��
@ mem.ksm.pages_to_scan Number of pages to scan at a time��
@ mem.ksm.pages_unshared The number of nodes in the unstable tree��
@ mem.ksm.pages_volatile Number of pages that are candidate to be shared��
@ mem.ksm.run_state Whether the KSM daemon has run and/or is running��
@ mem.ksm.sleep_time Time ksmd should sleep between batches��
@ swap.length total swap available metric from /proc/meminfo��
@ swap.used swap used metric from /proc/meminfo��
@ swap.free swap free metric from /proc/meminfo��
@ kernel.all.load 1, 5 and 15 minute load average��
@ kernel.all.nusers number of user sessions on the system (including root)��
@ kernel.all.nroots number of root user sessions on the system (only root)��
@ kernel.all.nsessions number of utmp sessions (login records)��
@ kernel.all.cpu.user total user CPU time from /proc/stat for all CPUs, including guest CPU time��
@ kernel.all.cpu.vuser total user CPU time from /proc/stat for all CPUs, excluding guest CPU time��
@ kernel.all.cpu.intr total interrupt CPU time from /proc/stat for all CPUs�Total time spent processing interrupts on all CPUs.
This value includes both soft and hard interrupt processing time.�
@ kernel.all.cpu.wait.total total wait CPU time from /proc/stat for all CPUs��
@ kernel.all.cpu.nice total nice user CPU time from /proc/stat for all CPUs, including guest time��
@ kernel.all.cpu.vnice total nice user CPU time from /proc/stat for all CPUs, excluding guest time��
@ kernel.all.cpu.sys total sys CPU time from /proc/stat for all CPUs��
@ kernel.all.cpu.idle total idle CPU time from /proc/stat for all CPUs��
@ kernel.all.cpu.irq.soft soft interrupt CPU time from /proc/stat for all CPUs�Total soft interrupt CPU time (deferred interrupt handling code,
not run in the initial interrupt handler).�
@ kernel.all.cpu.irq.hard hard interrupt CPU time from /proc/stat for all CPUs�Total hard interrupt CPU time ("hard" interrupt handling code
is the code run directly on receipt of the initial hardware
interrupt, and does not include "soft" interrupt handling code
which is deferred until later).�
@ kernel.all.cpu.steal total virtualisation CPU steal time for all CPUs�Total CPU time when a CPU had a runnable process, but the hypervisor
(virtualisation layer) chose to run something else instead.�
@ kernel.all.cpu.guest total virtual guest CPU time for all CPUs�Total CPU time spent running virtual guest operating systems.�
@ kernel.all.cpu.guest_nice total virtual guest CPU nice time for all CPUs�Total CPU nice time spent running virtual guest operating systems.�
@ kernel.pernode.cpu.user total user CPU time from /proc/stat for each node, including guest CPU time��
@ kernel.pernode.cpu.vuser total user CPU time from /proc/stat for each node, excluding guest CPU time��
@ kernel.pernode.cpu.intr total interrupt CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.wait.total total wait CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.nice total nice user CPU time from /proc/stat for each node, including guest time��
@ kernel.pernode.cpu.vnice total nice user CPU time from /proc/stat for each node, excluding guest time��
@ kernel.pernode.cpu.sys total sys CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.idle total idle CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.irq.soft soft interrupt CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.irq.hard hard interrupt CPU time from /proc/stat for each node��
@ kernel.pernode.cpu.steal total virtualisation CPU steal time for each node��
@ kernel.pernode.cpu.guest total virtual guest CPU time for each node��
@ kernel.pernode.cpu.guest_nice total virtual nice guest CPU time for each node��
@ hinv.ninterface number of active (up) network interfaces��
@ network.interface.in.bytes network recv read bytes from /proc/net/dev per network interface�bytes column on the "Receive" side of /proc/net/dev (stats->rx_bytes counter in
rtnl_link_stats64)�
@ network.interface.in.packets network recv read packets from /proc/net/dev per network interface�packets column on the "Receive" side of /proc/net/dev
(stats->rx_packets counter in rtnl_link_stats64)�
@ network.interface.in.errors network recv read errors from /proc/net/dev per network interface�errors column on the "Receive" side of /proc/net/dev
(stats->rx_errors counter in rtnl_link_stats64)�
@ network.interface.in.drops network recv read drops from /proc/net/dev per network interface�drop column on the "Receive" side of /proc/net/dev
(stats->{rx_dropped + rx_missed_errors} counters in rtnl_link_stats64)
rx_dropped are the dropped packets due to no space in linux buffers and rx_missed
are due to the receiver NIC missing a packet.
Not all NICS use the rx_missed_errors counter.�
@ network.interface.in.mcasts network recv multicast packets from /proc/net/dev per network interface�multicast column on the "Receive" side of /proc/net/dev
(stats->multicast counter in rtnl_link_stats64)�
@ network.interface.in.fifo network recv fifo overrun errors from /proc/net/dev per network interface�fifo column on the "Receive" side of /proc/net/dev
(stats->rx_fifo_errors counter in rtnl_link_stats64)�
@ network.interface.in.frame network recv frames errors from /proc/net/dev per network interface�frame column on the "Receive" side of /proc/net/dev
(stats->{rx_length_errors + rx_over_errors + rx_crc_errors + rx_frame_errors}
counters in rtnl_link_stats64)�
@ network.interface.in.compressed network recv compressed from /proc/net/dev per network interface�compressed column on the "Receive" side of /proc/net/dev
(stats->rx_compressed counter in rtnl_link_stats64).
Almost exclusively used for CSLIP or HDLC devices�
@ network.interface.out.bytes network send bytes from /proc/net/dev per network interface�bytes column on the "Transmit" side of /proc/net/dev
(stats->tx_bytes counter in rtnl_link_stats64)�
@ network.interface.out.packets network send packets from /proc/net/dev per network interface�packets column on the "Transmit" side of /proc/net/dev
(stats->tx_packets counter in rtnl_link_stats64)�
@ network.interface.out.errors network send errors from /proc/net/dev per network interface�errors column on the "Transmit" side of /proc/net/dev
(stats->tx_errors counter in rtnl_link_stats64)�
@ network.interface.out.drops network send drops from /proc/net/dev per network interface�drop column on the "Transmit" side of /proc/net/dev
(stats->tx_dropped counter in rtnl_link_stats64)�
@ network.interface.out.fifo network send fifos from /proc/net/dev per network interface�fifo column on the "Transmit" side of /proc/net/dev
(stats->tx_fifo_errors counter in rtnl_link_stats64)�
@ network.interface.collisions network send collisions from /proc/net/dev per network interface�colls column on the "Transmit" side of /proc/net/dev
(stats->collisions counter in rtnl_link_stats64).
Counter only valid for outgoing packets.�
@ network.interface.out.carrier network send carrier from /proc/net/dev per network interface�carrier column on the "Transmit" side of /proc/net/dev
(stats->{tx_carrier_errors + tx_aborted_errors + tx_window_errors +
tx_heartbeat_errors} counters in rtnl_link_stats64).�
@ network.interface.out.compressed network send compressed from /proc/net/dev per network interface�compressed column on the "Transmit" side of /proc/net/dev
(stats->tx_compressed counter in rtnl_link_stats64).
Almost exclusively used for CSLIP or HDLC devices�
@ network.interface.total.bytes network total (in+out) bytes from /proc/net/dev per network interface��
@ network.interface.total.packets network total (in+out) packets from /proc/net/dev per network interface��
@ network.interface.total.errors network total (in+out) errors from /proc/net/dev per network interface��
@ network.interface.total.drops network total (in+out) drops from /proc/net/dev per network interface��
@ network.interface.total.mcasts network total (in) mcasts from /proc/net/dev per network interface�Linux does not account for outgoing mcast packets per device, so this counter
is identical to the incoming mcast metric.�
@ network.interface.mtu maximum transmission unit on network interface��
@ network.interface.speed interface speed in megabytes per second�The linespeed on the network interface, as reported by the kernel,
scaled from Megabits/second to Megabytes/second.
See also network.interface.baudrate for the bytes/second value.�
@ network.interface.baudrate interface speed in bytes per second�The linespeed on the network interface, as reported by the kernel,
scaled up from Megabits/second to bits/second and divided by 8 to convert
to bytes/second.
See also network.interface.speed for the Megabytes/second value.�
@ network.interface.duplex value one for half or two for full duplex interface��
@ network.interface.up boolean for whether interface is currently up or down��
@ network.interface.running boolean for whether interface has resources allocated��
@ network.interface.wireless boolean for whether interface is wireless��
@ network.interface.type sysfs interface name assignment type value��
@ network.interface.inet_addr string INET interface address (ifconfig style)��
@ network.interface.ipv6_addr string IPv6 interface address (ifconfig style)��
@ network.interface.ipv6_scope string IPv6 interface scope (ifconfig style)��
@ network.interface.hw_addr hardware address (from sysfs)��
@ network.sockstat.total total number of sockets used by the system.��
@ network.sockstat.tcp.inuse instantaneous number of tcp sockets currently in use��
@ network.sockstat.tcp.orphan instantaneous number of orphan sockets��
@ network.sockstat.tcp.tw instantaneous number of sockets waiting close��
@ network.sockstat.tcp.alloc instantaneous number of allocated sockets��
@ network.sockstat.tcp.mem instantaneous number of used memory for tcp��
@ network.sockstat.udp.inuse instantaneous number of udp sockets currently in use��
@ network.sockstat.udp.mem instantaneous number of used memory for udp��
@ network.sockstat.udplite.inuse instantaneous number of udplite sockets currently in use��
@ network.sockstat.raw.inuse instantaneous number of raw sockets currently in use��
@ network.sockstat.frag.inuse instantaneous number of frag sockets currently in use��
@ network.sockstat.frag.memory nstantaneous number of used memory for frag��
@ network.sockstat.tcp6.inuse instantaneous number of tcp6 sockets currently in use��
@ network.sockstat.udp6.inuse instantaneous number of udp6 sockets currently in use��
@ network.sockstat.udplite6.inuse instantaneous number of udplite6 sockets currently in use��
@ network.sockstat.raw6.inuse instantaneous number of raw6 sockets currently in use��
@ network.sockstat.frag6.inuse instantaneous number of frag6 sockets currently in use��
@ network.sockstat.frag6.memory instantaneous number of used memory for frag6��
@ network.softnet.processed number of packets (not including netpoll) received by the interrupt handler��
@ network.softnet.dropped number of packets that were dropped because netdev_max_backlog was exceeded��
@ network.softnet.time_squeeze number of times ksoftirq ran out of netdev_budget or time slice with work remaining��
@ network.softnet.cpu_collision number of times that two cpus collided trying to get the device queue lock��
@ network.softnet.received_rps number of times rps_trigger_softirq has been called��
@ network.softnet.flow_limit_count softnet_data flow limit counter��
@ network.softnet.percpu.processed number of packets (not including netpoll) received by the interrupt handler��
@ network.softnet.percpu.dropped number of packets that were dropped because netdev_max_backlog was exceeded��
@ network.softnet.percpu.time_squeeze number of times ksoftirq ran out of netdev_budget or time slice with work remaining��
@ network.softnet.percpu.cpu_collision number of times that two cpus collided trying to get the device queue lock��
@ network.softnet.percpu.received_rps number of times rps_trigger_softirq has been called��
@ network.softnet.percpu.flow_limit_count softnet_data flow limit counter�The network stack has to drop packets when a receive processing a CPUs
backlog reaches netdev_max_backlog.  The flow_limit_count counter is
the number of times very active flows have dropped their traffic earlier
to maintain capacity for other less active flows.�
@ network.ip6.inreceives count of ip6 inreceives��
@ network.ip6.inhdrerrors count of ip6 inhdrerrors��
@ network.ip6.intoobigerrors count of ip6 intoobigerrors��
@ network.ip6.innoroutes count of ip6 innoroutes��
@ network.ip6.inaddrerrors count of ip6 inaddrerrors��
@ network.ip6.inunknownprotos count of ip6 inunknownprotos��
@ network.ip6.intruncatedpkts count of ip6 intruncatedpkts��
@ network.ip6.indiscards count of ip6 indiscards��
@ network.ip6.indelivers count of ip6 indelivers��
@ network.ip6.outforwdatagrams count of ip6 outforwdatagrams��
@ network.ip6.outrequests count of ip6 outrequests��
@ network.ip6.outdiscards count of ip6 outdiscards��
@ network.ip6.outnoroutes count of ip6 outnoroutes��
@ network.ip6.reasmtimeout count of ip6 reasmtimeout��
@ network.ip6.reasmreqds count of ip6 reassembly requireds��
@ network.ip6.reasmoks count of ip6 reassembly oks��
@ network.ip6.reasmfails count of ip6 reassembly failures�The number of failures detected by the IPv6 re-assembly algorithm (for
whatever reason: timed out, errors, etc).  Note that this is not necessarily
a count of discarded IPv6 fragments since some algorithms can lose track
of the number of fragments by combining them as they are received.�
@ network.ip6.fragoks count of ip6 fragmentation oks��
@ network.ip6.fragfails count of ip6 fragmentation failures��
@ network.ip6.fragcreates count of ip6 fragmentation creations��
@ network.ip6.inmcastpkts count of ip6 multicast packets in��
@ network.ip6.outmcastpkts count of ip6 multicast packets out��
@ network.ip6.inoctets count of ip6 octets in��
@ network.ip6.outoctets count of ip6 octets out��
@ network.ip6.inmcastoctets count of ip6 multicast octets in��
@ network.ip6.outmcastoctets count of ip6 multicast octets out��
@ network.ip6.inbcastoctets count of ip6 broadcast octets in��
@ network.ip6.outbcastoctets count of ip6 broadcast octets uot��
@ network.ip6.innoectpkts count of ip6 packets received with NOECT��
@ network.ip6.inect1pkts count of ip6 packets received with ECT(1)��
@ network.ip6.inect0pkts count of ip6 packets received with ECT(0)��
@ network.ip6.incepkts count of ip6 Congestion Experimented packets in��
@ network.icmp6.inmsgs count of icmp6 inmsgs��
@ network.icmp6.inerrors count of icmp6 inerrors��
@ network.icmp6.outmsgs count of icmp6 outmsgs��
@ network.icmp6.outerrors count of icmp6 outerrors��
@ network.icmp6.incsumerrors count of icmp6 incsumerrors��
@ network.icmp6.indestunreachs count of icmp6 indestunreachs��
@ network.icmp6.inpkttoobigs count of icmp6 inpkttoobigs��
@ network.icmp6.intimeexcds count of icmp6 intimeexcds��
@ network.icmp6.inparmproblems count of icmp6 inparmprobs��
@ network.icmp6.inechos count of icmp6 inechos��
@ network.icmp6.inechoreplies count of icmp6 inechoreplies��
@ network.icmp6.ingroupmembqueries count of icmp6 ingroupmembqueries��
@ network.icmp6.ingroupmembresponses count of icmp6 ingroupmembresponses��
@ network.icmp6.ingroupmembreductions count of icmp6 ingroupmembreductions��
@ network.icmp6.inroutersolicits count of icmp6 inroutersolicits��
@ network.icmp6.inrouteradvertisements count of icmp6 inrouteradvertisements��
@ network.icmp6.inneighborsolicits count of icmp6 inneighborsolicits��
@ network.icmp6.inneighboradvertisements count of icmp6 inneighboradvertisements��
@ network.icmp6.inredirects count of icmp6 inredirects��
@ network.icmp6.inmldv2reports count of icmp6 inmldv2reports��
@ network.icmp6.outdestunreachs count of icmp6 outdestunreachs��
@ network.icmp6.outpkttoobigs count of icmp6 outpkttoobigs��
@ network.icmp6.outtimeexcds count of icmp6 outtimeexcds��
@ network.icmp6.outparmproblems count of icmp6 outparmproblems��
@ network.icmp6.outechos count of icmp6 outechos��
@ network.icmp6.outechoreplies count of icmp6 outechoreplies��
@ network.icmp6.outgroupmembqueries count of icmp6 outgroupmembqueries��
@ network.icmp6.outgroupmembresponses count of icmp6 outgroupmembresponses��
@ network.icmp6.outgroupmembreductions count of icmp6 outgroupmembreductions��
@ network.icmp6.outroutersolicits count of icmp6 outroutersolicits��
@ network.icmp6.outrouteradvertisements count of icmp6 outrouteradvertisements��
@ network.icmp6.outneighborsolicits count of icmp6 outneighborsolicits��
@ network.icmp6.outneighboradvertisements count of icmp6 outneighboradvertisements��
@ network.icmp6.outredirects count of icmp6 outredirects��
@ network.icmp6.outmldv2reports count of icmp6 outmldv2reports��
@ network.udp6.indatagrams count of udp6 indatagrams��
@ network.udp6.noports count of udp6 noports��
@ network.udp6.inerrors count of udp6 inerrors��
@ network.udp6.outdatagrams count of udp6 outdatagrams��
@ network.udp6.rcvbuferrors count of udp6 rcvbuferrors��
@ network.udp6.sndbuferrors count of udp6 sndbuferrors��
@ network.udp6.incsumerrors count of udp6 incsumerrors��
@ network.udp6.ignoredmulti count of udp6 ignoredmulti��
@ network.udplite6.indatagrams count of udplite6 indatagrams��
@ network.udplite6.noports count of udplite6 noports��
@ network.udplite6.inerrors count of udplite6 inerrors��
@ network.udplite6.outdatagrams count of udplite6 outdatagrams��
@ network.udplite6.rcvbuferrors count of udplite6 receive buffer errors��
@ network.udplite6.sndbuferrors count of udplite6 send buffer errors��
@ network.udplite6.incsumerrors count of udplite6 in checksum errors��
@ network.tcpconn6.established Number of established tcp6 connections��
@ network.tcpconn6.syn_sent Number of SYN_SENT tcp6 connections��
@ network.tcpconn6.syn_recv Number of SYN_RECV tcp6 connections��
@ network.tcpconn6.fin_wait1 Number of FIN_WAIT1 tcp6 connections��
@ network.tcpconn6.fin_wait2 Number of FIN_WAIT2 tcp6 connections��
@ network.tcpconn6.time_wait Number of TIME_WAIT tcp6 connections��
@ network.tcpconn6.close Number of CLOSE tcp6 connections��
@ network.tcpconn6.close_wait Number of CLOSE_WAIT tcp6 connections��
@ network.tcpconn6.last_ack Number of LAST_ACK tcp6 connections��
@ network.tcpconn6.listen Number of LISTEN tcp6 connections��
@ network.tcpconn6.closing Number of CLOSING tcp6 connections��
@ network.rawconn.count Number of raw socket connections��
@ network.rawconn6.count Number of raw6 socket connections��
@ network.udpconn.established Number of established udp connections��
@ network.udpconn.listen Number of udp connections in listen state��
@ network.udpconn6.established Number of established udp6 connections��
@ network.udpconn6.listen Number of udp6 connections in listen state��
@ network.unix.datagram.count Number of datagram unix domain sockets��
@ network.unix.stream.established Number of established unix domain socket streams��
@ network.unix.stream.listen Number of unix domain socket streams in listen state��
@ network.unix.stream.count Number of unix domain socket streams��
@ hinv.physmem total system memory metric from /proc/meminfo��
@ hinv.pagesize Memory page size�The memory page size of the running kernel in bytes.�
@ hinv.hugepagesize Huge page size from /proc/meminfo�The memory huge page size of the running kernel in bytes.�
@ hinv.ncpu number of CPUs in the system��
@ hinv.ndisk number of disks in the system��
@ hinv.nfilesys number of (local) file systems currently mounted��
@ hinv.nnode number of NUMA nodes in the system��
@ hinv.map.scsi list of active SCSI devices�There is one string value for each SCSI device active in the system,
as extracted from /proc/scsi/scsi. The external instance name
for each device is in the format scsiD:C:I:L where
D is controller number, C is channel number, I is device ID
and L is the SCSI LUN number for the device. The values for this
metric are the actual device names (sd[a-z] are SCSI disks, st[0-9]
are SCSI tapes and scd[0-9] are SCSI CD-ROMS.�
@ filesys.capacity Total capacity of mounted filesystem (Kbytes)��
@ filesys.used Total space used on mounted filesystem (Kbytes)��
@ filesys.free Total space free on mounted filesystem (Kbytes)��
@ filesys.maxfiles Inodes capacity of mounted filesystem��
@ filesys.usedfiles Number of inodes allocated on mounted filesystem��
@ filesys.freefiles Number of unallocated inodes on mounted filesystem��
@ filesys.mountdir File system mount point��
@ filesys.full Percentage of filesystem in use��
@ filesys.blocksize Size of each block on mounted filesystem (Bytes)��
@ filesys.avail Total space free to non-superusers on mounted filesystem (Kbytes)��
@ filesys.readonly Indicates whether a filesystem is mounted readonly��
@ tmpfs.capacity Total capacity of mounted tmpfs filesystem (Kbytes)��
@ tmpfs.used Total space used on mounted tmpfs filesystem (Kbytes)��
@ tmpfs.free Total space free on mounted tmpfs filesystem (Kbytes)��
@ tmpfs.maxfiles Inodes capacity of mounted tmpfs filesystem��
@ tmpfs.usedfiles Number of inodes allocated on mounted tmpfs filesystem��
@ tmpfs.freefiles Number of unallocated inodes on mounted tmpfs filesystem��
@ tmpfs.full Percentage of tmpfs filesystem in use��
@ swapdev.free physical swap free space��
@ swapdev.length physical swap size��
@ swapdev.maxswap maximum swap length (same as swapdev.length on Linux)��
@ swapdev.vlength virtual swap size (always zero on Linux)�Virtual swap size (always zero on Linux since Linux does not support
virtual swap).

This metric is retained on Linux for interoperability with PCP monitor
tools running on IRIX.�
@ swapdev.priority swap resource priority��
@ nfs.client.calls cumulative total of client NFSv2 requests��
@ nfs.client.reqs cumulative total of client NFSv2 requests by request type��
@ nfs.server.calls cumulative total of server NFSv2 requests��
@ nfs.server.reqs cumulative total of client NFSv2 requests by request type��
@ nfs.server.threads.total number of nfsd threads running��
@ nfs.server.threads.pools number of thread pools��
@ nfs.server.threads.requests cumulative total of requests received��
@ nfs.server.threads.enqueued cumulative total of requests that had to wait to be processed��
@ nfs.server.threads.processed cumulative total of requests processed immediately��
@ nfs.server.threads.timedout cumulative total of threads that timedout from inactivity��
@ nfs3.client.calls cumulative total of client NFSv3 requests��
@ nfs3.client.reqs cumulative total of client NFSv3 requests by request type��
@ nfs3.server.calls cumulative total of server NFSv3 requests��
@ nfs3.server.reqs cumulative total of client NFSv3 requests by request type��
@ nfs4.client.calls cumulative total of client NFSv4 requests��
@ nfs4.client.reqs cumulative total for each client NFSv4 request type��
@ nfs4.server.calls cumulative total of server NFSv4 operations, plus NULL requests��
@ nfs4.server.reqs cumulative total for each server NFSv4 operation, and for NULL requests��
@ rpc.client.rpccnt cumulative total of client RPC requests��
@ rpc.client.rpcretrans cumulative total of client RPC retransmissions��
@ rpc.client.rpcauthrefresh cumulative total of client RPC auth refreshes��
@ rpc.client.netcnt cumulative total of client RPC network layer requests��
@ rpc.client.netudpcnt cumulative total of client RPC UDP network layer requests��
@ rpc.client.nettcpcnt cumulative total of client RPC TCP network layer requests��
@ rpc.client.nettcpconn cumulative total of client RPC TCP network layer connection requests��
@ rpc.server.rpccnt cumulative total of server RPC requests��
@ rpc.server.rpcerr cumulative total of server RPC errors��
@ rpc.server.rpcbadfmt cumulative total of server RPC bad format errors��
@ rpc.server.rpcbadauth cumulative total of server RPC bad auth errors��
@ rpc.server.rpcbadclnt cumulative total of server RPC bad client errors��
@ rpc.server.rchits cumulative total of request-reply-cache hits��
@ rpc.server.rcmisses cumulative total of request-reply-cache misses��
@ rpc.server.rcnocache cumulative total of uncached request-reply-cache requests��
@ rpc.server.fh_cached cumulative total of file handle cache requests��
@ rpc.server.fh_valid cumulative total of file handle cache validations��
@ rpc.server.fh_fixup cumulative total of file handle cache fixup validations��
@ rpc.server.fh_lookup cumulative total of file handle cache new lookups��
@ rpc.server.fh_stale cumulative total of stale file handle cache errors��
@ rpc.server.fh_concurrent cumulative total of concurrent file handle cache requests��
@ rpc.server.netcnt cumulative total of server RPC network layer requests��
@ rpc.server.netudpcnt cumulative total of server RPC UDP network layer requests��
@ rpc.server.nettcpcnt cumulative total of server RPC TCP network layer requests��
@ rpc.server.nettcpconn cumulative total of server RPC TCP network layer connection requests��
@ rpc.server.fh_anon cumulative total anonymous file dentries returned��
@ rpc.server.fh_nocache_dir count of directory file handles not found cached��
@ rpc.server.fh_nocache_nondir count of non-directory file handles not found cached��
@ rpc.server.io_read cumulative count of bytes returned from read requests��
@ rpc.server.io_write cumulative count of bytes passed into write requests��
@ rpc.server.th_cnt available nfsd threads��
@ rpc.server.th_fullcnt number of times the last free nfsd thread was used��
@ rpc.server.ra_size size of read-ahead params cache��
@ rpc.server.ra_hits count of read-ahead params cache hits��
@ rpc.server.ra_misses count of read-ahead params cache misses��
@ network.ip.forwarding count of ip forwarding��
@ network.ip.defaultttl count of ip defaultttl��
@ network.ip.inreceives count of ip inreceives��
@ network.ip.inhdrerrors count of ip inhdrerrors��
@ network.ip.inaddrerrors count of ip inaddrerrors��
@ network.ip.forwdatagrams count of ip forwdatagrams��
@ network.ip.inunknownprotos count of ip inunknownprotos��
@ network.ip.indiscards count of ip indiscards��
@ network.ip.indelivers count of ip indelivers��
@ network.ip.outrequests count of ip outrequests��
@ network.ip.outdiscards count of ip outdiscards��
@ network.ip.outnoroutes count of ip outnoroutes��
@ network.ip.reasmtimeout count of ip reasmtimeout��
@ network.ip.reasmreqds count of ip reasmreqds��
@ network.ip.reasmoks count of ip reasmoks��
@ network.ip.reasmfails count of ip reasmfails�The number of failures detected by the IP re-assembly algorithm (for whatever
reason: timed out, errors, etc).  Note that this is not necessarily a
count of discarded IP fragments since some algorithms can lose track of
the number of fragments by combining them as they are received.�
@ network.ip.fragoks count of ip fragoks��
@ network.ip.fragfails count of ip fragfails��
@ network.ip.fragcreates count of ip fragcreates��
@ network.icmp.inmsgs count of icmp inmsgs��
@ network.icmp.inerrors count of icmp inerrors��
@ network.icmp.indestunreachs count of icmp indestunreachs��
@ network.icmp.intimeexcds count of icmp intimeexcds��
@ network.icmp.inparmprobs count of icmp inparmprobs��
@ network.icmp.insrcquenchs count of icmp insrcquenchs��
@ network.icmp.inredirects count of icmp inredirects��
@ network.icmp.inechos count of icmp inechos��
@ network.icmp.inechoreps count of icmp inechoreps��
@ network.icmp.intimestamps count of icmp intimestamps��
@ network.icmp.intimestampreps count of icmp intimestampreps��
@ network.icmp.inaddrmasks count of icmp inaddrmasks��
@ network.icmp.inaddrmaskreps count of icmp inaddrmaskreps��
@ network.icmp.outmsgs count of icmp outmsgs��
@ network.icmp.outerrors count of icmp outerrors��
@ network.icmp.outdestunreachs count of icmp outdestunreachs��
@ network.icmp.outtimeexcds count of icmp outtimeexcds��
@ network.icmp.outparmprobs count of icmp outparmprobs��
@ network.icmp.outsrcquenchs count of icmp outsrcquenchs��
@ network.icmp.outredirects count of icmp outredirects��
@ network.icmp.outechos count of icmp outechos��
@ network.icmp.outechoreps count of icmp outechoreps��
@ network.icmp.outtimestamps count of icmp outtimestamps��
@ network.icmp.outtimestampreps count of icmp outtimestampreps��
@ network.icmp.outaddrmasks count of icmp outaddrmasks��
@ network.icmp.outaddrmaskreps count of icmp outaddrmaskreps��
@ network.icmp.incsumerrors count of icmp in checksum errors��
@ network.icmpmsg.intype count of icmp message types recvd��
@ network.icmpmsg.outtype count of icmp message types sent��
@ network.tcp.rtoalgorithm the retransmission timeout algorithm in use��
@ network.tcp.rtomin minimum retransmission timeout��
@ network.tcp.rtomax maximum retransmission timeout��
@ network.tcp.maxconn limit on tcp connections��
@ network.tcp.activeopens count of tcp activeopens��
@ network.tcp.passiveopens count of tcp passiveopens��
@ network.tcp.attemptfails count of tcp attemptfails��
@ network.tcp.estabresets count of tcp estabresets��
@ network.tcp.currestab current established tcp connections��
@ network.tcp.insegs count of tcp segments received��
@ network.tcp.outsegs count of tcp segments sent��
@ network.tcp.retranssegs count of tcp segments retransmitted��
@ network.tcp.inerrs count of tcp segments received in error��
@ network.tcp.outrsts count of tcp segments sent with RST flag��
@ network.tcp.incsumerrors count of tcp segments received with checksum errors��
@ network.tcpconn.established Number of established connections��
@ network.tcpconn.syn_sent Number of SYN_SENT connections��
@ network.tcpconn.syn_recv Number of SYN_RECV connections��
@ network.tcpconn.fin_wait1 Number of FIN_WAIT1 connections��
@ network.tcpconn.fin_wait2 Number of FIN_WAIT2 connections��
@ network.tcpconn.time_wait Number of TIME_WAIT connections��
@ network.tcpconn.close Number of CLOSE connections��
@ network.tcpconn.close_wait Number of CLOSE_WAIT connections��
@ network.tcpconn.last_ack Number of LAST_ACK connections��
@ network.tcpconn.listen Number of LISTEN connections��
@ network.tcpconn.closing Number of CLOSING connections��
@ network.udp.indatagrams count of udp indatagrams��
@ network.udp.noports count of udp noports��
@ network.udp.inerrors count of udp inerrors��
@ network.udp.outdatagrams count of udp outdatagrams��
@ network.udp.recvbuferrors count of udp receive buffer errors��
@ network.udp.sndbuferrors count of udp send buffer errors��
@ network.udp.incsumerrors count of udp in checksum errors��
@ network.udplite.indatagrams count of udplite indatagrams��
@ network.udplite.noports count of udplite noports��
@ network.udplite.inerrors count of udplite inerrors��
@ network.udplite.outdatagrams count of udplite outdatagrams��
@ network.udplite.recvbuferrors count of udplite receive buffer errors��
@ network.udplite.sndbuferrors count of udplite send buffer errors��
@ network.udplite.incsumerrors count of udplite in checksum errors��
@ network.ip.innoroutes Number of IP datagrams discarded due to no routes in forwarding path��
@ network.ip.intruncatedpkts Number of IP datagrams discarded due to frame not carrying enough data��
@ network.ip.inmcastpkts Number of received IP multicast datagrams��
@ network.ip.outmcastpkts Number of sent IP multicast datagrams��
@ network.ip.inbcastpkts Number of received IP broadcast datagrams��
@ network.ip.outbcastpkts Number of sent IP bradcast datagrams��
@ network.ip.inoctets Number of received octets��
@ network.ip.outoctets Number of sent octets��
@ network.ip.inmcastoctets Number of received IP multicast octets��
@ network.ip.outmcastoctets Number of sent IP multicast octets��
@ network.ip.inbcastoctets Number of received IP broadcast octets��
@ network.ip.outbcastoctets Number of sent IP broadcast octets��
@ network.ip.csumerrors Number of IP datagrams with checksum errors��
@ network.ip.noectpkts Number of packets received with NOECT��
@ network.ip.ect1pkts Number of packets received with ECT(1)��
@ network.ip.ect0pkts Number of packets received with ECT(0)��
@ network.ip.cepkts Number of packets received with Congestion Experimented��
@ network.tcp.syncookiessent Number of sent SYN cookies��
@ network.tcp.syncookiesrecv Number of received SYN cookies��
@ network.tcp.syncookiesfailed Number of failed SYN cookies��
@ network.tcp.embryonicrsts Number of resets received for embryonic SYN_RECV sockets��
@ network.tcp.prunecalled Number of packets pruned from receive queue because of socket buffer overrun��
@ network.tcp.rcvpruned Number of packets pruned from receive queue��
@ network.tcp.ofopruned Number of packets dropped from out-of-order queue because of socket buffer overrun��
@ network.tcp.outofwindowicmps Number of dropped out of window ICMPs��
@ network.tcp.lockdroppedicmps Number of dropped ICMP because socket was locked��
@ network.tcp.arpfilter Number of arp packets filtered��
@ network.tcp.timewaited Number of TCP sockets finished time wait in fast timer��
@ network.tcp.timewaitrecycled Number of time wait sockets recycled by time stamp��
@ network.tcp.timewaitkilled Number of TCP sockets finished time wait in slow timer��
@ network.tcp.pawspassiverejected Number of passive connections rejected because of timestamp��
@ network.tcp.pawsactiverejected Number of active connections rejected because of timestamp��
@ network.tcp.pawsestabrejected Number of packets rejects in established connections because of timestamp��
@ network.tcp.delayedacks Number of delayed acks sent��
@ network.tcp.delayedacklocked Number of delayed acks further delayed because of locked socket��
@ network.tcp.delayedacklost Number of times quick ack mode was activated times��
@ network.tcp.listenoverflows Number of times the listen queue of a socket overflowed��
@ network.tcp.listendrops Number of SYNs to LISTEN sockets dropped��
@ network.tcp.prequeued Number of packets directly queued to recvmsg prequeue��
@ network.tcp.directcopyfrombacklog Number of bytes directly in process context from backlog��
@ network.tcp.directcopyfromprequeue Number of bytes directly received in process context from prequeue��
@ network.tcp.prequeueddropped Number of packets dropped from prequeue��
@ network.tcp.hphits Number of packet headers predicted��
@ network.tcp.hphitstouser Number of packets header predicted and directly queued to user��
@ network.tcp.pureacks Number of acknowledgments not containing data payload received��
@ network.tcp.hpacks Number of predicted acknowledgments��
@ network.tcp.renorecovery Number of times recovered from packet loss due to fast retransmit��
@ network.tcp.sackrecovery Number of times recovered from packet loss by selective acknowledgements��
@ network.tcp.sackreneging Number of bad SACK blocks received��
@ network.tcp.fackreorder Number of times detected reordering using FACK��
@ network.tcp.sackreorder Number of times detected reordering using SACK��
@ network.tcp.renoreorder Number of times detected reordering using reno fast retransmit��
@ network.tcp.tsreorder Number of times detected reordering times using time stamp��
@ network.tcp.fullundo Number of congestion windows fully recovered without slow start��
@ network.tcp.partialundo Number of congestion windows partially recovered using Hoe heuristic��
@ network.tcp.dsackundo Number of congestion windows recovered without slow start using DSACK��
@ network.tcp.lossundo Number of congestion windows recovered without slow start after partial ack��
@ network.tcp.lostretransmit Number of retransmits lost��
@ network.tcp.renofailures Number of timeouts after reno fast retransmit��
@ network.tcp.sackfailures Number of timeouts after SACK recovery��
@ network.tcp.lossfailures Number of timeouts in loss state��
@ network.tcp.fastretrans Number of fast retransmits��
@ network.tcp.forwardretrans Number of forward retransmits��
@ network.tcp.slowstartretrans Number of retransmits in slow start��
@ network.tcp.timeouts Number of other TCP timeouts��
@ network.tcp.lossprobes Number of sent TCP loss probes��
@ network.tcp.lossproberecovery Number of TCP loss probe recoveries��
@ network.tcp.renorecoveryfail Number of reno fast retransmits failed��
@ network.tcp.sackrecoveryfail Number of SACK retransmits failed��
@ network.tcp.schedulerfail Number of times receiver scheduled too late for direct processing��
@ network.tcp.rcvcollapsed Number of packets collapsed in receive queue due to low socket buffer��
@ network.tcp.dsackoldsent Number of DSACKs sent for old packets��
@ network.tcp.dsackofosent Number of DSACKs sent for out of order packets��
@ network.tcp.dsackrecv Number of DSACKs received��
@ network.tcp.dsackoforecv Number of DSACKs for out of order packets received��
@ network.tcp.abortondata Number of connections reset due to unexpected data��
@ network.tcp.abortonclose Number of connections reset due to early user close��
@ network.tcp.abortonmemory Number of connections aborted due to memory pressure��
@ network.tcp.abortontimeout Number of connections aborted due to timeout��
@ network.tcp.abortonlinger Number of connections aborted after user close in linger timeout��
@ network.tcp.abortfailed Number of times unable to send RST due to no memory��
@ network.tcp.memorypressures Numer of times TCP ran low on memory��
@ network.tcp.sackdiscard Number of SACKs discarded��
@ network.tcp.dsackignoredold Number of ignored old duplicate SACKs��
@ network.tcp.dsackignorednoundo Number of ignored duplicate SACKs with undo_marker not set��
@ network.tcp.spuriousrtos Number of FRTO's successfully detected spurious RTOs��
@ network.tcp.md5notfound Number of times MD5 hash expected but not found��
@ network.tcp.md5unexpected Number of times MD5 hash unexpected but found��
@ network.tcp.sackshifted Number of SACKs shifted��
@ network.tcp.sackmerged Number of SACKs merged��
@ network.tcp.sackshiftfallback Number of SACKs fallbacks��
@ network.tcp.backlogdrop Number of frames dropped because of full backlog queue��
@ network.tcp.minttldrop Number of frames dropped when TTL is under the minimum��
@ network.tcp.deferacceptdrop Number of dropped ACK frames when socket is in SYN-RECV state�Due to SYNACK retrans count lower than defer_accept value�
@ network.tcp.iprpfilter Number of packets dropped in input path because of rp_filter settings��
@ network.tcp.timewaitoverflow Number of occurrences of time wait bucket overflow��
@ network.tcp.reqqfulldocookies Number of times a SYNCOOKIE was replied to client��
@ network.tcp.reqqfulldrop Number of times a SYN request was dropped due to disabled syncookies��
@ network.tcp.retransfail Number of failed tcp_retransmit_skb() calls��
@ network.tcp.rcvcoalesce Number of times tried to coalesce the receive queue��
@ network.tcp.ofoqueue Number of packets queued in OFO queue��
@ network.tcp.ofodrop Number of packets meant to be queued in OFO but dropped due to limits hit�Number of packets meant to be queued in OFO but dropped because socket rcvbuf
limit reached.�
@ network.tcp.ofomerge Number of packets in OFO that were merged with other packets��
@ network.tcp.challengeack Number of challenge ACKs sent (RFC 5961 3.2)��
@ network.tcp.synchallenge Number of challenge ACKs sent in response to SYN packets��
@ network.tcp.fastopenactive Number of successful active fast opens��
@ network.tcp.fastopenactivefail Number of fast open attempts failed due to remote not accepting it or time outs��
@ network.tcp.fastopenpassive Number of successful passive fast opens��
@ network.tcp.fastopenpassivefail Number of passive fast open attempts failed��
@ network.tcp.fastopenlistenoverflow Number of times the fastopen listen queue overflowed��
@ network.tcp.fastopencookiereqd Number of fast open cookies requested��
@ network.tcp.spuriousrtxhostqueues Number of times that the fast clone is not yet freed in tcp_transmit_skb()��
@ network.tcp.busypollrxpackets Number of low latency application-fetched packets��
@ network.tcp.autocorking Number of times stack detected skb was underused and its flush was deferred��
@ network.tcp.fromzerowindowadv Number of times window went from zero to non-zero��
@ network.tcp.tozerowindowadv Number of times window went from non-zero to zero��
@ network.tcp.wantzerowindowadv Number of times zero window announced��
@ network.tcp.synretrans Number of SYN-SYN/ACK retransmits�Number of SYN-SYN/ACK retransmits to break down retransmissions in SYN, fast/timeout
retransmits.�
@ network.tcp.origdatasent Number of outgoing packets with original data�Excluding retransmission but including data-in-SYN). This counter is different from
TcpOutSegs because TcpOutSegs also tracks pure ACKs. TCPOrigDataSent is
more useful to track the TCP retransmission rate.�
@ network.mptcp.mpcapablesynrx Multipath TCP received SYN with MP_CAPABLE��
@ network.mptcp.mpcapableackrx Multipath TCP received third ACK with MP_CAPABLE��
@ network.mptcp.mpcapablefallbackack Multipath TCP server-side fallback during 3-way handshake��
@ network.mptcp.mpcapablefallbacksynack Multipath TCP client-side fallback during 3-way handshake��
@ network.mptcp.mptcpretrans Multipath TCP segments retransmitted at the MPTCP-level��
@ network.mptcp.mpjoinnotokenfound Multipath TCP received MP_JOIN but the token was not found��
@ network.mptcp.mpjoinsynrx Multipath TCP received a SYN and MP_JOIN��
@ network.mptcp.mpjoinsynackrx Multipath TCP received a SYN/ACK and MP_JOIN��
@ network.mptcp.mpjoinsynackhmacfailure Multipath TCP HMAC was wrong on SYN/ACK and MP_JOIN��
@ network.mptcp.mpjoinackrx Multipath TCP received an ACK and MP_JOIN��
@ network.mptcp.mpjoinackhmacfailure Multipath TCP HMAC was wrong on ACK and MP_JOIN��
@ network.mptcp.dssnotmatching Multipath TCP received a new mapping that did not match the previous one��
@ network.mptcp.infinitemaprx Multipath TCP received an infinite mapping��
@ pmda.uname identity and type of current system�Identity and type of current system.  The concatenation of the values
returned from utsname(2), also similar to uname -a.

See also the kernel.uname.* metrics�
@ pmda.version build version of Linux PMDA��
@ hinv.map.cpu_num logical to physical CPU mapping for each CPU��
@ hinv.map.cpu_node logical CPU to NUMA node mapping for each CPU��
@ hinv.machine hardware identifier as reported by uname(2)��
@ hinv.cpu.clock clock rate in Mhz for each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.vendor manufacturer of each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.model model number of each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.model_name model name of each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.stepping stepping of each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.cache primary cache size of each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.bogomips bogo mips rating for each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.flags Hardware capability flags for each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.cache_alignment Cache alignment for each CPU as reported by /proc/cpuinfo��
@ hinv.cpu.online CPU online state from /sys/devices/system/cpu/*/online��
@ hinv.node.online NUMA node online state from /sys/devices/system/node/*/online��
@ hinv.cpu.thermal_throttle.core.count CPU core throttles from /sys/devices/system/cpu/*/thermal_throttle��
@ hinv.cpu.thermal_throttle.core.time CPU core throttle time from /sys/devices/system/cpu/*/thermal_throttle��
@ hinv.cpu.thermal_throttle.package.count CPU package throttles from /sys/devices/system/cpu/*/thermal_throttle��
@ hinv.cpu.thermal_throttle.package.time CPU package throttle time from /sys/devices/system/cpu/*/thermal_throttle��
@ hinv.cpu.frequency_scaling.count CPU frequency scaled count from /sys/devices/system/cpu/*/cpufreq��
@ hinv.cpu.frequency_scaling.time CPU frequency scaled time from /sys/devices/system/cpu/*/cpufreq��
@ hinv.cpu.frequency_scaling.max Minimum scaled CPU frequency from /sys/devices/system/cpu/*/cpufreq��
@ hinv.cpu.frequency_scaling.min Maximum scaled CPU frequency from /sys/devices/system/cpu/*/cpufreq��
@ kernel.all.hz value of HZ (jiffies/second) for the currently running kernel��
@ kernel.all.uptime time the current kernel has been running��
@ kernel.all.idletime time the current kernel has been idle since boot��
@ kernel.all.lastpid most recently allocated process identifier��
@ kernel.all.pid_max maximum process identifier from /proc/sys/kernel/pid_max��
@ kernel.all.nptys number of in-use pseudo-ttys from /proc/sys/kernel/pty/nr��
@ kernel.all.runnable total number of processes in the (per-CPU) run queues��
@ kernel.all.nprocs total number of processes (lightweight)��
@ mem.slabinfo.objects.active number of active objects in each cache��
@ mem.slabinfo.objects.total total number of objects in each cache��
@ mem.slabinfo.objects.size size of individual objects of each cache��
@ mem.slabinfo.slabs.active number of active slabs comprising each cache��
@ mem.slabinfo.slabs.total total number of slabs comprising each cache��
@ mem.slabinfo.slabs.pages_per_slab number of pages in each slab��
@ mem.slabinfo.slabs.objects_per_slab number of objects in each slab��
@ mem.slabinfo.slabs.total_size total number of bytes allocated for active objects in each slab��
@ ipc.sem.max_semmap maximum number of entries in a semaphore map (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_semid maximum number of semaphore identifiers (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_sem maximum number of semaphores in system (from semctl(..,IPC_INFO,..))��
@ ipc.sem.num_undo number of undo structures in system (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_perid maximum number of semaphores per identifier (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_ops maximum number of operations per semop call (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_undoent maximum number of undo entries per process (from semctl(..,IPC_INFO,..))��
@ ipc.sem.sz_semundo size of struct sem_undo (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_semval semaphore maximum value (from semctl(..,IPC_INFO,..))��
@ ipc.sem.max_exit adjust on exit maximum value (from semctl(..,IPC_INFO,..))��
@ ipc.sem.used_sem number of semaphore sets currently on the system (from semctl(..,SEM_INFO,..))��
@ ipc.sem.tot_sem number of semaphores in all sets on the system (from semctl(..,SEM_INFO,..))��
@ ipc.sem.key key of these semaphore (from msgctl(..,SEM_STAT,..))��
@ ipc.sem.owner username of owner (from msgctl(..,SEM_STAT,..))��
@ ipc.sem.perms access permissions (from msgctl(..,SEM_STAT,..))��
@ ipc.sem.nsems number of semaphore (from semctl(..,SEM_STAT,..))��
@ ipc.msg.sz_pool size of message pool in kilobytes (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.mapent number of entries in a message map (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.max_msgsz maximum size of a message in bytes (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.max_defmsgq default maximum size of a message queue (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.max_msgqid maximum number of message queue identifiers (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.max_msgseg message segment size (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.num_smsghdr number of system message headers (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.max_seg maximum number of message segments (from msgctl(..,IPC_INFO,..))��
@ ipc.msg.used_queues number of message queues that currently exist (from msgctl(..,MSG_INFO,..))��
@ ipc.msg.tot_msg total number of messages in all queues (from msgctl(..,MSG_INFO,..))��
@ ipc.msg.tot_bytes number of bytes in all messages in all queues (from msgctl(..,MSG_INFO,..))��
@ ipc.msg.key name of these messages slot (from msgctl(..,MSG_STAT,..))��
@ ipc.msg.owner username of owner (from msgctl(..,MSG_STAT,..))��
@ ipc.msg.perms access permissions (from msgctl(..,MSG_STAT,..))��
@ ipc.msg.msgsz used size in bytes (from msgctl(..,MSG_STAT,..))��
@ ipc.msg.messages number of messages currently queued (from msgctl(..,MSG_STAT,..))��
@ ipc.msg.last_send_pid last process to send on each message queue��
@ ipc.msg.last_recv_pid last process to recv on each message queue��
@ ipc.shm.max_segsz maximum shared segment size in bytes (from shmctl(..,IPC_INFO,..))��
@ ipc.shm.min_segsz minimum shared segment size in bytes (from shmctl(..,IPC_INFO,..))��
@ ipc.shm.max_seg maximum number of shared segments in system (from shmctl(..,IPC_INFO,..))��
@ ipc.shm.max_segproc maximum number of shared segments per process (from shmctl(..,IPC_INFO,..))��
@ ipc.shm.max_shmsys maximum amount of shared memory in system in pages (from shmctl(..,IPC_INFO,..))��
@ ipc.shm.tot total number of shared memory pages (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.rss number of resident shared memory pages (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.swp number of swapped shared memory pages (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.used_ids number of currently existing segments (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.swap_attempts number of swap attempts (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.swap_successes number of swap successes (from shmctl(..,SHM_INFO,..))��
@ ipc.shm.key Key supplied to shmget (from shmctl(.., SHM_STAT, ..))��
@ ipc.shm.owner share memory segment owner (rom shmctl(.., SHM_STAT, ..))��
@ ipc.shm.perms operation perms (from shmctl(.., SHM_STAT, ..))��
@ ipc.shm.segsz size of segment (bytes) (from shmctl(.., SHM_STAT, ..))��
@ ipc.shm.nattch no. of current attaches (from shmctl(.., SHM_STAT, ..))��
@ ipc.shm.status share memory segment status (from shmctl(.., SHM_STAT, ..))�The string value may contain the space-separated values "dest" (a shared memory
segment marked for destruction on last detach) and "locked" or the empty string.�
@ ipc.shm.creator_pid process creating each shared memory segment��
@ ipc.shm.last_access_pid process last accessing each shared memory segment��
@ vfs.files.count number of in-use file structures��
@ vfs.files.free number of available file structures��
@ vfs.files.max hard maximum on number of file structures��
@ vfs.inodes.count number of in-use inode structures��
@ vfs.inodes.free number of available inode structures��
@ vfs.dentry.count number of in-use directory entry structures��
@ vfs.dentry.free number of available directory entry structures��
@ vfs.aio.count number of in-use asynchronous IO structures��
@ vfs.aio.max hard maximum on number of asynchronous IO structures��
@ vfs.locks.posix.read number of POSIX locks held for reading��
@ vfs.locks.posix.write number of POSIX locks held for writing��
@ vfs.locks.posix.count number of POSIX lock structures��
@ vfs.locks.flock.read number of advisory file locks held for reading��
@ vfs.locks.flock.write number of advisory file locks held for writing��
@ vfs.locks.flock.count number of advisory file lock structures��
@ vfs.locks.lease.read number of file leases held for reading��
@ vfs.locks.lease.write number of file leases held for writing��
@ vfs.locks.lease.count number of file lease structures��
@ sysfs.kernel.uevent_seqnum counter of the number of uevents processed by the udev subsystem��
@ hinv.ntape number of Linux scsi tape devices��
@ tape.dev.in_flight number of I/Os currently outstanding to this tape device��
@ tape.dev.io_ns cumulative amount of time spent waiting for all I/O to complete to tape device�The amount of time spent waiting (in nanoseconds) for all I/O to complete
(including read and write). This includes tape movement commands such as seeking
between file or set marks and implicit tape movement such as when rewind on close
tape devices are used.�
@ tape.dev.other_cnt number of I/Os issued to the tape drive other than read or write commands��
@ tape.dev.read_byte_cnt number of bytes read from the tape drive��
@ tape.dev.read_cnt number of read requests issued to the tape drive��
@ tape.dev.read_ns cumulative amount of time spent waiting for read requests to complete��
@ tape.dev.resid_cnt count of read or write residual data, per tape device�Number of times during a read or write we found the residual amount to be non-zero.
For reads this means a program is issuing a read larger than the block size on tape.
For writes it means not all data made it to tape.�
@ tape.dev.write_byte_cnt number of bytes written to the tape drive��
@ tape.dev.write_cnt number of write requests issued to the tape drive��
@ tape.dev.write_ns cumulative amount of time spent waiting for write requests to complete��
@ tty.serial.rx Number of receive interrupts for current serial line.��
@ tty.serial.tx Number of transmit interrupts for current serial line.��
@ tty.serial.frame Number of frame errors for current serial line.��
@ tty.serial.parity Number of parity errors for current serial line.��
@ tty.serial.brk Number of breaks for current serial line.��
@ tty.serial.overrun Number of overrun errors for current serial line.��
@ tty.serial.irq IRQ number.��
@ kernel.all.pressure.cpu.some.avg Percentage of time runnable processes delayed for CPU resources�Indicates the time in which at least some tasks stalled on CPU resources.
The ratios are tracked as recent trends over ten second, one minute,
and five minute windows.
Pressure stall information (PSI) from /proc/pressure/cpu.�
@ kernel.all.pressure.cpu.some.total Total time processes stalled for CPU resources�Indicates the time in which at least some tasks stalled on CPU resources.
Pressure stall information (PSI) from /proc/pressure/cpu.�
@ kernel.all.pressure.memory.some.avg Percentage of time runnable processes delayed for memory resources�Indicates the time in which at least some tasks stalled on memory resources.
The ratios are tracked as recent trends over ten second, one minute,
and five minute windows.
Pressure stall information (PSI) from /proc/pressure/memory.�
@ kernel.all.pressure.memory.some.total Total time processes stalled for memory resources�The CPU time for which at least some tasks stalled on memory resources.
Pressure stall information (PSI) from /proc/pressure/memory.�
@ kernel.all.pressure.memory.full.avg Percentage of time all work is delayed from memory pressure�Indicates the time in which all tasks stalled on memory resources.
The ratios are tracked as recent trends over ten second, one minute,
and five minute windows.
Pressure stall information (PSI) from /proc/pressure/memory.�
@ kernel.all.pressure.memory.full.total Total time when all tasks stall on memory resources�The CPU time for which all tasks stalled on memory resources.
Pressure stall information (PSI) from /proc/pressure/memory.�
@ kernel.all.pressure.io.some.avg Percentage of time runnable processes delayed for IO resources�Indicates the time in which at least some tasks stalled on IO resources.
The ratios are tracked as recent trends over ten second, one minute,
and five minute windows.
Pressure stall information (PSI) from /proc/pressure/io.�
@ kernel.all.pressure.io.some.total Total time processes stalled for IO resources�The CPU time in which at least some tasks stalled on IO resources.
Pressure stall information (PSI) from /proc/pressure/io.�
@ kernel.all.pressure.io.full.avg Percentage of time all work is delayed from IO pressure�Indicates the time in which all tasks stalled on IO resources.
The ratios are tracked as recent trends over ten second, one minute,
and five minute windows.
Pressure stall information (PSI) from /proc/pressure/io.�
@ kernel.all.pressure.io.full.total Total time when all tasks stall on IO resources�The CPU time in which all tasks stalled on IO resources.
Pressure stall information (PSI) from /proc/pressure/io.�
@ zram.read read operations metric for compressed memory�Cumulative number of disk read operations since system boot time
(subject to counter wrap) for compressed memory devices.�
@ zram.write write operations metric for compressed memory�Cumulative number of disk write operations since system boot time
(subject to counter wrap) for compressed memory devices.�
@ zram.total total (read+write) I/O operations for compressed memory�Cumulative number of disk read and write operations since system boot
time (subject to counter wrap) for compressed memory devices.�
@ zram.blkread block read operations metric for compressed memory�Cumulative number of disk block read operations since system boot time
(subject to counter wrap) for compressed memory devices.�
@ zram.blkwrite block write operations metric for compressed memory�Cumulative number of disk block write operations since system boot time
(subject to counter wrap) for compressed memory devices.�
@ zram.blktotal total (read+write) block operations for compressed memory�Cumulative number of disk block read and write operations since system
boot time (subject to counter wrap) for compressed memory devices.�
@ zram.read_bytes number of bytes read for compressed memory devices�Cumulative number of bytes read since system boot time (subject to
counter wrap) for compressed memory devices.�
@ zram.write_bytes number of bytes written for compressed memory devices�Cumulative number of bytes written since system boot time (subject to
counter wrap) for compressed memory devices.�
@ zram.total_bytes total number of bytes read and written for compressed memory�Cumulative number of bytes read and written since system boot time
(subject to counter wrap) for compressed memory devices.�
@ zram.read_merge per-compressed-memory-device count of merged read requests��
@ zram.write_merge per-compressed-memory-device count of merged write requests��
@ zram.avactive per-compressed-memory-device count of active time�Counts the number of milliseconds for which at least one I/O is in
progress for each compressed memory device.

When converted to a rate, this metric represents the average utilization
of the compressed memory device during the sampling interval.  A value of 0.5
(or 50%) means the memory device was active (i.e. busy) half the time.�
@ zram.aveq per-compressed-memory-device time averaged count of request queue length��
@ zram.read_rawactive per-compressed-memory-device raw count of read response time�For each completed read on each compressed memory device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding reads for a compressed memory device.  When
divided by the number of completed reads for a zram device (zram.read),
the value represents the stochastic average of the read response (or wait)
time for that compressed memory device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.zram.r_await = delta(zram.read_rawactive) / delta(zram.read)�
@ zram.write_rawactive per-compressed-memory-device raw count of write response time�For each completed write on each compressed memory device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding writes for a zram device.
When divided by the number of completed writes for a zram device
(zram.write), the value represents the stochastic average of
the write response (or wait) time for that zram device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.zram.w_await = delta(zram.write_rawactive) / delta(zram.write)�
@ zram.total_rawactive per-compressed-memory-device raw count of I/O response time�For each completed I/O on each compressed memory device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time
average of the number of outstanding I/Os for a zram device.
When divided by the number of completed I/Os for a zram device
(zram.total), the value represents the stochastic average of
the I/O response (or wait) time for that zram device.�
@ zram.capacity per-compressed-memory-device capacity�Total space presented by each zram device, from /proc/partitions.�
@ zram.discard discard operations count for compressed memory devices��
@ zram.blkdiscard block discard operations count for compressed memory device��
@ zram.discard_bytes number of discard bytes for compressed memory device��
@ zram.discard_merge per-zram-device count of merged discard requests��
@ zram.discard_rawactive per-zram-device raw count of discard response time�For each completed discard on each compressed memory device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding discards for a zram device.  When divided by
the number of completed discards for a zram device (zram.discard),
the value represents the stochastic average of the discard response (or wait)
time for that compressed memory device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.zram.d_await = delta(zram.discard_rawactive) / delta(zram.discard)�
@ zram.flush flush operations metric for compressed memory devices��
@ zram.flush_rawactive per-zram-device raw count of flush response time�For each completed flush on each compressed memory device the response time
(queue time plus service time) in milliseconds is added to the associated
instance of this metric.

When converted to a normalized rate, the value represents the time average
of the number of outstanding flushes for a zram device.  When divided by
the number of completed flushes for a compressed memory device (zram.flush),
the value represents the stochastic average of the flush response (or wait)
time for that compressed memory device.

It is suitable mainly for use in calculations with other metrics,
e.g. mirroring the results from existing performance tools:

iostat.zram.f_await = delta(zram.flush_rawactive) / delta(zram.flush)�
@ zram.io_stat.failed.reads number of failed reads��
@ zram.io_stat.failed.writes number of failed writes��
@ zram.io_stat.invalid number of non-page-size-aligned I/O requests��
@ zram.io_stat.notify_free number of pages freed�Depending on device usage scenario it may account:
- the number of pages freed because of swap slot free notifications
- the number of pages freed because of REQ_OP_DISCARD requests sent by
bio. The former ones are sent to a swap block device when a swap slot
is freed, which implies that this disk is being used as a swap disk.

The latter ones are sent by filesystem mounted with discard option,
whenever some data blocks are getting discarded.�
@ zram.mm_stat.data_size.original uncompressed data stored in this disk��
@ zram.mm_stat.data_size.compressed compressed data stored in this disk��
@ zram.mm_stat.mem.used_total amount of memory allocated for this disk�This includes allocator fragmentation and metadata overhead, allocated
for this disk.  So, allocator space efficiency can be calculated using
zram.mm_stat.data_size.compressed and this statistic.�
@ zram.mm_stat.mem.limit maximum amount of memory zram can use to store compressed data��
@ zram.mm_stat.mem.max_used maximum amount of memory zram has consumed to store the data��
@ zram.mm_stat.pages.same number of same element filled pages written to this disk��
@ zram.mm_stat.pages.compacted pages freed during compaction��
@ zram.mm_stat.pages.huge the number of incompressible pages��
@ zram.bd_stat.count size of data written in backing device��
@ zram.bd_stat.reads the number of reads from backing device��
@ zram.bd_stat.writes the number of writes to backing device��
@ network.all.in.bytes network recv bytes from physical network interfaces�Sum of bytes column on the "Receive" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.in.packets network recv packets from physical network interfaces�Sum of packets column on the "Receive" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.in.errors network recv errors from physical network interfaces�Sum of errors column on the "Receive" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.in.drops network recv drops from physical network interfaces�Sum of drop column on the "Receive" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.out.bytes network bytes sent from physical network interfaces�Sum of bytes column on the "Transmit" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.out.packets network packets sent from physical network interfaces�Sum of packets column on the "Transmit" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.out.errors network send errors from physical network interfaces�Sum of errors column on the "Transmit" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.out.drops network send drops from physical network interfaces�Sum of drop column on the "Transmit" side of /proc/net/dev for network
interfaces deemed to be 'physical' interfaces, using regular expression
pattern described in the $PCP_SYSCONF_DIR/linux/interfaces.conf file.�
@ network.all.total.bytes total bytes on physical network interfaces�Sum of network.all.in.bytes and network.all.out.bytes metrics.�
@ network.all.total.packets total packets on physical network interfaces�Sum of network.all.in.packets and network.all.out.packets metrics.�
@ network.all.total.errors network errors on physical network interfaces�Sum of network.all.in.errors and network.all.out.errors metrics.�
@ network.all.total.drops network drops on physical network interfaces�Sum of network.all.in.drops and network.all.out.drops metrics.�
@ fchost.in.frames Count of fibre channel frames received�The total number of Fibre Channel (FC) frames that have been received
by the Host Bus Adapter (HBA).
This count is across all protocols and classes, which includes general
services etc. as well as frames associated with SCSI traffic.�
@ fchost.in.bytes Count of fibre channel bytes received�The total number of received bytes by the Host Bus Adapter (HBA).
This count is across all protocols and classes, which includes general
services etc. as well as frames associated with SCSI traffic.
Also this counts all received words, frame headers, CRC, and so on,
and not only "user" data bytes.�
@ fchost.out.frames Count of fibre channel frames transmitted�The total number of Fibre Channel (FC) frames that have been transmitted
by the Host Bus Adapter (HBA).
This count is across all protocols and classes, which includes general
services etc. as well as frames associated with SCSI traffic.�
@ fchost.out.bytes Count of fibre channel bytes transmitted�The total number of transmitted bytes by the Host Bus Adapter (HBA).
This count is across all protocols and classes, which includes general
services etc. as well as frames associated with SCSI traffic.
Also this counts all transmitted words, frame headers, CRC, and so on,
and not only "user" data bytes.�
@ fchost.lip_count Count of loop initializaton FC primitives�The number of LIP (loop initializaton FC primitive) resets that have
occurred.
This should be the number initiated by the HBA unless there are other
ports on the link the HBA is connected to that can also issue the LIP
primitive.
One can send a LIP by writing to /sys/class/fc_host/hostN/issue_lip
A LIP will cause temporary loss of link (link down/link up events).�
@ fchost.nos_count Count of not operational FC primitives�The number of NOS (not operational) FC primitives that have occurred
on the switched fabric.
This would typically be recieved by the HBA during link initialization
between the HBA port and the switch if the switch detected a problem -
typically NOS is sent by a port that is offline or has detected a link
problem or failure of some type.  This being non-zero implies problems
at the link level or with the switch port the HBA is connected to.�
@ fchost.error_frames Count of FC frames received in error��
@ fchost.dumped_frames Count of FC frames lost due to buffers�The count of FC frames that were lost due to lack of local resources
(buffers).
A frame arrives at the HBA nport, but there is no place to capture it
due to lack of available buffers within the adapter.
The frame is "dumped", i.e. dropped and the firmware never sees it.
Something is not working with buffer credits between ports at a lower
FC link level if this is happening (one guess as to why dumped frames
could occur).�
@ hinv.nfchost Number of fibre channel host bus adapters from /sys/class/fc_host/host*��
@ network.mptcp.mpcapablesynackrx Received SYN/ACK with MP_CAPABLE flag�The MPCapableSYNACKRX field of the MPTcp line from /proc/net/netstat�
@ network.mptcp.mpcapablesyntx Sent SYN with MP_CAPABLE flag�The MPCapableSYNTX field of the MPTcp line from /proc/net/netstat�
@ network.mptcp.mpfallbacktokeninit Could not init/allocate token�The MPFallbackTokenInit field of the MPTcp line from /proc/net/netstat�
@ network.tcp.pfmemallocdrop Dropped skb allocated from pfmemalloc�Just counts the cases for packets which did not have the SOCK_MEMALLOC flag
set.

The PFMemallocDrop field of the Tcp line from /proc/net/netstat�
@ network.tcp.tcpbacklogcoalesce Number of coalesced packets that were in the backlog queue�The TCPBacklogCoalesce field of the Tcp line from /proc/net/netstat�
@ network.tcp.tcpfastopenactivefail Fast Open attempts (SYN/data) failed�Failure because the remote does not accept it or the attempts timed out.

The TCPFastOpenActiveFail field of the Tcp line from /proc/net/netstat�
@ network.tcp.tcpfastopenblackhole Number of times the TFO blackhole has been enabled�The TCPFastOpenBlackhole field of the Tcp line from /proc/net/netstat�
@ network.tcp.tcpmd5failure Counter for drops caused by md5 mismatches�The TCPMD5Failure field of the Tcp line from /proc/net/netstat�
@ network.tcp.tcpmemorypressureschrono Cumulative counter tracking duration of memory pressure events�the Z field of the Tcp line from /proc/net/netstat�
@ network.udp.ignoredmulti the IgnoredMulti field of the Udp line from /proc/net/netstat��
@ network.udp.memerrors the MemErrors field of the Udp line from /proc/net/netstat��
@ network.udplite.ignoredmulti the IgnoredMulti field of the UdpLite line from /proc/net/netstat��
@ network.udplite.memerrors the MemErrors field of the UdpLite line from /proc/net/netstat��
@ network.tcp.tcphystarttraindetect the TCPHystartTrainDetect field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcphystarttraincwnd the TCPHystartTrainCwnd field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcphystartdelaydetect the TCPHystartDelayDetect field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcphystartdelaycwnd the TCPHystartDelayCwnd field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedsynrecv the TCPACKSkippedSynRecv field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedpaws the TCPACKSkippedPAWS field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedseq the TCPACKSkippedSeq field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedfinwait2 the TCPACKSkippedFinWait2 field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedtimewait the TCPACKSkippedTimeWait field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackskippedchallenge the TCPACKSkippedChallenge field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpwinprobe the TCPWinProbe field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpkeepalive the TCPKeepAlive field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpmtupfail the TCPMTUPFail field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpmtupsuccess the TCPMTUPSuccess field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpdelivered the TCPDelivered field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpdeliveredce the TCPDeliveredCE field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpackcompressed the TCPAckCompressed field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpzerowindowdrop the TCPZeroWindowDrop field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcprcvqdrop the TCPRcvQDrop field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpwqueuetoobig the TCPWqueueTooBig field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpfastopenpassivealtkey the TCPFastOpenPassiveAltKey field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcptimeoutrehash the TcpTimeoutRehash field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpduplicatedatarehash the TcpDuplicateDataRehash field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpdsackrecvsegs the TCPDSACKRecvSegs field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpdsackignoreddubious the TCPDSACKIgnoredDubious field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpmigratereqsuccess the TCPMigrateReqSuccess field of the Tcp line from /proc/net/netstat��
@ network.tcp.tcpmigratereqfailure the TCPMigrateReqFailure field of the Tcp line from /proc/net/netstat��
@ network.ip.reasmoverlaps the ReasmOverlaps field of the Ip line from /proc/net/netstat��
@ network.mptcp.dssnomatchtcp the DSSNoMatchTCP field of the MPTcp line from /proc/net/netstat�Multipath TCP DSS-mapping did not map with TCP's sequence numbers.�
@ network.mptcp.datacsumerr the DataCsumErr field of the MPTcp line from /proc/net/netstat�Multipath TCP data checksum failures.�
@ network.mptcp.ofoqueuetail the OFOQueueTail field of the MPTcp line from /proc/net/netstat�Multipath TCP segments inserted into OoO queue tail.�
@ network.mptcp.ofoqueue the OFOQueue field of the MPTcp line from /proc/net/netstat�Multipath TCP segments inserted into OoO queue.�
@ network.mptcp.ofomerge the OFOMerge field of the MPTcp line from /proc/net/netstat�Multipath TCP segments merged in OoO queue.�
@ network.mptcp.nodssinwindow the NoDSSInWindow field of the MPTcp line from /proc/net/netstat�Multipath TCP segments not in MPTCP windows.�
@ network.mptcp.duplicatedata the DuplicateData field of the MPTcp line from /proc/net/netstat�Multipath TCP segments discarded due to duplicate DSS.�
@ network.mptcp.addaddr the AddAddr field of the MPTcp line from /proc/net/netstat�Multipath TCP received ADD_ADDR with echo-flag=0.�
@ network.mptcp.echoadd the EchoAdd field of the MPTcp line from /proc/net/netstat�Multipath TCP received ADD_ADDR with echo-flag=1.�
@ network.mptcp.portadd the PortAdd field of the MPTcp line from /proc/net/netstat�Multipath TCP received ADD_ADDR with a port-number.�
@ network.mptcp.mpjoinportsynrx the MPJoinPortSynRx field of the MPTcp line from /proc/net/netstat�Multipath TCP received a SYN MP_JOIN with a different port-number.�
@ network.mptcp.mpjoinportsynackrx the MPJoinPortSynAckRx field of the MPTcp line from /proc/net/netstat�Multipath TCP received a SYNACK MP_JOIN with a different port-number.�
@ network.mptcp.mpjoinportackrx the MPJoinPortAckRx field of the MPTcp line from /proc/net/netstat�Multipath TCP received an ACK MP_JOIN with a different port-number.�
@ network.mptcp.mismatchportsynrx the MismatchPortSynRx field of the MPTcp line from /proc/net/netstat�Multipath TCP received a SYN MP_JOIN with a mismatched port-number.�
@ network.mptcp.mismatchportackrx the MismatchPortAckRx field of the MPTcp line from /proc/net/netstat�Multipath TCP received an ACK MP_JOIN with a mismatched port-number.�
@ network.mptcp.rmaddr the RmAddr field of the MPTcp line from /proc/net/netstat�Multipath TCP RM_ADDR receives.�
@ network.mptcp.rmsubflow the RmSubflow field of the MPTcp line from /proc/net/netstat�Multipath TCP subflows removed.�
@ network.mptcp.mppriotx the MPPrioTx field of the MPTcp line from /proc/net/netstat�Multipath TCP MP_PRIO transmits.�
@ network.mptcp.mppriorx the MPPrioRx field of the MPTcp line from /proc/net/netstat�Multipath TCP MP_PRIO receives.�
@ network.mptcp.rcvpruned the RcvPruned field of the MPTcp line from /proc/net/netstat�Incoming multipath packet dropped due to memory limit.�
@ network.tcp.tcploss the TCPLOSS field of the Tcp line from /proc/net/netstat��
@ network.mptcp.mpfailtx the MPFailTx field of the MPTcp line from /proc/net/netstat�Multipath TCP MP_FAIL transmits.�
@ network.mptcp.mpfailrx the MPFailRx field of the MPTcp line from /proc/net/netstat�Multipath TCP MP_FAIL receives.�
@ network.mptcp.subflowstale the SubflowStale field of the MPTcp line from /proc/net/netstat�Multipath TCP subflows entered 'stale' status.�
@ network.mptcp.subflowrecover the SubflowRecover field of the MPTcp line from /proc/net/netstat�Multipath TCP subflows returned to active status after being stale.�
@ network.mptcp.addaddrdrop the AddAddrDrop field of the MPTcp line from /proc/net/netstat�Multipath TCP incoming ADD_ADDR drops.�
@ network.mptcp.rmaddrdrop the RmAddrDrop field of the MPTcp line from /proc/net/netstat�Multipath TCP incoming RM_ADDR drops.�
@ disk.wwid.scsi_paths per-device/wwid scsi paths to same physical device�The value of this string typed metric is the sd paths to the same physical
device identified by it's WWID. Non-multipath devices will have only one
such path. Multipath devices can have more than one scsi device path, e.g
via multiple HBAs to the same physical LUN via an FC switch.�
@ disk.wwid.read per-device read operations aggregated for all scsi paths��
@ disk.wwid.write per-device write operations aggregated for all scsi paths��
@ disk.wwid.total per-device total (read+write) operations aggregated for all scsi paths��
@ disk.wwid.blkread per-device block read operations aggregated for all scsi paths��
@ disk.wwid.blkwrite per-device block write operations aggregated for all scsi paths��
@ disk.wwid.blktotal per-device total (read+write) block operations aggregated for all scsi paths��
@ disk.wwid.read_bytes per-device count of bytes read aggregated for all scsi paths��
@ disk.wwid.write_bytes per-device count of bytes written aggregated for all scsi paths��
@ disk.wwid.total_bytes per-device count of total bytes read and written aggregated for all scsi paths��
@ disk.wwid.scheduler per-device I/O scheduler aggregated for all scsi paths��
@ disk.wwid.avactive per-device count of active time aggregated for all scsi paths��
@ disk.wwid.read_rawactive per-device raw count of read response time aggregated for all scsi paths��
@ disk.wwid.write_rawactive per-device raw count of write response time aggregated for all scsi paths��
@ disk.wwid.total_rawactive per-device raw count of I/O response time aggregated for all scsi paths��
@ disk.wwid.aveq per-device time averaged count of request queue length aggregated for all scsi paths��
@ disk.wwid.read_merge per-device count of merged read requests aggregated for all scsi paths��
@ disk.wwid.write_merge per-device count of merged write requests aggregated for all scsi paths��
@ disk.wwid.capacity per-device physical device capacity aggregated for all scsi paths��
@ disk.wwid.discard per-device discard operations aggregated for all scsi paths��
@ disk.wwid.blkdiscard per-device block discard operations aggregated for all scsi paths��
@ disk.wwid.discard_bytes per-device count of bytes discard'ed aggregated for all scsi paths��
@ disk.wwid.discard_merge per-device count of merged discard requests aggregated for all scsi paths��
@ disk.wwid.discard_rawactive per-device raw count of discard response time aggregated for all scsi paths��
@ disk.wwid.flush per-device flush operations aggregated for all scsi paths��
@ disk.wwid.flush_rawactive per-device raw count of flush response time aggregated for all scsi paths��

${this.title}

Code Editor : help.pag