mirror_zfs/man/man1/arcstat.1

611 lines
6.4 KiB
Groff
Raw Normal View History

.\"
.\" This file and its contents are supplied under the terms of the
.\" Common Development and Distribution License ("CDDL"), version 1.0.
.\" You may only use this file in accordance with the terms of version
.\" 1.0 of the CDDL.
.\"
.\" A full copy of the text of the CDDL should have accompanied this
.\" source. A copy of the CDDL is also available via the Internet at
.\" http://www.illumos.org/license/CDDL.
.\"
.\"
.\" Copyright 2014 Adam Stevko. All rights reserved.
.\" Copyright (c) 2015 by Delphix. All rights reserved.
.\" Copyright (c) 2020 by AJ Jordan. All rights reserved.
.\"
.TH ARCSTAT 1 "Oct 20, 2020" OpenZFS
.SH NAME
arcstat \- report ZFS ARC and L2ARC statistics
.SH SYNOPSIS
.LP
.nf
\fBarcstat\fR [\fB-havxp\fR] [\fB-f field[,field]...\fR] [\fB-o file\fR] [\fB-s string\fR] [\fBinterval\fR [\fBcount\fR]]
.fi
.SH DESCRIPTION
.LP
The \fBarcstat\fR utility print various ZFS ARC and L2ARC statistics in
vmstat-like fashion.
.sp
.sp
.LP
The \fBarcstat\fR command reports the following information:
.sp
.ne 2
.\"
.sp
.ne 1
.na
\fBc \fR
.ad
.RS 14n
ARC target size
.RE
.sp
.ne 2
.na
\fBdh% \fR
.ad
.RS 14n
Demand data hit percentage
.RE
.sp
.ne 2
.na
\fBdm% \fR
.ad
.RS 14n
Demand data miss percentage
.RE
.sp
.ne 2
.na
\fBmfu \fR
.ad
.RS 14n
MFU list hits per second
.RE
.sp
.ne 2
.na
\fBmh% \fR
.ad
.RS 14n
Metadata hit percentage
.RE
.sp
.ne 2
.na
\fBmm% \fR
.ad
.RS 14n
Metadata miss percentage
.RE
.sp
.ne 2
.na
\fBmru \fR
.ad
.RS 14n
MRU list hits per second
.RE
.sp
.ne 2
.na
\fBph% \fR
.ad
.RS 14n
Prefetch hits percentage
.RE
.sp
.ne 2
.na
\fBpm% \fR
.ad
.RS 14n
Prefetch miss percentage
.RE
.sp
.ne 2
.na
\fBdhit \fR
.ad
.RS 14n
Demand data hits per second
.RE
.sp
.ne 2
.na
\fBdmis \fR
.ad
.RS 14n
Demand data misses per second
.RE
.sp
.ne 2
.na
\fBhit% \fR
.ad
.RS 14n
ARC hit percentage
.RE
.sp
.ne 2
.na
\fBhits \fR
.ad
.RS 14n
ARC reads per second
.RE
.sp
.ne 2
.na
\fBmfug \fR
.ad
.RS 14n
MFU ghost list hits per second
.RE
.sp
.ne 2
.na
\fBmhit \fR
.ad
.RS 14n
Metadata hits per second
.RE
.sp
.ne 2
.na
\fBmiss \fR
.ad
.RS 14n
ARC misses per second
.RE
.sp
.ne 2
.na
\fBmmis \fR
.ad
.RS 14n
Metadata misses per second
.RE
.sp
.ne 2
.na
\fBmrug \fR
.ad
.RS 14n
MRU ghost list hits per second
.RE
.sp
.ne 2
.na
\fBphit \fR
.ad
.RS 14n
Prefetch hits per second
.RE
.sp
.ne 2
.na
\fBpmis \fR
.ad
.RS 14n
Prefetch misses per second
.RE
.sp
.ne 2
.na
\fBread \fR
.ad
.RS 14n
Total ARC accesses per second
.RE
.sp
.ne 2
.na
\fBtime \fR
.ad
.RS 14n
Time
.RE
.sp
.ne 2
.na
\fBsize \fR
.ad
.RS 14n
ARC size
.RE
.sp
.ne 2
.na
\fBarcsz \fR
.ad
.RS 14n
Alias for \fBsize\fR
.RE
.sp
.ne 2
.na
\fBdread \fR
.ad
.RS 14n
Demand data accesses per second
.RE
.sp
.ne 2
.na
\fBeskip \fR
.ad
.RS 14n
evict_skip per second
.RE
.sp
.ne 2
.na
\fBmiss% \fR
.ad
.RS 14n
ARC miss percentage
.RE
.sp
.ne 2
.na
\fBmread \fR
.ad
.RS 14n
Metadata accesses per second
.RE
.sp
.ne 2
.na
\fBpread \fR
.ad
.RS 14n
Prefetch accesses per second
.RE
.sp
.ne 2
.na
\fBl2hit% \fR
.ad
.RS 14n
L2ARC access hit percentage
.RE
.sp
.ne 2
.na
\fBl2hits \fR
.ad
.RS 14n
L2ARC hits per second
.RE
.sp
.ne 2
.na
\fBl2miss \fR
.ad
.RS 14n
L2ARC misses per second
.RE
.sp
.ne 2
.na
\fBl2read \fR
.ad
.RS 14n
Total L2ARC accesses per second
.RE
Add L2ARC arcstats for MFU/MRU buffers and buffer content type Currently the ARC state (MFU/MRU) of cached L2ARC buffer and their content type is unknown. Knowing this information may prove beneficial in adjusting the L2ARC caching policy. This commit adds L2ARC arcstats that display the aligned size (in bytes) of L2ARC buffers according to their content type (data/metadata) and according to their ARC state (MRU/MFU or prefetch). It also expands the existing evict_l2_eligible arcstat to differentiate between MFU and MRU buffers. L2ARC caches buffers from the MRU and MFU lists of ARC. Upon caching a buffer, its ARC state (MRU/MFU) is stored in the L2 header (b_arcs_state). The l2_m{f,r}u_asize arcstats reflect the aligned size (in bytes) of L2ARC buffers according to their ARC state (based on b_arcs_state). We also account for the case where an L2ARC and ARC cached MRU or MRU_ghost buffer transitions to MFU. The l2_prefetch_asize reflects the alinged size (in bytes) of L2ARC buffers that were cached while they had the prefetch flag set in ARC. This is dynamically updated as the prefetch flag of L2ARC buffers changes. When buffers are evicted from ARC, if they are determined to be L2ARC eligible then their logical size is recorded in evict_l2_eligible_m{r,f}u arcstats according to their ARC state upon eviction. Persistent L2ARC: When committing an L2ARC buffer to a log block (L2ARC metadata) its b_arcs_state and prefetch flag is also stored. If the buffer changes its arcstate or prefetch flag this is reflected in the above arcstats. However, the L2ARC metadata cannot currently be updated to reflect this change. Example: L2ARC caches an MRU buffer. L2ARC metadata and arcstats count this as an MRU buffer. The buffer transitions to MFU. The arcstats are updated to reflect this. Upon pool re-import or on/offlining the L2ARC device the arcstats are cleared and the buffer will now be counted as an MRU buffer, as the L2ARC metadata were not updated. Bug fix: - If l2arc_noprefetch is set, arc_read_done clears the L2CACHE flag of an ARC buffer. However, prefetches may be issued in a way that arc_read_done() is bypassed. Instead, move the related code in l2arc_write_eligible() to account for those cases too. Also add a test and update manpages for l2arc_mfuonly module parameter, and update the manpages and code comments for l2arc_noprefetch. Move persist_l2arc tests to l2arc. Reviewed-by: Ryan Moeller <freqlabs@FreeBSD.org> Reviewed-by: Richard Elling <Richard.Elling@RichardElling.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: George Amanakis <gamanakis@gmail.com> Closes #10743
2020-09-14 20:10:44 +03:00
.sp
.ne 2
.na
\fBl2pref \fR
.ad
.RS 14n
L2ARC prefetch allocated size per second
.RE
.sp
.ne 2
.na
\fBl2pref% \fR
.ad
.RS 14n
L2ARC prefetch allocated size percentage
.RE
.sp
.ne 2
.na
\fBl2mfu \fR
.ad
.RS 14n
L2ARC MFU allocated size per second
.RE
.sp
.ne 2
.na
\fBl2mfu% \fR
.ad
.RS 14n
L2ARC MFU allocated size percentage
.RE
.sp
.ne 2
.na
\fBl2mru \fR
.ad
.RS 14n
L2ARC MRU allocated size per second
.RE
.sp
.ne 2
.na
\fBl2mru% \fR
.ad
.RS 14n
L2ARC MRU allocated size percentage
.RE
.sp
.ne 2
.na
\fBl2data \fR
.ad
.RS 14n
L2ARC data (buf content) allocated size per second
.RE
.sp
.ne 2
.na
\fBl2data% \fR
.ad
.RS 14n
L2ARC data (buf content) allocated size percentage
.RE
.sp
.ne 2
.na
\fBl2meta \fR
.ad
.RS 14n
L2ARC metadata (buf content) allocated size per second
.RE
.sp
.ne 2
.na
\fBl2meta% \fR
.ad
.RS 14n
L2ARC metadata (buf content) allocated size percentage
.RE
.sp
.ne 2
.na
\fBl2size \fR
.ad
.RS 14n
Size of the L2ARC
.RE
.sp
.ne 2
.na
\fBmtxmis \fR
.ad
.RS 14n
mutex_miss per second
.RE
.sp
.ne 2
.na
\fBl2bytes \fR
.ad
.RS 14n
Bytes read per second from the L2ARC
.RE
.sp
.ne 2
.na
\fBl2miss% \fR
.ad
.RS 14n
L2ARC access miss percentage
.RE
.sp
.ne 2
.na
\fBl2asize \fR
.ad
.RS 14n
Actual (compressed) size of the L2ARC
.RE
.sp
.ne 2
.na
\fBgrow \fR
.ad
.RS 14n
ARC grow disabled
.RE
.sp
.ne 2
.na
\fBneed \fR
.ad
.RS 14n
ARC reclaim needed
.RE
.sp
.ne 2
.na
\fBfree \fR
.ad
.RS 14n
The ARC's idea of how much free memory there is, which includes evictable memory in the page cache.
Since the ARC tries to keep \fBavail\fR above zero, \fBavail\fR is usually more instructive to observe than \fBfree\fR.
.RE
.sp
.ne 2
.na
\fBavail \fR
.ad
.RS 14n
The ARC's idea of how much free memory is available to it, which is a bit less than \fBfree\fR.
May temporarily be negative, in which case the ARC will reduce the target size \fBc\fR.
.RE
.\"
.SH OPTIONS
.LP
The following options are supported:
.sp
.ne 2
.na
\fB\fB-a\fR\fR
.ad
.RS 12n
Print all possible stats.
.RE
.sp
.ne 2
.na
\fB\fB-f\fR\fR
.ad
.RS 12n
Display only specific fields. See \fBDESCRIPTION\fR for supported statistics.
.RE
.sp
.ne 2
.na
\fB\fB-h\fR\fR
.ad
.RS 12n
Display help message.
.RE
.sp
.ne 2
.na
\fB\fB-o\fR\fR
.ad
.RS 12n
Report statistics to a file instead of the standard output.
.RE
.sp
.ne 2
.na
\fB\fB-p\fR\fR
.ad
.RS 12n
Disable auto-scaling of numerical fields (for raw, machine-parsable values).
.RE
.sp
.ne 2
.na
\fB\fB-s\fR\fR
.ad
.RS 12n
Display data with a specified separator (default: 2 spaces).
.RE
.sp
.ne 2
.na
\fB\fB-x\fR\fR
.ad
.RS 12n
Print extended stats (same as -f time,mfu,mru,mfug,mrug,eskip,mtxmis,dread,pread,read).
.RE
.sp
.ne 2
.na
\fB\fB-v\fR\fR
.ad
.RS 12n
Show field headers and definitions
.RE
.SH OPERANDS
.LP
The following operands are supported:
.sp
.ne 2
.na
\fB\fIcount\fR\fR
.ad
.RS 12n
Display only \fIcount\fR reports.
.RE
.sp
.ne 2
.na
\fB\fIinterval\fR\fR
.ad
.RS 12n
Specify the sampling interval in seconds.
.RE
.SH AUTHORS
.LP
arcstat was originally written in Perl by Neelakanth Nadgir and supported only ZFS ARC statistics.
Mike Harsch updated it to support L2ARC statistics.
John Hixson ported it to Python for FreeNAS over some beer, after which many individuals from the OpenZFS community continued to maintain and improve it.