blob: 59763d695b1412f8cc460718efb9d6f3ff3d449f [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3------------------------------------------------------------------------------
4/proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
6
72.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8------------------------------------------------------------------------------
9Version 1.3 Kernel version 2.2.12
10 Kernel version 2.4.0-test11-pre4
11------------------------------------------------------------------------------
12
13Table of Contents
14-----------------
15
16 0 Preface
17 0.1 Introduction/Credits
18 0.2 Legal Stuff
19
20 1 Collecting System Information
21 1.1 Process-Specific Subdirectories
22 1.2 Kernel data
23 1.3 IDE devices in /proc/ide
24 1.4 Networking info in /proc/net
25 1.5 SCSI info
26 1.6 Parallel port info in /proc/parport
27 1.7 TTY info in /proc/tty
28 1.8 Miscellaneous kernel statistics in /proc/stat
29
30 2 Modifying System Parameters
31 2.1 /proc/sys/fs - File system data
32 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
33 2.3 /proc/sys/kernel - general kernel parameters
34 2.4 /proc/sys/vm - The virtual memory subsystem
35 2.5 /proc/sys/dev - Device specific parameters
36 2.6 /proc/sys/sunrpc - Remote procedure calls
37 2.7 /proc/sys/net - Networking stuff
38 2.8 /proc/sys/net/ipv4 - IPV4 settings
39 2.9 Appletalk
40 2.10 IPX
41 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -070042 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
43 2.13 /proc/<pid>/oom_score - Display current oom-killer score
Linus Torvalds1da177e2005-04-16 15:20:36 -070044
45------------------------------------------------------------------------------
46Preface
47------------------------------------------------------------------------------
48
490.1 Introduction/Credits
50------------------------
51
52This documentation is part of a soon (or so we hope) to be released book on
53the SuSE Linux distribution. As there is no complete documentation for the
54/proc file system and we've used many freely available sources to write these
55chapters, it seems only fair to give the work back to the Linux community.
56This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
57afraid it's still far from complete, but we hope it will be useful. As far as
58we know, it is the first 'all-in-one' document about the /proc file system. It
59is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
60SPARC, AXP, etc., features, you probably won't find what you are looking for.
61It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
62additions and patches are welcome and will be added to this document if you
63mail them to Bodo.
64
65We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
66other people for help compiling this documentation. We'd also like to extend a
67special thank you to Andi Kleen for documentation, which we relied on heavily
68to create this document, as well as the additional information he provided.
69Thanks to everybody else who contributed source or docs to the Linux kernel
70and helped create a great piece of software... :)
71
72If you have any comments, corrections or additions, please don't hesitate to
73contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
74document.
75
76The latest version of this document is available online at
77http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
78
79If the above direction does not works for you, ypu could try the kernel
80mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
81comandante@zaralinux.com.
82
830.2 Legal Stuff
84---------------
85
86We don't guarantee the correctness of this document, and if you come to us
87complaining about how you screwed up your system because of incorrect
88documentation, we won't feel responsible...
89
90------------------------------------------------------------------------------
91CHAPTER 1: COLLECTING SYSTEM INFORMATION
92------------------------------------------------------------------------------
93
94------------------------------------------------------------------------------
95In This Chapter
96------------------------------------------------------------------------------
97* Investigating the properties of the pseudo file system /proc and its
98 ability to provide information on the running Linux system
99* Examining /proc's structure
100* Uncovering various information about the kernel and the processes running
101 on the system
102------------------------------------------------------------------------------
103
104
105The proc file system acts as an interface to internal data structures in the
106kernel. It can be used to obtain information about the system and to change
107certain kernel parameters at runtime (sysctl).
108
109First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
110show you how you can use /proc/sys to change settings.
111
1121.1 Process-Specific Subdirectories
113-----------------------------------
114
115The directory /proc contains (among other things) one subdirectory for each
116process running on the system, which is named after the process ID (PID).
117
118The link self points to the process reading the file system. Each process
119subdirectory has the entries listed in Table 1-1.
120
121
122Table 1-1: Process specific entries in /proc
123..............................................................................
124 File Content
125 cmdline Command line arguments
Uwe Zeisbergerc30fe7f2006-03-24 18:23:14 +0100126 cpu Current and last cpu in which it was executed (2.4)(smp)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700127 cwd Link to the current working directory
128 environ Values of environment variables
129 exe Link to the executable of this process
130 fd Directory, which contains all file descriptors
131 maps Memory maps to executables and library files (2.4)
132 mem Memory held by this process
133 root Link to the root directory of this process
134 stat Process status
135 statm Process memory status information
136 status Process status in human readable form
137 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
Mauricio Line070ad42005-09-03 15:55:10 -0700138 smaps Extension based on maps, presenting the rss size for each mapped file
Linus Torvalds1da177e2005-04-16 15:20:36 -0700139..............................................................................
140
141For example, to get the status information of a process, all you have to do is
142read the file /proc/PID/status:
143
144 >cat /proc/self/status
145 Name: cat
146 State: R (running)
147 Pid: 5452
148 PPid: 743
149 TracerPid: 0 (2.4)
150 Uid: 501 501 501 501
151 Gid: 100 100 100 100
152 Groups: 100 14 16
153 VmSize: 1112 kB
154 VmLck: 0 kB
155 VmRSS: 348 kB
156 VmData: 24 kB
157 VmStk: 12 kB
158 VmExe: 8 kB
159 VmLib: 1044 kB
160 SigPnd: 0000000000000000
161 SigBlk: 0000000000000000
162 SigIgn: 0000000000000000
163 SigCgt: 0000000000000000
164 CapInh: 00000000fffffeff
165 CapPrm: 0000000000000000
166 CapEff: 0000000000000000
167
168
169This shows you nearly the same information you would get if you viewed it with
170the ps command. In fact, ps uses the proc file system to obtain its
171information. The statm file contains more detailed information about the
172process memory usage. Its seven fields are explained in Table 1-2.
173
174
175Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
176..............................................................................
177 Field Content
178 size total program size (pages) (same as VmSize in status)
179 resident size of memory portions (pages) (same as VmRSS in status)
180 shared number of pages that are shared (i.e. backed by a file)
181 trs number of pages that are 'code' (not including libs; broken,
182 includes data segment)
183 lrs number of pages of library (always 0 on 2.6)
184 drs number of pages of data/stack (including libs; broken,
185 includes library text)
186 dt number of dirty pages (always 0 on 2.6)
187..............................................................................
188
1891.2 Kernel data
190---------------
191
192Similar to the process entries, the kernel data files give information about
193the running kernel. The files used to obtain this information are contained in
194/proc and are listed in Table 1-3. Not all of these will be present in your
195system. It depends on the kernel configuration and the loaded modules, which
196files are there, and which are missing.
197
198Table 1-3: Kernel info in /proc
199..............................................................................
200 File Content
201 apm Advanced power management info
202 buddyinfo Kernel memory allocator information (see text) (2.5)
203 bus Directory containing bus specific information
204 cmdline Kernel command line
205 cpuinfo Info about the CPU
206 devices Available devices (block and character)
207 dma Used DMS channels
208 filesystems Supported filesystems
209 driver Various drivers grouped here, currently rtc (2.4)
210 execdomains Execdomains, related to security (2.4)
211 fb Frame Buffer devices (2.4)
212 fs File system parameters, currently nfs/exports (2.4)
213 ide Directory containing info about the IDE subsystem
214 interrupts Interrupt usage
215 iomem Memory map (2.4)
216 ioports I/O port usage
217 irq Masks for irq to cpu affinity (2.4)(smp?)
218 isapnp ISA PnP (Plug&Play) Info (2.4)
219 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
220 kmsg Kernel messages
221 ksyms Kernel symbol table
222 loadavg Load average of last 1, 5 & 15 minutes
223 locks Kernel locks
224 meminfo Memory info
225 misc Miscellaneous
226 modules List of loaded modules
227 mounts Mounted filesystems
228 net Networking info (see text)
229 partitions Table of partitions known to the system
230 pci Depreciated info of PCI bus (new way -> /proc/bus/pci/,
231 decoupled by lspci (2.4)
232 rtc Real time clock
233 scsi SCSI info (see text)
234 slabinfo Slab pool info
235 stat Overall statistics
236 swaps Swap space utilization
237 sys See chapter 2
238 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
239 tty Info of tty drivers
240 uptime System uptime
241 version Kernel version
242 video bttv info of video resources (2.4)
243..............................................................................
244
245You can, for example, check which interrupts are currently in use and what
246they are used for by looking in the file /proc/interrupts:
247
248 > cat /proc/interrupts
249 CPU0
250 0: 8728810 XT-PIC timer
251 1: 895 XT-PIC keyboard
252 2: 0 XT-PIC cascade
253 3: 531695 XT-PIC aha152x
254 4: 2014133 XT-PIC serial
255 5: 44401 XT-PIC pcnet_cs
256 8: 2 XT-PIC rtc
257 11: 8 XT-PIC i82365
258 12: 182918 XT-PIC PS/2 Mouse
259 13: 1 XT-PIC fpu
260 14: 1232265 XT-PIC ide0
261 15: 7 XT-PIC ide1
262 NMI: 0
263
264In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
265output of a SMP machine):
266
267 > cat /proc/interrupts
268
269 CPU0 CPU1
270 0: 1243498 1214548 IO-APIC-edge timer
271 1: 8949 8958 IO-APIC-edge keyboard
272 2: 0 0 XT-PIC cascade
273 5: 11286 10161 IO-APIC-edge soundblaster
274 8: 1 0 IO-APIC-edge rtc
275 9: 27422 27407 IO-APIC-edge 3c503
276 12: 113645 113873 IO-APIC-edge PS/2 Mouse
277 13: 0 0 XT-PIC fpu
278 14: 22491 24012 IO-APIC-edge ide0
279 15: 2183 2415 IO-APIC-edge ide1
280 17: 30564 30414 IO-APIC-level eth0
281 18: 177 164 IO-APIC-level bttv
282 NMI: 2457961 2457959
283 LOC: 2457882 2457881
284 ERR: 2155
285
286NMI is incremented in this case because every timer interrupt generates a NMI
287(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
288
289LOC is the local interrupt counter of the internal APIC of every CPU.
290
291ERR is incremented in the case of errors in the IO-APIC bus (the bus that
292connects the CPUs in a SMP system. This means that an error has been detected,
293the IO-APIC automatically retry the transmission, so it should not be a big
294problem, but you should read the SMP-FAQ.
295
296In this context it could be interesting to note the new irq directory in 2.4.
297It could be used to set IRQ to CPU affinity, this means that you can "hook" an
298IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
299irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
300
301For example
302 > ls /proc/irq/
303 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
304 1 11 13 15 17 19 3 5 7 9
305 > ls /proc/irq/0/
306 smp_affinity
307
308The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
309is the same by default:
310
311 > cat /proc/irq/0/smp_affinity
312 ffffffff
313
Uwe Zeisbergerc30fe7f2006-03-24 18:23:14 +0100314It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can
Linus Torvalds1da177e2005-04-16 15:20:36 -0700315set it by doing:
316
317 > echo 1 > /proc/irq/prof_cpu_mask
318
319This means that only the first CPU will handle the IRQ, but you can also echo 5
Uwe Zeisbergerc30fe7f2006-03-24 18:23:14 +0100320which means that only the first and fourth CPU can handle the IRQ.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700321
322The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
323between all the CPUs which are allowed to handle it. As usual the kernel has
324more info than you and does a better job than you, so the defaults are the
325best choice for almost everyone.
326
327There are three more important subdirectories in /proc: net, scsi, and sys.
328The general rule is that the contents, or even the existence of these
329directories, depend on your kernel configuration. If SCSI is not enabled, the
330directory scsi may not exist. The same is true with the net, which is there
331only when networking support is present in the running kernel.
332
333The slabinfo file gives information about memory usage at the slab level.
334Linux uses slab pools for memory management above page level in version 2.2.
335Commonly used objects have their own slab pool (such as network buffers,
336directory cache, and so on).
337
338..............................................................................
339
340> cat /proc/buddyinfo
341
342Node 0, zone DMA 0 4 5 4 4 3 ...
343Node 0, zone Normal 1 0 0 1 101 8 ...
344Node 0, zone HighMem 2 0 0 1 1 0 ...
345
346Memory fragmentation is a problem under some workloads, and buddyinfo is a
347useful tool for helping diagnose these problems. Buddyinfo will give you a
348clue as to how big an area you can safely allocate, or why a previous
349allocation failed.
350
351Each column represents the number of pages of a certain order which are
352available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
353ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
354available in ZONE_NORMAL, etc...
355
356..............................................................................
357
358meminfo:
359
360Provides information about distribution and utilization of memory. This
361varies by architecture and compile options. The following is from a
36216GB PIII, which has highmem enabled. You may not have all of these fields.
363
364> cat /proc/meminfo
365
366
367MemTotal: 16344972 kB
368MemFree: 13634064 kB
369Buffers: 3656 kB
370Cached: 1195708 kB
371SwapCached: 0 kB
372Active: 891636 kB
373Inactive: 1077224 kB
374HighTotal: 15597528 kB
375HighFree: 13629632 kB
376LowTotal: 747444 kB
377LowFree: 4432 kB
378SwapTotal: 0 kB
379SwapFree: 0 kB
380Dirty: 968 kB
381Writeback: 0 kB
382Mapped: 280372 kB
383Slab: 684068 kB
384CommitLimit: 7669796 kB
385Committed_AS: 100056 kB
386PageTables: 24448 kB
387VmallocTotal: 112216 kB
388VmallocUsed: 428 kB
389VmallocChunk: 111088 kB
390
391 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
392 bits and the kernel binary code)
393 MemFree: The sum of LowFree+HighFree
394 Buffers: Relatively temporary storage for raw disk blocks
395 shouldn't get tremendously large (20MB or so)
396 Cached: in-memory cache for files read from the disk (the
397 pagecache). Doesn't include SwapCached
398 SwapCached: Memory that once was swapped out, is swapped back in but
399 still also is in the swapfile (if memory is needed it
400 doesn't need to be swapped out AGAIN because it is already
401 in the swapfile. This saves I/O)
402 Active: Memory that has been used more recently and usually not
403 reclaimed unless absolutely necessary.
404 Inactive: Memory which has been less recently used. It is more
405 eligible to be reclaimed for other purposes
406 HighTotal:
407 HighFree: Highmem is all memory above ~860MB of physical memory
408 Highmem areas are for use by userspace programs, or
409 for the pagecache. The kernel must use tricks to access
410 this memory, making it slower to access than lowmem.
411 LowTotal:
412 LowFree: Lowmem is memory which can be used for everything that
Matt LaPlante3f6dee92006-10-03 22:45:33 +0200413 highmem can be used for, but it is also available for the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700414 kernel's use for its own data structures. Among many
415 other things, it is where everything from the Slab is
416 allocated. Bad things happen when you're out of lowmem.
417 SwapTotal: total amount of swap space available
418 SwapFree: Memory which has been evicted from RAM, and is temporarily
419 on the disk
420 Dirty: Memory which is waiting to get written back to the disk
421 Writeback: Memory which is actively being written back to the disk
422 Mapped: files which have been mmaped, such as libraries
Adrian Bunke82443c2006-01-10 00:20:30 +0100423 Slab: in-kernel data structures cache
Linus Torvalds1da177e2005-04-16 15:20:36 -0700424 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
425 this is the total amount of memory currently available to
426 be allocated on the system. This limit is only adhered to
427 if strict overcommit accounting is enabled (mode 2 in
428 'vm.overcommit_memory').
429 The CommitLimit is calculated with the following formula:
430 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
431 For example, on a system with 1G of physical RAM and 7G
432 of swap with a `vm.overcommit_ratio` of 30 it would
433 yield a CommitLimit of 7.3G.
434 For more details, see the memory overcommit documentation
435 in vm/overcommit-accounting.
436Committed_AS: The amount of memory presently allocated on the system.
437 The committed memory is a sum of all of the memory which
438 has been allocated by processes, even if it has not been
439 "used" by them as of yet. A process which malloc()'s 1G
440 of memory, but only touches 300M of it will only show up
441 as using 300M of memory even if it has the address space
442 allocated for the entire 1G. This 1G is memory which has
443 been "committed" to by the VM and can be used at any time
444 by the allocating application. With strict overcommit
445 enabled on the system (mode 2 in 'vm.overcommit_memory'),
446 allocations which would exceed the CommitLimit (detailed
447 above) will not be permitted. This is useful if one needs
448 to guarantee that processes will not fail due to lack of
449 memory once that memory has been successfully allocated.
450 PageTables: amount of memory dedicated to the lowest level of page
451 tables.
452VmallocTotal: total size of vmalloc memory area
453 VmallocUsed: amount of vmalloc area which is used
454VmallocChunk: largest contigious block of vmalloc area which is free
455
456
4571.3 IDE devices in /proc/ide
458----------------------------
459
460The subdirectory /proc/ide contains information about all IDE devices of which
461the kernel is aware. There is one subdirectory for each IDE controller, the
462file drivers and a link for each IDE device, pointing to the device directory
463in the controller specific subtree.
464
465The file drivers contains general information about the drivers used for the
466IDE devices:
467
468 > cat /proc/ide/drivers
469 ide-cdrom version 4.53
470 ide-disk version 1.08
471
472More detailed information can be found in the controller specific
473subdirectories. These are named ide0, ide1 and so on. Each of these
474directories contains the files shown in table 1-4.
475
476
477Table 1-4: IDE controller info in /proc/ide/ide?
478..............................................................................
479 File Content
480 channel IDE channel (0 or 1)
481 config Configuration (only for PCI/IDE bridge)
482 mate Mate name
483 model Type/Chipset of IDE controller
484..............................................................................
485
486Each device connected to a controller has a separate subdirectory in the
487controllers directory. The files listed in table 1-5 are contained in these
488directories.
489
490
491Table 1-5: IDE device information
492..............................................................................
493 File Content
494 cache The cache
495 capacity Capacity of the medium (in 512Byte blocks)
496 driver driver and version
497 geometry physical and logical geometry
498 identify device identify block
499 media media type
500 model device identifier
501 settings device setup
502 smart_thresholds IDE disk management thresholds
503 smart_values IDE disk management values
504..............................................................................
505
506The most interesting file is settings. This file contains a nice overview of
507the drive parameters:
508
509 # cat /proc/ide/ide0/hda/settings
510 name value min max mode
511 ---- ----- --- --- ----
512 bios_cyl 526 0 65535 rw
513 bios_head 255 0 255 rw
514 bios_sect 63 0 63 rw
515 breada_readahead 4 0 127 rw
516 bswap 0 0 1 r
517 file_readahead 72 0 2097151 rw
518 io_32bit 0 0 3 rw
519 keepsettings 0 0 1 rw
520 max_kb_per_request 122 1 127 rw
521 multcount 0 0 8 rw
522 nice1 1 0 1 rw
523 nowerr 0 0 1 rw
524 pio_mode write-only 0 255 w
525 slow 0 0 1 rw
526 unmaskirq 0 0 1 rw
527 using_dma 0 0 1 rw
528
529
5301.4 Networking info in /proc/net
531--------------------------------
532
533The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
534additional values you get for IP version 6 if you configure the kernel to
535support this. Table 1-7 lists the files and their meaning.
536
537
538Table 1-6: IPv6 info in /proc/net
539..............................................................................
540 File Content
541 udp6 UDP sockets (IPv6)
542 tcp6 TCP sockets (IPv6)
543 raw6 Raw device statistics (IPv6)
544 igmp6 IP multicast addresses, which this host joined (IPv6)
545 if_inet6 List of IPv6 interface addresses
546 ipv6_route Kernel routing table for IPv6
547 rt6_stats Global IPv6 routing tables statistics
548 sockstat6 Socket statistics (IPv6)
549 snmp6 Snmp data (IPv6)
550..............................................................................
551
552
553Table 1-7: Network info in /proc/net
554..............................................................................
555 File Content
556 arp Kernel ARP table
557 dev network devices with statistics
558 dev_mcast the Layer2 multicast groups a device is listening too
559 (interface index, label, number of references, number of bound
560 addresses).
561 dev_stat network device status
562 ip_fwchains Firewall chain linkage
563 ip_fwnames Firewall chain names
564 ip_masq Directory containing the masquerading tables
565 ip_masquerade Major masquerading table
566 netstat Network statistics
567 raw raw device statistics
568 route Kernel routing table
569 rpc Directory containing rpc info
570 rt_cache Routing cache
571 snmp SNMP data
572 sockstat Socket statistics
573 tcp TCP sockets
574 tr_rif Token ring RIF routing table
575 udp UDP sockets
576 unix UNIX domain sockets
577 wireless Wireless interface data (Wavelan etc)
578 igmp IP multicast addresses, which this host joined
579 psched Global packet scheduler parameters.
580 netlink List of PF_NETLINK sockets
581 ip_mr_vifs List of multicast virtual interfaces
582 ip_mr_cache List of multicast routing cache
583..............................................................................
584
585You can use this information to see which network devices are available in
586your system and how much traffic was routed over those devices:
587
588 > cat /proc/net/dev
589 Inter-|Receive |[...
590 face |bytes packets errs drop fifo frame compressed multicast|[...
591 lo: 908188 5596 0 0 0 0 0 0 [...
592 ppp0:15475140 20721 410 0 0 410 0 0 [...
593 eth0: 614530 7085 0 0 0 0 0 1 [...
594
595 ...] Transmit
596 ...] bytes packets errs drop fifo colls carrier compressed
597 ...] 908188 5596 0 0 0 0 0 0
598 ...] 1375103 17405 0 0 0 0 0 0
599 ...] 1703981 5535 0 0 0 3 0 0
600
601In addition, each Channel Bond interface has it's own directory. For
602example, the bond0 device will have a directory called /proc/net/bond0/.
603It will contain information that is specific to that bond, such as the
604current slaves of the bond, the link status of the slaves, and how
605many times the slaves link has failed.
606
6071.5 SCSI info
608-------------
609
610If you have a SCSI host adapter in your system, you'll find a subdirectory
611named after the driver for this adapter in /proc/scsi. You'll also see a list
612of all recognized SCSI devices in /proc/scsi:
613
614 >cat /proc/scsi/scsi
615 Attached devices:
616 Host: scsi0 Channel: 00 Id: 00 Lun: 00
617 Vendor: IBM Model: DGHS09U Rev: 03E0
618 Type: Direct-Access ANSI SCSI revision: 03
619 Host: scsi0 Channel: 00 Id: 06 Lun: 00
620 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
621 Type: CD-ROM ANSI SCSI revision: 02
622
623
624The directory named after the driver has one file for each adapter found in
625the system. These files contain information about the controller, including
626the used IRQ and the IO address range. The amount of information shown is
627dependent on the adapter you use. The example shows the output for an Adaptec
628AHA-2940 SCSI adapter:
629
630 > cat /proc/scsi/aic7xxx/0
631
632 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
633 Compile Options:
634 TCQ Enabled By Default : Disabled
635 AIC7XXX_PROC_STATS : Disabled
636 AIC7XXX_RESET_DELAY : 5
637 Adapter Configuration:
638 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
639 Ultra Wide Controller
640 PCI MMAPed I/O Base: 0xeb001000
641 Adapter SEEPROM Config: SEEPROM found and used.
642 Adaptec SCSI BIOS: Enabled
643 IRQ: 10
644 SCBs: Active 0, Max Active 2,
645 Allocated 15, HW 16, Page 255
646 Interrupts: 160328
647 BIOS Control Word: 0x18b6
648 Adapter Control Word: 0x005b
649 Extended Translation: Enabled
650 Disconnect Enable Flags: 0xffff
651 Ultra Enable Flags: 0x0001
652 Tag Queue Enable Flags: 0x0000
653 Ordered Queue Tag Flags: 0x0000
654 Default Tag Queue Depth: 8
655 Tagged Queue By Device array for aic7xxx host instance 0:
656 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
657 Actual queue depth per device for aic7xxx host instance 0:
658 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
659 Statistics:
660 (scsi0:0:0:0)
661 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
662 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
663 Total transfers 160151 (74577 reads and 85574 writes)
664 (scsi0:0:6:0)
665 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
666 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
667 Total transfers 0 (0 reads and 0 writes)
668
669
6701.6 Parallel port info in /proc/parport
671---------------------------------------
672
673The directory /proc/parport contains information about the parallel ports of
674your system. It has one subdirectory for each port, named after the port
675number (0,1,2,...).
676
677These directories contain the four files shown in Table 1-8.
678
679
680Table 1-8: Files in /proc/parport
681..............................................................................
682 File Content
683 autoprobe Any IEEE-1284 device ID information that has been acquired.
684 devices list of the device drivers using that port. A + will appear by the
685 name of the device currently using the port (it might not appear
686 against any).
687 hardware Parallel port's base address, IRQ line and DMA channel.
688 irq IRQ that parport is using for that port. This is in a separate
689 file to allow you to alter it by writing a new value in (IRQ
690 number or none).
691..............................................................................
692
6931.7 TTY info in /proc/tty
694-------------------------
695
696Information about the available and actually used tty's can be found in the
697directory /proc/tty.You'll find entries for drivers and line disciplines in
698this directory, as shown in Table 1-9.
699
700
701Table 1-9: Files in /proc/tty
702..............................................................................
703 File Content
704 drivers list of drivers and their usage
705 ldiscs registered line disciplines
706 driver/serial usage statistic and status of single tty lines
707..............................................................................
708
709To see which tty's are currently in use, you can simply look into the file
710/proc/tty/drivers:
711
712 > cat /proc/tty/drivers
713 pty_slave /dev/pts 136 0-255 pty:slave
714 pty_master /dev/ptm 128 0-255 pty:master
715 pty_slave /dev/ttyp 3 0-255 pty:slave
716 pty_master /dev/pty 2 0-255 pty:master
717 serial /dev/cua 5 64-67 serial:callout
718 serial /dev/ttyS 4 64-67 serial
719 /dev/tty0 /dev/tty0 4 0 system:vtmaster
720 /dev/ptmx /dev/ptmx 5 2 system
721 /dev/console /dev/console 5 1 system:console
722 /dev/tty /dev/tty 5 0 system:/dev/tty
723 unknown /dev/tty 4 1-63 console
724
725
7261.8 Miscellaneous kernel statistics in /proc/stat
727-------------------------------------------------
728
729Various pieces of information about kernel activity are available in the
730/proc/stat file. All of the numbers reported in this file are aggregates
731since the system first booted. For a quick look, simply cat the file:
732
733 > cat /proc/stat
734 cpu 2255 34 2290 22625563 6290 127 456
735 cpu0 1132 34 1441 11311718 3675 127 438
736 cpu1 1123 0 849 11313845 2614 0 18
737 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
738 ctxt 1990473
739 btime 1062191376
740 processes 2915
741 procs_running 1
742 procs_blocked 0
743
744The very first "cpu" line aggregates the numbers in all of the other "cpuN"
745lines. These numbers identify the amount of time the CPU has spent performing
746different kinds of work. Time units are in USER_HZ (typically hundredths of a
747second). The meanings of the columns are as follows, from left to right:
748
749- user: normal processes executing in user mode
750- nice: niced processes executing in user mode
751- system: processes executing in kernel mode
752- idle: twiddling thumbs
753- iowait: waiting for I/O to complete
754- irq: servicing interrupts
755- softirq: servicing softirqs
756
757The "intr" line gives counts of interrupts serviced since boot time, for each
758of the possible system interrupts. The first column is the total of all
759interrupts serviced; each subsequent column is the total for that particular
760interrupt.
761
762The "ctxt" line gives the total number of context switches across all CPUs.
763
764The "btime" line gives the time at which the system booted, in seconds since
765the Unix epoch.
766
767The "processes" line gives the number of processes and threads created, which
768includes (but is not limited to) those created by calls to the fork() and
769clone() system calls.
770
771The "procs_running" line gives the number of processes currently running on
772CPUs.
773
774The "procs_blocked" line gives the number of processes currently blocked,
775waiting for I/O to complete.
776
777
778------------------------------------------------------------------------------
779Summary
780------------------------------------------------------------------------------
781The /proc file system serves information about the running system. It not only
782allows access to process data but also allows you to request the kernel status
783by reading files in the hierarchy.
784
785The directory structure of /proc reflects the types of information and makes
786it easy, if not obvious, where to look for specific data.
787------------------------------------------------------------------------------
788
789------------------------------------------------------------------------------
790CHAPTER 2: MODIFYING SYSTEM PARAMETERS
791------------------------------------------------------------------------------
792
793------------------------------------------------------------------------------
794In This Chapter
795------------------------------------------------------------------------------
796* Modifying kernel parameters by writing into files found in /proc/sys
797* Exploring the files which modify certain parameters
798* Review of the /proc/sys file tree
799------------------------------------------------------------------------------
800
801
802A very interesting part of /proc is the directory /proc/sys. This is not only
803a source of information, it also allows you to change parameters within the
804kernel. Be very careful when attempting this. You can optimize your system,
805but you can also cause it to crash. Never alter kernel parameters on a
806production system. Set up a development machine and test to make sure that
807everything works the way you want it to. You may have no alternative but to
808reboot the machine once an error has been made.
809
810To change a value, simply echo the new value into the file. An example is
811given below in the section on the file system data. You need to be root to do
812this. You can create your own boot script to perform this every time your
813system boots.
814
815The files in /proc/sys can be used to fine tune and monitor miscellaneous and
816general things in the operation of the Linux kernel. Since some of the files
817can inadvertently disrupt your system, it is advisable to read both
818documentation and source before actually making adjustments. In any case, be
819very careful when writing to any of these files. The entries in /proc may
820change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
821review the kernel documentation in the directory /usr/src/linux/Documentation.
822This chapter is heavily based on the documentation included in the pre 2.2
823kernels, and became part of it in version 2.2.1 of the Linux kernel.
824
8252.1 /proc/sys/fs - File system data
826-----------------------------------
827
828This subdirectory contains specific file system, file handle, inode, dentry
829and quota information.
830
831Currently, these files are in /proc/sys/fs:
832
833dentry-state
834------------
835
836Status of the directory cache. Since directory entries are dynamically
837allocated and deallocated, this file indicates the current status. It holds
838six values, in which the last two are not used and are always zero. The others
839are listed in table 2-1.
840
841
842Table 2-1: Status files of the directory cache
843..............................................................................
844 File Content
845 nr_dentry Almost always zero
846 nr_unused Number of unused cache entries
847 age_limit
848 in seconds after the entry may be reclaimed, when memory is short
849 want_pages internally
850..............................................................................
851
852dquot-nr and dquot-max
853----------------------
854
855The file dquot-max shows the maximum number of cached disk quota entries.
856
857The file dquot-nr shows the number of allocated disk quota entries and the
858number of free disk quota entries.
859
860If the number of available cached disk quotas is very low and you have a large
861number of simultaneous system users, you might want to raise the limit.
862
863file-nr and file-max
864--------------------
865
866The kernel allocates file handles dynamically, but doesn't free them again at
867this time.
868
869The value in file-max denotes the maximum number of file handles that the
870Linux kernel will allocate. When you get a lot of error messages about running
871out of file handles, you might want to raise this limit. The default value is
87210% of RAM in kilobytes. To change it, just write the new number into the
873file:
874
875 # cat /proc/sys/fs/file-max
876 4096
877 # echo 8192 > /proc/sys/fs/file-max
878 # cat /proc/sys/fs/file-max
879 8192
880
881
882This method of revision is useful for all customizable parameters of the
883kernel - simply echo the new value to the corresponding file.
884
885Historically, the three values in file-nr denoted the number of allocated file
886handles, the number of allocated but unused file handles, and the maximum
887number of file handles. Linux 2.6 always reports 0 as the number of free file
888handles -- this is not an error, it just means that the number of allocated
889file handles exactly matches the number of used file handles.
890
891Attempts to allocate more file descriptors than file-max are reported with
892printk, look for "VFS: file-max limit <number> reached".
893
894inode-state and inode-nr
895------------------------
896
897The file inode-nr contains the first two items from inode-state, so we'll skip
898to that file...
899
900inode-state contains two actual numbers and five dummy values. The numbers
901are nr_inodes and nr_free_inodes (in order of appearance).
902
903nr_inodes
904~~~~~~~~~
905
906Denotes the number of inodes the system has allocated. This number will
907grow and shrink dynamically.
908
909nr_free_inodes
910--------------
911
912Represents the number of free inodes. Ie. The number of inuse inodes is
913(nr_inodes - nr_free_inodes).
914
Linus Torvalds1da177e2005-04-16 15:20:36 -0700915aio-nr and aio-max-nr
916---------------------
917
918aio-nr is the running total of the number of events specified on the
919io_setup system call for all currently active aio contexts. If aio-nr
920reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
921raising aio-max-nr does not result in the pre-allocation or re-sizing
922of any kernel data structures.
923
9242.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
925-----------------------------------------------------------
926
927Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
928handles the kernel support for miscellaneous binary formats.
929
930Binfmt_misc provides the ability to register additional binary formats to the
931Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
932needs to know magic numbers at the beginning or the filename extension of the
933binary.
934
935It works by maintaining a linked list of structs that contain a description of
936a binary format, including a magic with size (or the filename extension),
937offset and mask, and the interpreter name. On request it invokes the given
938interpreter with the original program as argument, as binfmt_java and
939binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
940binary-formats, you have to register an additional binary-format.
941
942There are two general files in binfmt_misc and one file per registered format.
943The two general files are register and status.
944
945Registering a new binary format
946-------------------------------
947
948To register a new binary format you have to issue the command
949
950 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
951
952
953
954with appropriate name (the name for the /proc-dir entry), offset (defaults to
9550, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
956last but not least, the interpreter that is to be invoked (for example and
957testing /bin/echo). Type can be M for usual magic matching or E for filename
958extension matching (give extension in place of magic).
959
960Check or reset the status of the binary format handler
961------------------------------------------------------
962
963If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
964current status (enabled/disabled) of binfmt_misc. Change the status by echoing
9650 (disables) or 1 (enables) or -1 (caution: this clears all previously
966registered binary formats) to status. For example echo 0 > status to disable
967binfmt_misc (temporarily).
968
969Status of a single handler
970--------------------------
971
972Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
973perform the same function as status, but their scope is limited to the actual
974binary format. By cating this file, you also receive all related information
975about the interpreter/magic of the binfmt.
976
977Example usage of binfmt_misc (emulate binfmt_java)
978--------------------------------------------------
979
980 cd /proc/sys/fs/binfmt_misc
981 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
982 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
983 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
984 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
985
986
987These four lines add support for Java executables and Java applets (like
988binfmt_java, additionally recognizing the .html extension with no need to put
989<!--applet> to every applet file). You have to install the JDK and the
990shell-script /usr/local/java/bin/javawrapper too. It works around the
991brokenness of the Java filename handling. To add a Java binary, just create a
992link to the class-file somewhere in the path.
993
9942.3 /proc/sys/kernel - general kernel parameters
995------------------------------------------------
996
997This directory reflects general kernel behaviors. As I've said before, the
998contents depend on your configuration. Here you'll find the most important
999files, along with descriptions of what they mean and how to use them.
1000
1001acct
1002----
1003
1004The file contains three values; highwater, lowwater, and frequency.
1005
1006It exists only when BSD-style process accounting is enabled. These values
1007control its behavior. If the free space on the file system where the log lives
1008goes below lowwater percentage, accounting suspends. If it goes above
1009highwater percentage, accounting resumes. Frequency determines how often you
1010check the amount of free space (value is in seconds). Default settings are: 4,
10112, and 30. That is, suspend accounting if there is less than 2 percent free;
1012resume it if we have a value of 3 or more percent; consider information about
1013the amount of free space valid for 30 seconds
1014
1015ctrl-alt-del
1016------------
1017
1018When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1019program to handle a graceful restart. However, when the value is greater that
1020zero, Linux's reaction to this key combination will be an immediate reboot,
1021without syncing its dirty buffers.
1022
1023[NOTE]
1024 When a program (like dosemu) has the keyboard in raw mode, the
1025 ctrl-alt-del is intercepted by the program before it ever reaches the
1026 kernel tty layer, and it is up to the program to decide what to do with
1027 it.
1028
1029domainname and hostname
1030-----------------------
1031
1032These files can be controlled to set the NIS domainname and hostname of your
1033box. For the classic darkstar.frop.org a simple:
1034
1035 # echo "darkstar" > /proc/sys/kernel/hostname
1036 # echo "frop.org" > /proc/sys/kernel/domainname
1037
1038
1039would suffice to set your hostname and NIS domainname.
1040
1041osrelease, ostype and version
1042-----------------------------
1043
1044The names make it pretty obvious what these fields contain:
1045
1046 > cat /proc/sys/kernel/osrelease
1047 2.2.12
1048
1049 > cat /proc/sys/kernel/ostype
1050 Linux
1051
1052 > cat /proc/sys/kernel/version
1053 #4 Fri Oct 1 12:41:14 PDT 1999
1054
1055
1056The files osrelease and ostype should be clear enough. Version needs a little
1057more clarification. The #4 means that this is the 4th kernel built from this
1058source base and the date after it indicates the time the kernel was built. The
1059only way to tune these values is to rebuild the kernel.
1060
1061panic
1062-----
1063
1064The value in this file represents the number of seconds the kernel waits
1065before rebooting on a panic. When you use the software watchdog, the
1066recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1067is disabled, which is the default setting.
1068
1069printk
1070------
1071
1072The four values in printk denote
1073* console_loglevel,
1074* default_message_loglevel,
1075* minimum_console_loglevel and
1076* default_console_loglevel
1077respectively.
1078
1079These values influence printk() behavior when printing or logging error
1080messages, which come from inside the kernel. See syslog(2) for more
1081information on the different log levels.
1082
1083console_loglevel
1084----------------
1085
1086Messages with a higher priority than this will be printed to the console.
1087
1088default_message_level
1089---------------------
1090
1091Messages without an explicit priority will be printed with this priority.
1092
1093minimum_console_loglevel
1094------------------------
1095
1096Minimum (highest) value to which the console_loglevel can be set.
1097
1098default_console_loglevel
1099------------------------
1100
1101Default value for console_loglevel.
1102
1103sg-big-buff
1104-----------
1105
1106This file shows the size of the generic SCSI (sg) buffer. At this point, you
1107can't tune it yet, but you can change it at compile time by editing
1108include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1109
1110If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1111this to a higher value. Refer to the SANE documentation on this issue.
1112
1113modprobe
1114--------
1115
1116The location where the modprobe binary is located. The kernel uses this
1117program to load modules on demand.
1118
1119unknown_nmi_panic
1120-----------------
1121
1122The value in this file affects behavior of handling NMI. When the value is
1123non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1124debugging information is displayed on console.
1125
1126NMI switch that most IA32 servers have fires unknown NMI up, for example.
1127If a system hangs up, try pressing the NMI switch.
1128
Don Zickuse33e89a2006-09-26 10:52:27 +02001129nmi_watchdog
1130------------
1131
1132Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
1133the NMI watchdog is enabled and will continuously test all online cpus to
1134determine whether or not they are still functioning properly.
1135
1136Because the NMI watchdog shares registers with oprofile, by disabling the NMI
1137watchdog, oprofile may have more registers to utilize.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001138
1139
11402.4 /proc/sys/vm - The virtual memory subsystem
1141-----------------------------------------------
1142
1143The files in this directory can be used to tune the operation of the virtual
1144memory (VM) subsystem of the Linux kernel.
1145
1146vfs_cache_pressure
1147------------------
1148
1149Controls the tendency of the kernel to reclaim the memory which is used for
1150caching of directory and inode objects.
1151
1152At the default value of vfs_cache_pressure=100 the kernel will attempt to
1153reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1154swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1155to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1156causes the kernel to prefer to reclaim dentries and inodes.
1157
1158dirty_background_ratio
1159----------------------
1160
1161Contains, as a percentage of total system memory, the number of pages at which
1162the pdflush background writeback daemon will start writing out dirty data.
1163
1164dirty_ratio
1165-----------------
1166
1167Contains, as a percentage of total system memory, the number of pages at which
1168a process which is generating disk writes will itself start writing out dirty
1169data.
1170
1171dirty_writeback_centisecs
1172-------------------------
1173
1174The pdflush writeback daemons will periodically wake up and write `old' data
1175out to disk. This tunable expresses the interval between those wakeups, in
1176100'ths of a second.
1177
1178Setting this to zero disables periodic writeback altogether.
1179
1180dirty_expire_centisecs
1181----------------------
1182
1183This tunable is used to define when dirty data is old enough to be eligible
1184for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1185Data which has been dirty in-memory for longer than this interval will be
1186written out next time a pdflush daemon wakes up.
1187
1188legacy_va_layout
1189----------------
1190
1191If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1192will use the legacy (2.4) layout for all processes.
1193
1194lower_zone_protection
1195---------------------
1196
1197For some specialised workloads on highmem machines it is dangerous for
1198the kernel to allow process memory to be allocated from the "lowmem"
1199zone. This is because that memory could then be pinned via the mlock()
1200system call, or by unavailability of swapspace.
1201
1202And on large highmem machines this lack of reclaimable lowmem memory
1203can be fatal.
1204
1205So the Linux page allocator has a mechanism which prevents allocations
1206which _could_ use highmem from using too much lowmem. This means that
1207a certain amount of lowmem is defended from the possibility of being
1208captured into pinned user memory.
1209
1210(The same argument applies to the old 16 megabyte ISA DMA region. This
1211mechanism will also defend that region from allocations which could use
1212highmem or lowmem).
1213
1214The `lower_zone_protection' tunable determines how aggressive the kernel is
1215in defending these lower zones. The default value is zero - no
1216protection at all.
1217
1218If you have a machine which uses highmem or ISA DMA and your
1219applications are using mlock(), or if you are running with no swap then
1220you probably should increase the lower_zone_protection setting.
1221
1222The units of this tunable are fairly vague. It is approximately equal
1223to "megabytes". So setting lower_zone_protection=100 will protect around 100
1224megabytes of the lowmem zone from user allocations. It will also make
1225those 100 megabytes unavaliable for use by applications and by
1226pagecache, so there is a cost.
1227
1228The effects of this tunable may be observed by monitoring
1229/proc/meminfo:LowFree. Write a single huge file and observe the point
1230at which LowFree ceases to fall.
1231
1232A reasonable value for lower_zone_protection is 100.
1233
1234page-cluster
1235------------
1236
1237page-cluster controls the number of pages which are written to swap in
1238a single attempt. The swap I/O size.
1239
1240It is a logarithmic value - setting it to zero means "1 page", setting
1241it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1242
1243The default value is three (eight pages at a time). There may be some
1244small benefits in tuning this to a different value if your workload is
1245swap-intensive.
1246
1247overcommit_memory
1248-----------------
1249
Chuck Ebbertaf97c722005-09-09 13:10:15 -07001250Controls overcommit of system memory, possibly allowing processes
1251to allocate (but not use) more memory than is actually available.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001252
Chuck Ebbertaf97c722005-09-09 13:10:15 -07001253
12540 - Heuristic overcommit handling. Obvious overcommits of
1255 address space are refused. Used for a typical system. It
1256 ensures a seriously wild allocation fails while allowing
1257 overcommit to reduce swap usage. root is allowed to
1258 allocate slighly more memory in this mode. This is the
1259 default.
1260
12611 - Always overcommit. Appropriate for some scientific
1262 applications.
1263
12642 - Don't overcommit. The total address space commit
1265 for the system is not permitted to exceed swap plus a
1266 configurable percentage (default is 50) of physical RAM.
1267 Depending on the percentage you use, in most situations
1268 this means a process will not be killed while attempting
1269 to use already-allocated memory but will receive errors
1270 on memory allocation as appropriate.
1271
1272overcommit_ratio
1273----------------
1274
1275Percentage of physical memory size to include in overcommit calculations
1276(see above.)
1277
1278Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
1279
1280 swapspace = total size of all swap areas
1281 physmem = size of physical memory in system
Linus Torvalds1da177e2005-04-16 15:20:36 -07001282
1283nr_hugepages and hugetlb_shm_group
1284----------------------------------
1285
1286nr_hugepages configures number of hugetlb page reserved for the system.
1287
1288hugetlb_shm_group contains group id that is allowed to create SysV shared
1289memory segment using hugetlb page.
1290
1291laptop_mode
1292-----------
1293
1294laptop_mode is a knob that controls "laptop mode". All the things that are
1295controlled by this knob are discussed in Documentation/laptop-mode.txt.
1296
1297block_dump
1298----------
1299
1300block_dump enables block I/O debugging when set to a nonzero value. More
1301information on block I/O debugging is in Documentation/laptop-mode.txt.
1302
1303swap_token_timeout
1304------------------
1305
1306This file contains valid hold time of swap out protection token. The Linux
1307VM has token based thrashing control mechanism and uses the token to prevent
1308unnecessary page faults in thrashing situation. The unit of the value is
1309second. The value would be useful to tune thrashing behavior.
1310
Andrew Morton9d0243b2006-01-08 01:00:39 -08001311drop_caches
1312-----------
1313
1314Writing to this will cause the kernel to drop clean caches, dentries and
1315inodes from memory, causing that memory to become free.
1316
1317To free pagecache:
1318 echo 1 > /proc/sys/vm/drop_caches
1319To free dentries and inodes:
1320 echo 2 > /proc/sys/vm/drop_caches
1321To free pagecache, dentries and inodes:
1322 echo 3 > /proc/sys/vm/drop_caches
1323
1324As this is a non-destructive operation and dirty objects are not freeable, the
1325user should run `sync' first.
1326
1327
Linus Torvalds1da177e2005-04-16 15:20:36 -070013282.5 /proc/sys/dev - Device specific parameters
1329----------------------------------------------
1330
1331Currently there is only support for CDROM drives, and for those, there is only
1332one read-only file containing information about the CD-ROM drives attached to
1333the system:
1334
1335 >cat /proc/sys/dev/cdrom/info
1336 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1337
1338 drive name: sr0 hdb
1339 drive speed: 32 40
1340 drive # of slots: 1 0
1341 Can close tray: 1 1
1342 Can open tray: 1 1
1343 Can lock tray: 1 1
1344 Can change speed: 1 1
1345 Can select disk: 0 1
1346 Can read multisession: 1 1
1347 Can read MCN: 1 1
1348 Reports media changed: 1 1
1349 Can play audio: 1 1
1350
1351
1352You see two drives, sr0 and hdb, along with a list of their features.
1353
13542.6 /proc/sys/sunrpc - Remote procedure calls
1355---------------------------------------------
1356
1357This directory contains four files, which enable or disable debugging for the
1358RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1359be set to one to turn debugging on. (The default value is 0 for each)
1360
13612.7 /proc/sys/net - Networking stuff
1362------------------------------------
1363
1364The interface to the networking parts of the kernel is located in
1365/proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1366some of them, depending on your kernel's configuration.
1367
1368
1369Table 2-3: Subdirectories in /proc/sys/net
1370..............................................................................
1371 Directory Content Directory Content
1372 core General parameter appletalk Appletalk protocol
1373 unix Unix domain sockets netrom NET/ROM
1374 802 E802 protocol ax25 AX25
1375 ethernet Ethernet protocol rose X.25 PLP layer
1376 ipv4 IP version 4 x25 X.25 protocol
1377 ipx IPX token-ring IBM token ring
1378 bridge Bridging decnet DEC net
1379 ipv6 IP version 6
1380..............................................................................
1381
1382We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1383only minor players in the Linux world, we'll skip them in this chapter. You'll
1384find some short info on Appletalk and IPX further on in this chapter. Review
1385the online documentation and the kernel source to get a detailed view of the
1386parameters for those protocols. In this section we'll discuss the
1387subdirectories printed in bold letters in the table above. As default values
1388are suitable for most needs, there is no need to change these values.
1389
1390/proc/sys/net/core - Network core options
1391-----------------------------------------
1392
1393rmem_default
1394------------
1395
1396The default setting of the socket receive buffer in bytes.
1397
1398rmem_max
1399--------
1400
1401The maximum receive socket buffer size in bytes.
1402
1403wmem_default
1404------------
1405
1406The default setting (in bytes) of the socket send buffer.
1407
1408wmem_max
1409--------
1410
1411The maximum send socket buffer size in bytes.
1412
1413message_burst and message_cost
1414------------------------------
1415
1416These parameters are used to limit the warning messages written to the kernel
1417log from the networking code. They enforce a rate limit to make a
1418denial-of-service attack impossible. A higher message_cost factor, results in
1419fewer messages that will be written. Message_burst controls when messages will
1420be dropped. The default settings limit warning messages to one every five
1421seconds.
1422
1423netdev_max_backlog
1424------------------
1425
1426Maximum number of packets, queued on the INPUT side, when the interface
1427receives packets faster than kernel can process them.
1428
1429optmem_max
1430----------
1431
1432Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1433of struct cmsghdr structures with appended data.
1434
1435/proc/sys/net/unix - Parameters for Unix domain sockets
1436-------------------------------------------------------
1437
1438There are only two files in this subdirectory. They control the delays for
1439deleting and destroying socket descriptors.
1440
14412.8 /proc/sys/net/ipv4 - IPV4 settings
1442--------------------------------------
1443
1444IP version 4 is still the most used protocol in Unix networking. It will be
1445replaced by IP version 6 in the next couple of years, but for the moment it's
1446the de facto standard for the internet and is used in most networking
1447environments around the world. Because of the importance of this protocol,
1448we'll have a deeper look into the subtree controlling the behavior of the IPv4
1449subsystem of the Linux kernel.
1450
1451Let's start with the entries in /proc/sys/net/ipv4.
1452
1453ICMP settings
1454-------------
1455
1456icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1457----------------------------------------------------
1458
1459Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1460just those to broadcast and multicast addresses.
1461
1462Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1463destination address your network may be used as an exploder for denial of
1464service packet flooding attacks to other hosts.
1465
1466icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1467---------------------------------------------------------------------------------------
1468
1469Sets limits for sending ICMP packets to specific targets. A value of zero
1470disables all limiting. Any positive value sets the maximum package rate in
1471hundredth of a second (on Intel systems).
1472
1473IP settings
1474-----------
1475
1476ip_autoconfig
1477-------------
1478
1479This file contains the number one if the host received its IP configuration by
1480RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1481
1482ip_default_ttl
1483--------------
1484
1485TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1486hops a packet may travel.
1487
1488ip_dynaddr
1489----------
1490
1491Enable dynamic socket address rewriting on interface address change. This is
1492useful for dialup interface with changing IP addresses.
1493
1494ip_forward
1495----------
1496
1497Enable or disable forwarding of IP packages between interfaces. Changing this
1498value resets all other parameters to their default values. They differ if the
1499kernel is configured as host or router.
1500
1501ip_local_port_range
1502-------------------
1503
1504Range of ports used by TCP and UDP to choose the local port. Contains two
1505numbers, the first number is the lowest port, the second number the highest
1506local port. Default is 1024-4999. Should be changed to 32768-61000 for
1507high-usage systems.
1508
1509ip_no_pmtu_disc
1510---------------
1511
1512Global switch to turn path MTU discovery off. It can also be set on a per
1513socket basis by the applications or on a per route basis.
1514
1515ip_masq_debug
1516-------------
1517
1518Enable/disable debugging of IP masquerading.
1519
1520IP fragmentation settings
1521-------------------------
1522
1523ipfrag_high_trash and ipfrag_low_trash
1524--------------------------------------
1525
1526Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1527of memory is allocated for this purpose, the fragment handler will toss
1528packets until ipfrag_low_thresh is reached.
1529
1530ipfrag_time
1531-----------
1532
1533Time in seconds to keep an IP fragment in memory.
1534
1535TCP settings
1536------------
1537
1538tcp_ecn
1539-------
1540
1541This file controls the use of the ECN bit in the IPv4 headers, this is a new
1542feature about Explicit Congestion Notification, but some routers and firewalls
1543block trafic that has this bit set, so it could be necessary to echo 0 to
1544/proc/sys/net/ipv4/tcp_ecn, if you want to talk to this sites. For more info
1545you could read RFC2481.
1546
1547tcp_retrans_collapse
1548--------------------
1549
1550Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1551larger packets to work around bugs in certain TCP stacks. Can be turned off by
1552setting it to zero.
1553
1554tcp_keepalive_probes
1555--------------------
1556
1557Number of keep alive probes TCP sends out, until it decides that the
1558connection is broken.
1559
1560tcp_keepalive_time
1561------------------
1562
1563How often TCP sends out keep alive messages, when keep alive is enabled. The
1564default is 2 hours.
1565
1566tcp_syn_retries
1567---------------
1568
1569Number of times initial SYNs for a TCP connection attempt will be
1570retransmitted. Should not be higher than 255. This is only the timeout for
1571outgoing connections, for incoming connections the number of retransmits is
1572defined by tcp_retries1.
1573
1574tcp_sack
1575--------
1576
1577Enable select acknowledgments after RFC2018.
1578
1579tcp_timestamps
1580--------------
1581
1582Enable timestamps as defined in RFC1323.
1583
1584tcp_stdurg
1585----------
1586
1587Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1588default is to use the BSD compatible interpretation of the urgent pointer
1589pointing to the first byte after the urgent data. The RFC793 interpretation is
1590to have it point to the last byte of urgent data. Enabling this option may
Matt LaPlante2fe0ae72006-10-03 22:50:39 +02001591lead to interoperability problems. Disabled by default.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001592
1593tcp_syncookies
1594--------------
1595
1596Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1597syncookies when the syn backlog queue of a socket overflows. This is to ward
1598off the common 'syn flood attack'. Disabled by default.
1599
1600Note that the concept of a socket backlog is abandoned. This means the peer
1601may not receive reliable error messages from an over loaded server with
1602syncookies enabled.
1603
1604tcp_window_scaling
1605------------------
1606
1607Enable window scaling as defined in RFC1323.
1608
1609tcp_fin_timeout
1610---------------
1611
1612The length of time in seconds it takes to receive a final FIN before the
1613socket is always closed. This is strictly a violation of the TCP
1614specification, but required to prevent denial-of-service attacks.
1615
1616tcp_max_ka_probes
1617-----------------
1618
1619Indicates how many keep alive probes are sent per slow timer run. Should not
1620be set too high to prevent bursts.
1621
1622tcp_max_syn_backlog
1623-------------------
1624
1625Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1626in listen(2) only specifies the length of the backlog queue of already
1627established sockets. When more connection requests arrive Linux starts to drop
1628packets. When syncookies are enabled the packets are still answered and the
1629maximum queue is effectively ignored.
1630
1631tcp_retries1
1632------------
1633
1634Defines how often an answer to a TCP connection request is retransmitted
1635before giving up.
1636
1637tcp_retries2
1638------------
1639
1640Defines how often a TCP packet is retransmitted before giving up.
1641
1642Interface specific settings
1643---------------------------
1644
1645In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1646interface the system knows about and one directory calls all. Changes in the
1647all subdirectory affect all interfaces, whereas changes in the other
1648subdirectories affect only one interface. All directories have the same
1649entries:
1650
1651accept_redirects
1652----------------
1653
1654This switch decides if the kernel accepts ICMP redirect messages or not. The
1655default is 'yes' if the kernel is configured for a regular host and 'no' for a
1656router configuration.
1657
1658accept_source_route
1659-------------------
1660
1661Should source routed packages be accepted or declined. The default is
1662dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1663hosts.
1664
1665bootp_relay
1666~~~~~~~~~~~
1667
1668Accept packets with source address 0.b.c.d with destinations not to this host
1669as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1670such packets.
1671
1672The default is 0, since this feature is not implemented yet (kernel version
16732.2.12).
1674
1675forwarding
1676----------
1677
1678Enable or disable IP forwarding on this interface.
1679
1680log_martians
1681------------
1682
1683Log packets with source addresses with no known route to kernel log.
1684
1685mc_forwarding
1686-------------
1687
1688Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1689multicast routing daemon is required.
1690
1691proxy_arp
1692---------
1693
1694Does (1) or does not (0) perform proxy ARP.
1695
1696rp_filter
1697---------
1698
1699Integer value determines if a source validation should be made. 1 means yes, 0
1700means no. Disabled by default, but local/broadcast address spoofing is always
1701on.
1702
1703If you set this to 1 on a router that is the only connection for a network to
1704the net, it will prevent spoofing attacks against your internal networks
1705(external addresses can still be spoofed), without the need for additional
1706firewall rules.
1707
1708secure_redirects
1709----------------
1710
1711Accept ICMP redirect messages only for gateways, listed in default gateway
1712list. Enabled by default.
1713
1714shared_media
1715------------
1716
1717If it is not set the kernel does not assume that different subnets on this
1718device can communicate directly. Default setting is 'yes'.
1719
1720send_redirects
1721--------------
1722
1723Determines whether to send ICMP redirects to other hosts.
1724
1725Routing settings
1726----------------
1727
1728The directory /proc/sys/net/ipv4/route contains several file to control
1729routing issues.
1730
1731error_burst and error_cost
1732--------------------------
1733
1734These parameters are used to limit how many ICMP destination unreachable to
1735send from the host in question. ICMP destination unreachable messages are
Matt LaPlante84eb8d02006-10-03 22:53:09 +02001736sent when we cannot reach the next hop while trying to transmit a packet.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001737It will also print some error messages to kernel logs if someone is ignoring
1738our ICMP redirects. The higher the error_cost factor is, the fewer
1739destination unreachable and error messages will be let through. Error_burst
1740controls when destination unreachable messages and error messages will be
1741dropped. The default settings limit warning messages to five every second.
1742
1743flush
1744-----
1745
1746Writing to this file results in a flush of the routing cache.
1747
1748gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
1749---------------------------------------------------------------------
1750
1751Values to control the frequency and behavior of the garbage collection
1752algorithm for the routing cache. gc_min_interval is deprecated and replaced
1753by gc_min_interval_ms.
1754
1755
1756max_size
1757--------
1758
1759Maximum size of the routing cache. Old entries will be purged once the cache
1760reached has this size.
1761
1762max_delay, min_delay
1763--------------------
1764
1765Delays for flushing the routing cache.
1766
1767redirect_load, redirect_number
1768------------------------------
1769
1770Factors which determine if more ICPM redirects should be sent to a specific
1771host. No redirects will be sent once the load limit or the maximum number of
1772redirects has been reached.
1773
1774redirect_silence
1775----------------
1776
1777Timeout for redirects. After this period redirects will be sent again, even if
1778this has been stopped, because the load or number limit has been reached.
1779
1780Network Neighbor handling
1781-------------------------
1782
1783Settings about how to handle connections with direct neighbors (nodes attached
1784to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
1785
1786As we saw it in the conf directory, there is a default subdirectory which
1787holds the default values, and one directory for each interface. The contents
1788of the directories are identical, with the single exception that the default
1789settings contain additional options to set garbage collection parameters.
1790
1791In the interface directories you'll find the following entries:
1792
1793base_reachable_time, base_reachable_time_ms
1794-------------------------------------------
1795
1796A base value used for computing the random reachable time value as specified
1797in RFC2461.
1798
1799Expression of base_reachable_time, which is deprecated, is in seconds.
1800Expression of base_reachable_time_ms is in milliseconds.
1801
1802retrans_time, retrans_time_ms
1803-----------------------------
1804
1805The time between retransmitted Neighbor Solicitation messages.
1806Used for address resolution and to determine if a neighbor is
1807unreachable.
1808
1809Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
1810IPv4) or in jiffies (for IPv6).
1811Expression of retrans_time_ms is in milliseconds.
1812
1813unres_qlen
1814----------
1815
1816Maximum queue length for a pending arp request - the number of packets which
1817are accepted from other layers while the ARP address is still resolved.
1818
1819anycast_delay
1820-------------
1821
1822Maximum for random delay of answers to neighbor solicitation messages in
1823jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
1824yet).
1825
1826ucast_solicit
1827-------------
1828
1829Maximum number of retries for unicast solicitation.
1830
1831mcast_solicit
1832-------------
1833
1834Maximum number of retries for multicast solicitation.
1835
1836delay_first_probe_time
1837----------------------
1838
1839Delay for the first time probe if the neighbor is reachable. (see
1840gc_stale_time)
1841
1842locktime
1843--------
1844
1845An ARP/neighbor entry is only replaced with a new one if the old is at least
1846locktime old. This prevents ARP cache thrashing.
1847
1848proxy_delay
1849-----------
1850
1851Maximum time (real time is random [0..proxytime]) before answering to an ARP
1852request for which we have an proxy ARP entry. In some cases, this is used to
1853prevent network flooding.
1854
1855proxy_qlen
1856----------
1857
1858Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
1859
1860app_solcit
1861----------
1862
1863Determines the number of requests to send to the user level ARP daemon. Use 0
1864to turn off.
1865
1866gc_stale_time
1867-------------
1868
1869Determines how often to check for stale ARP entries. After an ARP entry is
1870stale it will be resolved again (which is useful when an IP address migrates
1871to another machine). When ucast_solicit is greater than 0 it first tries to
1872send an ARP packet directly to the known host When that fails and
1873mcast_solicit is greater than 0, an ARP request is broadcasted.
1874
18752.9 Appletalk
1876-------------
1877
1878The /proc/sys/net/appletalk directory holds the Appletalk configuration data
1879when Appletalk is loaded. The configurable parameters are:
1880
1881aarp-expiry-time
1882----------------
1883
1884The amount of time we keep an ARP entry before expiring it. Used to age out
1885old hosts.
1886
1887aarp-resolve-time
1888-----------------
1889
1890The amount of time we will spend trying to resolve an Appletalk address.
1891
1892aarp-retransmit-limit
1893---------------------
1894
1895The number of times we will retransmit a query before giving up.
1896
1897aarp-tick-time
1898--------------
1899
1900Controls the rate at which expires are checked.
1901
1902The directory /proc/net/appletalk holds the list of active Appletalk sockets
1903on a machine.
1904
1905The fields indicate the DDP type, the local address (in network:node format)
1906the remote address, the size of the transmit pending queue, the size of the
1907received queue (bytes waiting for applications to read) the state and the uid
1908owning the socket.
1909
1910/proc/net/atalk_iface lists all the interfaces configured for appletalk.It
1911shows the name of the interface, its Appletalk address, the network range on
1912that address (or network number for phase 1 networks), and the status of the
1913interface.
1914
1915/proc/net/atalk_route lists each known network route. It lists the target
1916(network) that the route leads to, the router (may be directly connected), the
1917route flags, and the device the route is using.
1918
19192.10 IPX
1920--------
1921
1922The IPX protocol has no tunable values in proc/sys/net.
1923
1924The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
1925socket giving the local and remote addresses in Novell format (that is
1926network:node:port). In accordance with the strange Novell tradition,
1927everything but the port is in hex. Not_Connected is displayed for sockets that
1928are not tied to a specific remote address. The Tx and Rx queue sizes indicate
1929the number of bytes pending for transmission and reception. The state
1930indicates the state the socket is in and the uid is the owning uid of the
1931socket.
1932
1933The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
1934it gives the network number, the node number, and indicates if the network is
1935the primary network. It also indicates which device it is bound to (or
1936Internal for internal networks) and the Frame Type if appropriate. Linux
1937supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
1938IPX.
1939
1940The /proc/net/ipx_route table holds a list of IPX routes. For each route it
1941gives the destination network, the router node (or Directly) and the network
1942address of the router (or Connected) for internal networks.
1943
19442.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
1945----------------------------------------------------------
1946
1947The "mqueue" filesystem provides the necessary kernel features to enable the
1948creation of a user space library that implements the POSIX message queues
1949API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
1950Interfaces specification.)
1951
1952The "mqueue" filesystem contains values for determining/setting the amount of
1953resources used by the file system.
1954
1955/proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
1956maximum number of message queues allowed on the system.
1957
1958/proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
1959maximum number of messages in a queue value. In fact it is the limiting value
1960for another (user) limit which is set in mq_open invocation. This attribute of
1961a queue must be less or equal then msg_max.
1962
1963/proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
1964maximum message size value (it is every message queue's attribute set during
1965its creation).
1966
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -070019672.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
1968------------------------------------------------------
1969
1970This file can be used to adjust the score used to select which processes
1971should be killed in an out-of-memory situation. Giving it a high score will
1972increase the likelihood of this process being killed by the oom-killer. Valid
1973values are in the range -16 to +15, plus the special value -17, which disables
1974oom-killing altogether for this process.
1975
19762.13 /proc/<pid>/oom_score - Display current oom-killer score
1977-------------------------------------------------------------
1978
1979------------------------------------------------------------------------------
1980This file can be used to check the current score used by the oom-killer is for
1981any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1982process should be killed in an out-of-memory situation.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001983
1984------------------------------------------------------------------------------
1985Summary
1986------------------------------------------------------------------------------
1987Certain aspects of kernel behavior can be modified at runtime, without the
1988need to recompile the kernel, or even to reboot the system. The files in the
1989/proc/sys tree can not only be read, but also modified. You can use the echo
1990command to write value into these files, thereby changing the default settings
1991of the kernel.
1992------------------------------------------------------------------------------