1 ------------------------------------------------------------------------------
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>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 ------------------------------------------------------------------------------
9 Version 1.3 Kernel version 2.2.12
10 Kernel version 2.4.0-test11-pre4
11 ------------------------------------------------------------------------------
17 0.1 Introduction/Credits
20 1 Collecting System Information
21 1.1 Process-Specific Subdirectories
23 1.3 IDE devices in /proc/ide
24 1.4 Networking info in /proc/net
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
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
41 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
42 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
43 2.13 /proc/<pid>/oom_score - Display current oom-killer score
44 2.14 /proc/<pid>/io - Display the IO accounting fields
45 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
47 ------------------------------------------------------------------------------
49 ------------------------------------------------------------------------------
51 0.1 Introduction/Credits
52 ------------------------
54 This documentation is part of a soon (or so we hope) to be released book on
55 the SuSE Linux distribution. As there is no complete documentation for the
56 /proc file system and we've used many freely available sources to write these
57 chapters, it seems only fair to give the work back to the Linux community.
58 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
59 afraid it's still far from complete, but we hope it will be useful. As far as
60 we know, it is the first 'all-in-one' document about the /proc file system. It
61 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
62 SPARC, AXP, etc., features, you probably won't find what you are looking for.
63 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
64 additions and patches are welcome and will be added to this document if you
67 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
68 other people for help compiling this documentation. We'd also like to extend a
69 special thank you to Andi Kleen for documentation, which we relied on heavily
70 to create this document, as well as the additional information he provided.
71 Thanks to everybody else who contributed source or docs to the Linux kernel
72 and helped create a great piece of software... :)
74 If you have any comments, corrections or additions, please don't hesitate to
75 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
78 The latest version of this document is available online at
79 http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
81 If the above direction does not works for you, ypu could try the kernel
82 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
83 comandante@zaralinux.com.
88 We don't guarantee the correctness of this document, and if you come to us
89 complaining about how you screwed up your system because of incorrect
90 documentation, we won't feel responsible...
92 ------------------------------------------------------------------------------
93 CHAPTER 1: COLLECTING SYSTEM INFORMATION
94 ------------------------------------------------------------------------------
96 ------------------------------------------------------------------------------
98 ------------------------------------------------------------------------------
99 * Investigating the properties of the pseudo file system /proc and its
100 ability to provide information on the running Linux system
101 * Examining /proc's structure
102 * Uncovering various information about the kernel and the processes running
104 ------------------------------------------------------------------------------
107 The proc file system acts as an interface to internal data structures in the
108 kernel. It can be used to obtain information about the system and to change
109 certain kernel parameters at runtime (sysctl).
111 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
112 show you how you can use /proc/sys to change settings.
114 1.1 Process-Specific Subdirectories
115 -----------------------------------
117 The directory /proc contains (among other things) one subdirectory for each
118 process running on the system, which is named after the process ID (PID).
120 The link self points to the process reading the file system. Each process
121 subdirectory has the entries listed in Table 1-1.
124 Table 1-1: Process specific entries in /proc
125 ..............................................................................
127 clear_refs Clears page referenced bits shown in smaps output
128 cmdline Command line arguments
129 cpu Current and last cpu in which it was executed (2.4)(smp)
130 cwd Link to the current working directory
131 environ Values of environment variables
132 exe Link to the executable of this process
133 fd Directory, which contains all file descriptors
134 maps Memory maps to executables and library files (2.4)
135 mem Memory held by this process
136 root Link to the root directory of this process
138 statm Process memory status information
139 status Process status in human readable form
140 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
141 smaps Extension based on maps, the rss size for each mapped file
142 ..............................................................................
144 For example, to get the status information of a process, all you have to do is
145 read the file /proc/PID/status:
147 >cat /proc/self/status
163 SigPnd: 0000000000000000
164 SigBlk: 0000000000000000
165 SigIgn: 0000000000000000
166 SigCgt: 0000000000000000
167 CapInh: 00000000fffffeff
168 CapPrm: 0000000000000000
169 CapEff: 0000000000000000
172 This shows you nearly the same information you would get if you viewed it with
173 the ps command. In fact, ps uses the proc file system to obtain its
174 information. The statm file contains more detailed information about the
175 process memory usage. Its seven fields are explained in Table 1-2. The stat
176 file contains details information about the process itself. Its fields are
177 explained in Table 1-3.
180 Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
181 ..............................................................................
183 size total program size (pages) (same as VmSize in status)
184 resident size of memory portions (pages) (same as VmRSS in status)
185 shared number of pages that are shared (i.e. backed by a file)
186 trs number of pages that are 'code' (not including libs; broken,
187 includes data segment)
188 lrs number of pages of library (always 0 on 2.6)
189 drs number of pages of data/stack (including libs; broken,
190 includes library text)
191 dt number of dirty pages (always 0 on 2.6)
192 ..............................................................................
195 Table 1-3: Contents of the stat files (as of 2.6.22-rc3)
196 ..............................................................................
199 tcomm filename of the executable
200 state state (R is running, S is sleeping, D is sleeping in an
201 uninterruptible wait, Z is zombie, T is traced or stopped)
202 ppid process id of the parent process
203 pgrp pgrp of the process
205 tty_nr tty the process uses
206 tty_pgrp pgrp of the tty
208 min_flt number of minor faults
209 cmin_flt number of minor faults with child's
210 maj_flt number of major faults
211 cmaj_flt number of major faults with child's
212 utime user mode jiffies
213 stime kernel mode jiffies
214 cutime user mode jiffies with child's
215 cstime kernel mode jiffies with child's
216 priority priority level
218 num_threads number of threads
219 start_time time the process started after system boot
220 vsize virtual memory size
221 rss resident set memory size
222 rsslim current limit in bytes on the rss
223 start_code address above which program text can run
224 end_code address below which program text can run
225 start_stack address of the start of the stack
226 esp current value of ESP
227 eip current value of EIP
228 pending bitmap of pending signals (obsolete)
229 blocked bitmap of blocked signals (obsolete)
230 sigign bitmap of ignored signals (obsolete)
231 sigcatch bitmap of catched signals (obsolete)
232 wchan address where process went to sleep
235 exit_signal signal to send to parent thread on exit
236 task_cpu which CPU the task is scheduled on
237 rt_priority realtime priority
238 policy scheduling policy (man sched_setscheduler)
239 blkio_ticks time spent waiting for block IO
240 ..............................................................................
246 Similar to the process entries, the kernel data files give information about
247 the running kernel. The files used to obtain this information are contained in
248 /proc and are listed in Table 1-4. Not all of these will be present in your
249 system. It depends on the kernel configuration and the loaded modules, which
250 files are there, and which are missing.
252 Table 1-4: Kernel info in /proc
253 ..............................................................................
255 apm Advanced power management info
256 buddyinfo Kernel memory allocator information (see text) (2.5)
257 bus Directory containing bus specific information
258 cmdline Kernel command line
259 cpuinfo Info about the CPU
260 devices Available devices (block and character)
261 dma Used DMS channels
262 filesystems Supported filesystems
263 driver Various drivers grouped here, currently rtc (2.4)
264 execdomains Execdomains, related to security (2.4)
265 fb Frame Buffer devices (2.4)
266 fs File system parameters, currently nfs/exports (2.4)
267 ide Directory containing info about the IDE subsystem
268 interrupts Interrupt usage
269 iomem Memory map (2.4)
270 ioports I/O port usage
271 irq Masks for irq to cpu affinity (2.4)(smp?)
272 isapnp ISA PnP (Plug&Play) Info (2.4)
273 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
275 ksyms Kernel symbol table
276 loadavg Load average of last 1, 5 & 15 minutes
280 modules List of loaded modules
281 mounts Mounted filesystems
282 net Networking info (see text)
283 partitions Table of partitions known to the system
284 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
285 decoupled by lspci (2.4)
287 scsi SCSI info (see text)
288 slabinfo Slab pool info
289 stat Overall statistics
290 swaps Swap space utilization
292 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
293 tty Info of tty drivers
295 version Kernel version
296 video bttv info of video resources (2.4)
297 ..............................................................................
299 You can, for example, check which interrupts are currently in use and what
300 they are used for by looking in the file /proc/interrupts:
302 > cat /proc/interrupts
304 0: 8728810 XT-PIC timer
305 1: 895 XT-PIC keyboard
307 3: 531695 XT-PIC aha152x
308 4: 2014133 XT-PIC serial
309 5: 44401 XT-PIC pcnet_cs
312 12: 182918 XT-PIC PS/2 Mouse
314 14: 1232265 XT-PIC ide0
318 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
319 output of a SMP machine):
321 > cat /proc/interrupts
324 0: 1243498 1214548 IO-APIC-edge timer
325 1: 8949 8958 IO-APIC-edge keyboard
326 2: 0 0 XT-PIC cascade
327 5: 11286 10161 IO-APIC-edge soundblaster
328 8: 1 0 IO-APIC-edge rtc
329 9: 27422 27407 IO-APIC-edge 3c503
330 12: 113645 113873 IO-APIC-edge PS/2 Mouse
332 14: 22491 24012 IO-APIC-edge ide0
333 15: 2183 2415 IO-APIC-edge ide1
334 17: 30564 30414 IO-APIC-level eth0
335 18: 177 164 IO-APIC-level bttv
340 NMI is incremented in this case because every timer interrupt generates a NMI
341 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
343 LOC is the local interrupt counter of the internal APIC of every CPU.
345 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
346 connects the CPUs in a SMP system. This means that an error has been detected,
347 the IO-APIC automatically retry the transmission, so it should not be a big
348 problem, but you should read the SMP-FAQ.
350 In this context it could be interesting to note the new irq directory in 2.4.
351 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
352 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
353 irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
357 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
358 1 11 13 15 17 19 3 5 7 9
362 The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
363 is the same by default:
365 > cat /proc/irq/0/smp_affinity
368 It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can
371 > echo 1 > /proc/irq/prof_cpu_mask
373 This means that only the first CPU will handle the IRQ, but you can also echo 5
374 which means that only the first and fourth CPU can handle the IRQ.
376 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
377 between all the CPUs which are allowed to handle it. As usual the kernel has
378 more info than you and does a better job than you, so the defaults are the
379 best choice for almost everyone.
381 There are three more important subdirectories in /proc: net, scsi, and sys.
382 The general rule is that the contents, or even the existence of these
383 directories, depend on your kernel configuration. If SCSI is not enabled, the
384 directory scsi may not exist. The same is true with the net, which is there
385 only when networking support is present in the running kernel.
387 The slabinfo file gives information about memory usage at the slab level.
388 Linux uses slab pools for memory management above page level in version 2.2.
389 Commonly used objects have their own slab pool (such as network buffers,
390 directory cache, and so on).
392 ..............................................................................
394 > cat /proc/buddyinfo
396 Node 0, zone DMA 0 4 5 4 4 3 ...
397 Node 0, zone Normal 1 0 0 1 101 8 ...
398 Node 0, zone HighMem 2 0 0 1 1 0 ...
400 Memory fragmentation is a problem under some workloads, and buddyinfo is a
401 useful tool for helping diagnose these problems. Buddyinfo will give you a
402 clue as to how big an area you can safely allocate, or why a previous
405 Each column represents the number of pages of a certain order which are
406 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
407 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
408 available in ZONE_NORMAL, etc...
410 ..............................................................................
414 Provides information about distribution and utilization of memory. This
415 varies by architecture and compile options. The following is from a
416 16GB PIII, which has highmem enabled. You may not have all of these fields.
421 MemTotal: 16344972 kB
428 HighTotal: 15597528 kB
429 HighFree: 13629632 kB
438 CommitLimit: 7669796 kB
439 Committed_AS: 100056 kB
441 VmallocTotal: 112216 kB
443 VmallocChunk: 111088 kB
445 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
446 bits and the kernel binary code)
447 MemFree: The sum of LowFree+HighFree
448 Buffers: Relatively temporary storage for raw disk blocks
449 shouldn't get tremendously large (20MB or so)
450 Cached: in-memory cache for files read from the disk (the
451 pagecache). Doesn't include SwapCached
452 SwapCached: Memory that once was swapped out, is swapped back in but
453 still also is in the swapfile (if memory is needed it
454 doesn't need to be swapped out AGAIN because it is already
455 in the swapfile. This saves I/O)
456 Active: Memory that has been used more recently and usually not
457 reclaimed unless absolutely necessary.
458 Inactive: Memory which has been less recently used. It is more
459 eligible to be reclaimed for other purposes
461 HighFree: Highmem is all memory above ~860MB of physical memory
462 Highmem areas are for use by userspace programs, or
463 for the pagecache. The kernel must use tricks to access
464 this memory, making it slower to access than lowmem.
466 LowFree: Lowmem is memory which can be used for everything that
467 highmem can be used for, but it is also available for the
468 kernel's use for its own data structures. Among many
469 other things, it is where everything from the Slab is
470 allocated. Bad things happen when you're out of lowmem.
471 SwapTotal: total amount of swap space available
472 SwapFree: Memory which has been evicted from RAM, and is temporarily
474 Dirty: Memory which is waiting to get written back to the disk
475 Writeback: Memory which is actively being written back to the disk
476 Mapped: files which have been mmaped, such as libraries
477 Slab: in-kernel data structures cache
478 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
479 this is the total amount of memory currently available to
480 be allocated on the system. This limit is only adhered to
481 if strict overcommit accounting is enabled (mode 2 in
482 'vm.overcommit_memory').
483 The CommitLimit is calculated with the following formula:
484 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
485 For example, on a system with 1G of physical RAM and 7G
486 of swap with a `vm.overcommit_ratio` of 30 it would
487 yield a CommitLimit of 7.3G.
488 For more details, see the memory overcommit documentation
489 in vm/overcommit-accounting.
490 Committed_AS: The amount of memory presently allocated on the system.
491 The committed memory is a sum of all of the memory which
492 has been allocated by processes, even if it has not been
493 "used" by them as of yet. A process which malloc()'s 1G
494 of memory, but only touches 300M of it will only show up
495 as using 300M of memory even if it has the address space
496 allocated for the entire 1G. This 1G is memory which has
497 been "committed" to by the VM and can be used at any time
498 by the allocating application. With strict overcommit
499 enabled on the system (mode 2 in 'vm.overcommit_memory'),
500 allocations which would exceed the CommitLimit (detailed
501 above) will not be permitted. This is useful if one needs
502 to guarantee that processes will not fail due to lack of
503 memory once that memory has been successfully allocated.
504 PageTables: amount of memory dedicated to the lowest level of page
506 VmallocTotal: total size of vmalloc memory area
507 VmallocUsed: amount of vmalloc area which is used
508 VmallocChunk: largest contigious block of vmalloc area which is free
511 1.3 IDE devices in /proc/ide
512 ----------------------------
514 The subdirectory /proc/ide contains information about all IDE devices of which
515 the kernel is aware. There is one subdirectory for each IDE controller, the
516 file drivers and a link for each IDE device, pointing to the device directory
517 in the controller specific subtree.
519 The file drivers contains general information about the drivers used for the
522 > cat /proc/ide/drivers
523 ide-cdrom version 4.53
524 ide-disk version 1.08
526 More detailed information can be found in the controller specific
527 subdirectories. These are named ide0, ide1 and so on. Each of these
528 directories contains the files shown in table 1-5.
531 Table 1-5: IDE controller info in /proc/ide/ide?
532 ..............................................................................
534 channel IDE channel (0 or 1)
535 config Configuration (only for PCI/IDE bridge)
537 model Type/Chipset of IDE controller
538 ..............................................................................
540 Each device connected to a controller has a separate subdirectory in the
541 controllers directory. The files listed in table 1-6 are contained in these
545 Table 1-6: IDE device information
546 ..............................................................................
549 capacity Capacity of the medium (in 512Byte blocks)
550 driver driver and version
551 geometry physical and logical geometry
552 identify device identify block
554 model device identifier
555 settings device setup
556 smart_thresholds IDE disk management thresholds
557 smart_values IDE disk management values
558 ..............................................................................
560 The most interesting file is settings. This file contains a nice overview of
561 the drive parameters:
563 # cat /proc/ide/ide0/hda/settings
564 name value min max mode
565 ---- ----- --- --- ----
566 bios_cyl 526 0 65535 rw
567 bios_head 255 0 255 rw
569 breada_readahead 4 0 127 rw
571 file_readahead 72 0 2097151 rw
573 keepsettings 0 0 1 rw
574 max_kb_per_request 122 1 127 rw
578 pio_mode write-only 0 255 w
584 1.4 Networking info in /proc/net
585 --------------------------------
587 The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
588 additional values you get for IP version 6 if you configure the kernel to
589 support this. Table 1-7 lists the files and their meaning.
592 Table 1-6: IPv6 info in /proc/net
593 ..............................................................................
595 udp6 UDP sockets (IPv6)
596 tcp6 TCP sockets (IPv6)
597 raw6 Raw device statistics (IPv6)
598 igmp6 IP multicast addresses, which this host joined (IPv6)
599 if_inet6 List of IPv6 interface addresses
600 ipv6_route Kernel routing table for IPv6
601 rt6_stats Global IPv6 routing tables statistics
602 sockstat6 Socket statistics (IPv6)
603 snmp6 Snmp data (IPv6)
604 ..............................................................................
607 Table 1-7: Network info in /proc/net
608 ..............................................................................
611 dev network devices with statistics
612 dev_mcast the Layer2 multicast groups a device is listening too
613 (interface index, label, number of references, number of bound
615 dev_stat network device status
616 ip_fwchains Firewall chain linkage
617 ip_fwnames Firewall chain names
618 ip_masq Directory containing the masquerading tables
619 ip_masquerade Major masquerading table
620 netstat Network statistics
621 raw raw device statistics
622 route Kernel routing table
623 rpc Directory containing rpc info
624 rt_cache Routing cache
626 sockstat Socket statistics
628 tr_rif Token ring RIF routing table
630 unix UNIX domain sockets
631 wireless Wireless interface data (Wavelan etc)
632 igmp IP multicast addresses, which this host joined
633 psched Global packet scheduler parameters.
634 netlink List of PF_NETLINK sockets
635 ip_mr_vifs List of multicast virtual interfaces
636 ip_mr_cache List of multicast routing cache
637 ..............................................................................
639 You can use this information to see which network devices are available in
640 your system and how much traffic was routed over those devices:
644 face |bytes packets errs drop fifo frame compressed multicast|[...
645 lo: 908188 5596 0 0 0 0 0 0 [...
646 ppp0:15475140 20721 410 0 0 410 0 0 [...
647 eth0: 614530 7085 0 0 0 0 0 1 [...
650 ...] bytes packets errs drop fifo colls carrier compressed
651 ...] 908188 5596 0 0 0 0 0 0
652 ...] 1375103 17405 0 0 0 0 0 0
653 ...] 1703981 5535 0 0 0 3 0 0
655 In addition, each Channel Bond interface has it's own directory. For
656 example, the bond0 device will have a directory called /proc/net/bond0/.
657 It will contain information that is specific to that bond, such as the
658 current slaves of the bond, the link status of the slaves, and how
659 many times the slaves link has failed.
664 If you have a SCSI host adapter in your system, you'll find a subdirectory
665 named after the driver for this adapter in /proc/scsi. You'll also see a list
666 of all recognized SCSI devices in /proc/scsi:
670 Host: scsi0 Channel: 00 Id: 00 Lun: 00
671 Vendor: IBM Model: DGHS09U Rev: 03E0
672 Type: Direct-Access ANSI SCSI revision: 03
673 Host: scsi0 Channel: 00 Id: 06 Lun: 00
674 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
675 Type: CD-ROM ANSI SCSI revision: 02
678 The directory named after the driver has one file for each adapter found in
679 the system. These files contain information about the controller, including
680 the used IRQ and the IO address range. The amount of information shown is
681 dependent on the adapter you use. The example shows the output for an Adaptec
682 AHA-2940 SCSI adapter:
684 > cat /proc/scsi/aic7xxx/0
686 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
688 TCQ Enabled By Default : Disabled
689 AIC7XXX_PROC_STATS : Disabled
690 AIC7XXX_RESET_DELAY : 5
691 Adapter Configuration:
692 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
693 Ultra Wide Controller
694 PCI MMAPed I/O Base: 0xeb001000
695 Adapter SEEPROM Config: SEEPROM found and used.
696 Adaptec SCSI BIOS: Enabled
698 SCBs: Active 0, Max Active 2,
699 Allocated 15, HW 16, Page 255
701 BIOS Control Word: 0x18b6
702 Adapter Control Word: 0x005b
703 Extended Translation: Enabled
704 Disconnect Enable Flags: 0xffff
705 Ultra Enable Flags: 0x0001
706 Tag Queue Enable Flags: 0x0000
707 Ordered Queue Tag Flags: 0x0000
708 Default Tag Queue Depth: 8
709 Tagged Queue By Device array for aic7xxx host instance 0:
710 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
711 Actual queue depth per device for aic7xxx host instance 0:
712 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
715 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
716 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
717 Total transfers 160151 (74577 reads and 85574 writes)
719 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
720 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
721 Total transfers 0 (0 reads and 0 writes)
724 1.6 Parallel port info in /proc/parport
725 ---------------------------------------
727 The directory /proc/parport contains information about the parallel ports of
728 your system. It has one subdirectory for each port, named after the port
731 These directories contain the four files shown in Table 1-8.
734 Table 1-8: Files in /proc/parport
735 ..............................................................................
737 autoprobe Any IEEE-1284 device ID information that has been acquired.
738 devices list of the device drivers using that port. A + will appear by the
739 name of the device currently using the port (it might not appear
741 hardware Parallel port's base address, IRQ line and DMA channel.
742 irq IRQ that parport is using for that port. This is in a separate
743 file to allow you to alter it by writing a new value in (IRQ
745 ..............................................................................
747 1.7 TTY info in /proc/tty
748 -------------------------
750 Information about the available and actually used tty's can be found in the
751 directory /proc/tty.You'll find entries for drivers and line disciplines in
752 this directory, as shown in Table 1-9.
755 Table 1-9: Files in /proc/tty
756 ..............................................................................
758 drivers list of drivers and their usage
759 ldiscs registered line disciplines
760 driver/serial usage statistic and status of single tty lines
761 ..............................................................................
763 To see which tty's are currently in use, you can simply look into the file
766 > cat /proc/tty/drivers
767 pty_slave /dev/pts 136 0-255 pty:slave
768 pty_master /dev/ptm 128 0-255 pty:master
769 pty_slave /dev/ttyp 3 0-255 pty:slave
770 pty_master /dev/pty 2 0-255 pty:master
771 serial /dev/cua 5 64-67 serial:callout
772 serial /dev/ttyS 4 64-67 serial
773 /dev/tty0 /dev/tty0 4 0 system:vtmaster
774 /dev/ptmx /dev/ptmx 5 2 system
775 /dev/console /dev/console 5 1 system:console
776 /dev/tty /dev/tty 5 0 system:/dev/tty
777 unknown /dev/tty 4 1-63 console
780 1.8 Miscellaneous kernel statistics in /proc/stat
781 -------------------------------------------------
783 Various pieces of information about kernel activity are available in the
784 /proc/stat file. All of the numbers reported in this file are aggregates
785 since the system first booted. For a quick look, simply cat the file:
788 cpu 2255 34 2290 22625563 6290 127 456
789 cpu0 1132 34 1441 11311718 3675 127 438
790 cpu1 1123 0 849 11313845 2614 0 18
791 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
798 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
799 lines. These numbers identify the amount of time the CPU has spent performing
800 different kinds of work. Time units are in USER_HZ (typically hundredths of a
801 second). The meanings of the columns are as follows, from left to right:
803 - user: normal processes executing in user mode
804 - nice: niced processes executing in user mode
805 - system: processes executing in kernel mode
806 - idle: twiddling thumbs
807 - iowait: waiting for I/O to complete
808 - irq: servicing interrupts
809 - softirq: servicing softirqs
811 The "intr" line gives counts of interrupts serviced since boot time, for each
812 of the possible system interrupts. The first column is the total of all
813 interrupts serviced; each subsequent column is the total for that particular
816 The "ctxt" line gives the total number of context switches across all CPUs.
818 The "btime" line gives the time at which the system booted, in seconds since
821 The "processes" line gives the number of processes and threads created, which
822 includes (but is not limited to) those created by calls to the fork() and
823 clone() system calls.
825 The "procs_running" line gives the number of processes currently running on
828 The "procs_blocked" line gives the number of processes currently blocked,
829 waiting for I/O to complete.
832 ------------------------------------------------------------------------------
834 ------------------------------------------------------------------------------
835 The /proc file system serves information about the running system. It not only
836 allows access to process data but also allows you to request the kernel status
837 by reading files in the hierarchy.
839 The directory structure of /proc reflects the types of information and makes
840 it easy, if not obvious, where to look for specific data.
841 ------------------------------------------------------------------------------
843 ------------------------------------------------------------------------------
844 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
845 ------------------------------------------------------------------------------
847 ------------------------------------------------------------------------------
849 ------------------------------------------------------------------------------
850 * Modifying kernel parameters by writing into files found in /proc/sys
851 * Exploring the files which modify certain parameters
852 * Review of the /proc/sys file tree
853 ------------------------------------------------------------------------------
856 A very interesting part of /proc is the directory /proc/sys. This is not only
857 a source of information, it also allows you to change parameters within the
858 kernel. Be very careful when attempting this. You can optimize your system,
859 but you can also cause it to crash. Never alter kernel parameters on a
860 production system. Set up a development machine and test to make sure that
861 everything works the way you want it to. You may have no alternative but to
862 reboot the machine once an error has been made.
864 To change a value, simply echo the new value into the file. An example is
865 given below in the section on the file system data. You need to be root to do
866 this. You can create your own boot script to perform this every time your
869 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
870 general things in the operation of the Linux kernel. Since some of the files
871 can inadvertently disrupt your system, it is advisable to read both
872 documentation and source before actually making adjustments. In any case, be
873 very careful when writing to any of these files. The entries in /proc may
874 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
875 review the kernel documentation in the directory /usr/src/linux/Documentation.
876 This chapter is heavily based on the documentation included in the pre 2.2
877 kernels, and became part of it in version 2.2.1 of the Linux kernel.
879 2.1 /proc/sys/fs - File system data
880 -----------------------------------
882 This subdirectory contains specific file system, file handle, inode, dentry
883 and quota information.
885 Currently, these files are in /proc/sys/fs:
890 Status of the directory cache. Since directory entries are dynamically
891 allocated and deallocated, this file indicates the current status. It holds
892 six values, in which the last two are not used and are always zero. The others
893 are listed in table 2-1.
896 Table 2-1: Status files of the directory cache
897 ..............................................................................
899 nr_dentry Almost always zero
900 nr_unused Number of unused cache entries
902 in seconds after the entry may be reclaimed, when memory is short
903 want_pages internally
904 ..............................................................................
906 dquot-nr and dquot-max
907 ----------------------
909 The file dquot-max shows the maximum number of cached disk quota entries.
911 The file dquot-nr shows the number of allocated disk quota entries and the
912 number of free disk quota entries.
914 If the number of available cached disk quotas is very low and you have a large
915 number of simultaneous system users, you might want to raise the limit.
920 The kernel allocates file handles dynamically, but doesn't free them again at
923 The value in file-max denotes the maximum number of file handles that the
924 Linux kernel will allocate. When you get a lot of error messages about running
925 out of file handles, you might want to raise this limit. The default value is
926 10% of RAM in kilobytes. To change it, just write the new number into the
929 # cat /proc/sys/fs/file-max
931 # echo 8192 > /proc/sys/fs/file-max
932 # cat /proc/sys/fs/file-max
936 This method of revision is useful for all customizable parameters of the
937 kernel - simply echo the new value to the corresponding file.
939 Historically, the three values in file-nr denoted the number of allocated file
940 handles, the number of allocated but unused file handles, and the maximum
941 number of file handles. Linux 2.6 always reports 0 as the number of free file
942 handles -- this is not an error, it just means that the number of allocated
943 file handles exactly matches the number of used file handles.
945 Attempts to allocate more file descriptors than file-max are reported with
946 printk, look for "VFS: file-max limit <number> reached".
948 inode-state and inode-nr
949 ------------------------
951 The file inode-nr contains the first two items from inode-state, so we'll skip
954 inode-state contains two actual numbers and five dummy values. The numbers
955 are nr_inodes and nr_free_inodes (in order of appearance).
960 Denotes the number of inodes the system has allocated. This number will
961 grow and shrink dynamically.
966 Represents the number of free inodes. Ie. The number of inuse inodes is
967 (nr_inodes - nr_free_inodes).
969 aio-nr and aio-max-nr
970 ---------------------
972 aio-nr is the running total of the number of events specified on the
973 io_setup system call for all currently active aio contexts. If aio-nr
974 reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
975 raising aio-max-nr does not result in the pre-allocation or re-sizing
976 of any kernel data structures.
978 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
979 -----------------------------------------------------------
981 Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
982 handles the kernel support for miscellaneous binary formats.
984 Binfmt_misc provides the ability to register additional binary formats to the
985 Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
986 needs to know magic numbers at the beginning or the filename extension of the
989 It works by maintaining a linked list of structs that contain a description of
990 a binary format, including a magic with size (or the filename extension),
991 offset and mask, and the interpreter name. On request it invokes the given
992 interpreter with the original program as argument, as binfmt_java and
993 binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
994 binary-formats, you have to register an additional binary-format.
996 There are two general files in binfmt_misc and one file per registered format.
997 The two general files are register and status.
999 Registering a new binary format
1000 -------------------------------
1002 To register a new binary format you have to issue the command
1004 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
1008 with appropriate name (the name for the /proc-dir entry), offset (defaults to
1009 0, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
1010 last but not least, the interpreter that is to be invoked (for example and
1011 testing /bin/echo). Type can be M for usual magic matching or E for filename
1012 extension matching (give extension in place of magic).
1014 Check or reset the status of the binary format handler
1015 ------------------------------------------------------
1017 If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
1018 current status (enabled/disabled) of binfmt_misc. Change the status by echoing
1019 0 (disables) or 1 (enables) or -1 (caution: this clears all previously
1020 registered binary formats) to status. For example echo 0 > status to disable
1021 binfmt_misc (temporarily).
1023 Status of a single handler
1024 --------------------------
1026 Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
1027 perform the same function as status, but their scope is limited to the actual
1028 binary format. By cating this file, you also receive all related information
1029 about the interpreter/magic of the binfmt.
1031 Example usage of binfmt_misc (emulate binfmt_java)
1032 --------------------------------------------------
1034 cd /proc/sys/fs/binfmt_misc
1035 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
1036 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
1037 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
1038 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
1041 These four lines add support for Java executables and Java applets (like
1042 binfmt_java, additionally recognizing the .html extension with no need to put
1043 <!--applet> to every applet file). You have to install the JDK and the
1044 shell-script /usr/local/java/bin/javawrapper too. It works around the
1045 brokenness of the Java filename handling. To add a Java binary, just create a
1046 link to the class-file somewhere in the path.
1048 2.3 /proc/sys/kernel - general kernel parameters
1049 ------------------------------------------------
1051 This directory reflects general kernel behaviors. As I've said before, the
1052 contents depend on your configuration. Here you'll find the most important
1053 files, along with descriptions of what they mean and how to use them.
1058 The file contains three values; highwater, lowwater, and frequency.
1060 It exists only when BSD-style process accounting is enabled. These values
1061 control its behavior. If the free space on the file system where the log lives
1062 goes below lowwater percentage, accounting suspends. If it goes above
1063 highwater percentage, accounting resumes. Frequency determines how often you
1064 check the amount of free space (value is in seconds). Default settings are: 4,
1065 2, and 30. That is, suspend accounting if there is less than 2 percent free;
1066 resume it if we have a value of 3 or more percent; consider information about
1067 the amount of free space valid for 30 seconds
1072 The file contains a single value denoting the limit on the argv array size
1073 for execve (in KiB). This limit is only applied when system call auditing for
1074 execve is enabled, otherwise the value is ignored.
1079 When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1080 program to handle a graceful restart. However, when the value is greater that
1081 zero, Linux's reaction to this key combination will be an immediate reboot,
1082 without syncing its dirty buffers.
1085 When a program (like dosemu) has the keyboard in raw mode, the
1086 ctrl-alt-del is intercepted by the program before it ever reaches the
1087 kernel tty layer, and it is up to the program to decide what to do with
1090 domainname and hostname
1091 -----------------------
1093 These files can be controlled to set the NIS domainname and hostname of your
1094 box. For the classic darkstar.frop.org a simple:
1096 # echo "darkstar" > /proc/sys/kernel/hostname
1097 # echo "frop.org" > /proc/sys/kernel/domainname
1100 would suffice to set your hostname and NIS domainname.
1102 osrelease, ostype and version
1103 -----------------------------
1105 The names make it pretty obvious what these fields contain:
1107 > cat /proc/sys/kernel/osrelease
1110 > cat /proc/sys/kernel/ostype
1113 > cat /proc/sys/kernel/version
1114 #4 Fri Oct 1 12:41:14 PDT 1999
1117 The files osrelease and ostype should be clear enough. Version needs a little
1118 more clarification. The #4 means that this is the 4th kernel built from this
1119 source base and the date after it indicates the time the kernel was built. The
1120 only way to tune these values is to rebuild the kernel.
1125 The value in this file represents the number of seconds the kernel waits
1126 before rebooting on a panic. When you use the software watchdog, the
1127 recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1128 is disabled, which is the default setting.
1133 The four values in printk denote
1135 * default_message_loglevel,
1136 * minimum_console_loglevel and
1137 * default_console_loglevel
1140 These values influence printk() behavior when printing or logging error
1141 messages, which come from inside the kernel. See syslog(2) for more
1142 information on the different log levels.
1147 Messages with a higher priority than this will be printed to the console.
1149 default_message_level
1150 ---------------------
1152 Messages without an explicit priority will be printed with this priority.
1154 minimum_console_loglevel
1155 ------------------------
1157 Minimum (highest) value to which the console_loglevel can be set.
1159 default_console_loglevel
1160 ------------------------
1162 Default value for console_loglevel.
1167 This file shows the size of the generic SCSI (sg) buffer. At this point, you
1168 can't tune it yet, but you can change it at compile time by editing
1169 include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1171 If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1172 this to a higher value. Refer to the SANE documentation on this issue.
1177 The location where the modprobe binary is located. The kernel uses this
1178 program to load modules on demand.
1183 The value in this file affects behavior of handling NMI. When the value is
1184 non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1185 debugging information is displayed on console.
1187 NMI switch that most IA32 servers have fires unknown NMI up, for example.
1188 If a system hangs up, try pressing the NMI switch.
1193 Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
1194 the NMI watchdog is enabled and will continuously test all online cpus to
1195 determine whether or not they are still functioning properly.
1197 Because the NMI watchdog shares registers with oprofile, by disabling the NMI
1198 watchdog, oprofile may have more registers to utilize.
1203 Enables/Disables the protection of the per-process proc entries "maps" and
1204 "smaps". When enabled, the contents of these files are visible only to
1205 readers that are allowed to ptrace() the given process.
1208 2.4 /proc/sys/vm - The virtual memory subsystem
1209 -----------------------------------------------
1211 The files in this directory can be used to tune the operation of the virtual
1212 memory (VM) subsystem of the Linux kernel.
1217 Controls the tendency of the kernel to reclaim the memory which is used for
1218 caching of directory and inode objects.
1220 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1221 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1222 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1223 to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1224 causes the kernel to prefer to reclaim dentries and inodes.
1226 dirty_background_ratio
1227 ----------------------
1229 Contains, as a percentage of total system memory, the number of pages at which
1230 the pdflush background writeback daemon will start writing out dirty data.
1235 Contains, as a percentage of total system memory, the number of pages at which
1236 a process which is generating disk writes will itself start writing out dirty
1239 dirty_writeback_centisecs
1240 -------------------------
1242 The pdflush writeback daemons will periodically wake up and write `old' data
1243 out to disk. This tunable expresses the interval between those wakeups, in
1244 100'ths of a second.
1246 Setting this to zero disables periodic writeback altogether.
1248 dirty_expire_centisecs
1249 ----------------------
1251 This tunable is used to define when dirty data is old enough to be eligible
1252 for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1253 Data which has been dirty in-memory for longer than this interval will be
1254 written out next time a pdflush daemon wakes up.
1259 If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1260 will use the legacy (2.4) layout for all processes.
1262 lower_zone_protection
1263 ---------------------
1265 For some specialised workloads on highmem machines it is dangerous for
1266 the kernel to allow process memory to be allocated from the "lowmem"
1267 zone. This is because that memory could then be pinned via the mlock()
1268 system call, or by unavailability of swapspace.
1270 And on large highmem machines this lack of reclaimable lowmem memory
1273 So the Linux page allocator has a mechanism which prevents allocations
1274 which _could_ use highmem from using too much lowmem. This means that
1275 a certain amount of lowmem is defended from the possibility of being
1276 captured into pinned user memory.
1278 (The same argument applies to the old 16 megabyte ISA DMA region. This
1279 mechanism will also defend that region from allocations which could use
1282 The `lower_zone_protection' tunable determines how aggressive the kernel is
1283 in defending these lower zones. The default value is zero - no
1286 If you have a machine which uses highmem or ISA DMA and your
1287 applications are using mlock(), or if you are running with no swap then
1288 you probably should increase the lower_zone_protection setting.
1290 The units of this tunable are fairly vague. It is approximately equal
1291 to "megabytes," so setting lower_zone_protection=100 will protect around 100
1292 megabytes of the lowmem zone from user allocations. It will also make
1293 those 100 megabytes unavailable for use by applications and by
1294 pagecache, so there is a cost.
1296 The effects of this tunable may be observed by monitoring
1297 /proc/meminfo:LowFree. Write a single huge file and observe the point
1298 at which LowFree ceases to fall.
1300 A reasonable value for lower_zone_protection is 100.
1305 page-cluster controls the number of pages which are written to swap in
1306 a single attempt. The swap I/O size.
1308 It is a logarithmic value - setting it to zero means "1 page", setting
1309 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1311 The default value is three (eight pages at a time). There may be some
1312 small benefits in tuning this to a different value if your workload is
1318 Controls overcommit of system memory, possibly allowing processes
1319 to allocate (but not use) more memory than is actually available.
1322 0 - Heuristic overcommit handling. Obvious overcommits of
1323 address space are refused. Used for a typical system. It
1324 ensures a seriously wild allocation fails while allowing
1325 overcommit to reduce swap usage. root is allowed to
1326 allocate slightly more memory in this mode. This is the
1329 1 - Always overcommit. Appropriate for some scientific
1332 2 - Don't overcommit. The total address space commit
1333 for the system is not permitted to exceed swap plus a
1334 configurable percentage (default is 50) of physical RAM.
1335 Depending on the percentage you use, in most situations
1336 this means a process will not be killed while attempting
1337 to use already-allocated memory but will receive errors
1338 on memory allocation as appropriate.
1343 Percentage of physical memory size to include in overcommit calculations
1346 Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
1348 swapspace = total size of all swap areas
1349 physmem = size of physical memory in system
1351 nr_hugepages and hugetlb_shm_group
1352 ----------------------------------
1354 nr_hugepages configures number of hugetlb page reserved for the system.
1356 hugetlb_shm_group contains group id that is allowed to create SysV shared
1357 memory segment using hugetlb page.
1359 hugepages_treat_as_movable
1360 --------------------------
1362 This parameter is only useful when kernelcore= is specified at boot time to
1363 create ZONE_MOVABLE for pages that may be reclaimed or migrated. Huge pages
1364 are not movable so are not normally allocated from ZONE_MOVABLE. A non-zero
1365 value written to hugepages_treat_as_movable allows huge pages to be allocated
1368 Once enabled, the ZONE_MOVABLE is treated as an area of memory the huge
1369 pages pool can easily grow or shrink within. Assuming that applications are
1370 not running that mlock() a lot of memory, it is likely the huge pages pool
1371 can grow to the size of ZONE_MOVABLE by repeatedly entering the desired value
1372 into nr_hugepages and triggering page reclaim.
1377 laptop_mode is a knob that controls "laptop mode". All the things that are
1378 controlled by this knob are discussed in Documentation/laptop-mode.txt.
1383 block_dump enables block I/O debugging when set to a nonzero value. More
1384 information on block I/O debugging is in Documentation/laptop-mode.txt.
1389 This file contains valid hold time of swap out protection token. The Linux
1390 VM has token based thrashing control mechanism and uses the token to prevent
1391 unnecessary page faults in thrashing situation. The unit of the value is
1392 second. The value would be useful to tune thrashing behavior.
1397 Writing to this will cause the kernel to drop clean caches, dentries and
1398 inodes from memory, causing that memory to become free.
1401 echo 1 > /proc/sys/vm/drop_caches
1402 To free dentries and inodes:
1403 echo 2 > /proc/sys/vm/drop_caches
1404 To free pagecache, dentries and inodes:
1405 echo 3 > /proc/sys/vm/drop_caches
1407 As this is a non-destructive operation and dirty objects are not freeable, the
1408 user should run `sync' first.
1411 2.5 /proc/sys/dev - Device specific parameters
1412 ----------------------------------------------
1414 Currently there is only support for CDROM drives, and for those, there is only
1415 one read-only file containing information about the CD-ROM drives attached to
1418 >cat /proc/sys/dev/cdrom/info
1419 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1423 drive # of slots: 1 0
1427 Can change speed: 1 1
1428 Can select disk: 0 1
1429 Can read multisession: 1 1
1431 Reports media changed: 1 1
1435 You see two drives, sr0 and hdb, along with a list of their features.
1437 2.6 /proc/sys/sunrpc - Remote procedure calls
1438 ---------------------------------------------
1440 This directory contains four files, which enable or disable debugging for the
1441 RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1442 be set to one to turn debugging on. (The default value is 0 for each)
1444 2.7 /proc/sys/net - Networking stuff
1445 ------------------------------------
1447 The interface to the networking parts of the kernel is located in
1448 /proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1449 some of them, depending on your kernel's configuration.
1452 Table 2-3: Subdirectories in /proc/sys/net
1453 ..............................................................................
1454 Directory Content Directory Content
1455 core General parameter appletalk Appletalk protocol
1456 unix Unix domain sockets netrom NET/ROM
1457 802 E802 protocol ax25 AX25
1458 ethernet Ethernet protocol rose X.25 PLP layer
1459 ipv4 IP version 4 x25 X.25 protocol
1460 ipx IPX token-ring IBM token ring
1461 bridge Bridging decnet DEC net
1463 ..............................................................................
1465 We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1466 only minor players in the Linux world, we'll skip them in this chapter. You'll
1467 find some short info on Appletalk and IPX further on in this chapter. Review
1468 the online documentation and the kernel source to get a detailed view of the
1469 parameters for those protocols. In this section we'll discuss the
1470 subdirectories printed in bold letters in the table above. As default values
1471 are suitable for most needs, there is no need to change these values.
1473 /proc/sys/net/core - Network core options
1474 -----------------------------------------
1479 The default setting of the socket receive buffer in bytes.
1484 The maximum receive socket buffer size in bytes.
1489 The default setting (in bytes) of the socket send buffer.
1494 The maximum send socket buffer size in bytes.
1496 message_burst and message_cost
1497 ------------------------------
1499 These parameters are used to limit the warning messages written to the kernel
1500 log from the networking code. They enforce a rate limit to make a
1501 denial-of-service attack impossible. A higher message_cost factor, results in
1502 fewer messages that will be written. Message_burst controls when messages will
1503 be dropped. The default settings limit warning messages to one every five
1509 This controls console messages from the networking stack that can occur because
1510 of problems on the network like duplicate address or bad checksums. Normally,
1511 this should be enabled, but if the problem persists the messages can be
1518 Maximum number of packets, queued on the INPUT side, when the interface
1519 receives packets faster than kernel can process them.
1524 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1525 of struct cmsghdr structures with appended data.
1527 /proc/sys/net/unix - Parameters for Unix domain sockets
1528 -------------------------------------------------------
1530 There are only two files in this subdirectory. They control the delays for
1531 deleting and destroying socket descriptors.
1533 2.8 /proc/sys/net/ipv4 - IPV4 settings
1534 --------------------------------------
1536 IP version 4 is still the most used protocol in Unix networking. It will be
1537 replaced by IP version 6 in the next couple of years, but for the moment it's
1538 the de facto standard for the internet and is used in most networking
1539 environments around the world. Because of the importance of this protocol,
1540 we'll have a deeper look into the subtree controlling the behavior of the IPv4
1541 subsystem of the Linux kernel.
1543 Let's start with the entries in /proc/sys/net/ipv4.
1548 icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1549 ----------------------------------------------------
1551 Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1552 just those to broadcast and multicast addresses.
1554 Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1555 destination address your network may be used as an exploder for denial of
1556 service packet flooding attacks to other hosts.
1558 icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1559 ---------------------------------------------------------------------------------------
1561 Sets limits for sending ICMP packets to specific targets. A value of zero
1562 disables all limiting. Any positive value sets the maximum package rate in
1563 hundredth of a second (on Intel systems).
1571 This file contains the number one if the host received its IP configuration by
1572 RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1577 TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1578 hops a packet may travel.
1583 Enable dynamic socket address rewriting on interface address change. This is
1584 useful for dialup interface with changing IP addresses.
1589 Enable or disable forwarding of IP packages between interfaces. Changing this
1590 value resets all other parameters to their default values. They differ if the
1591 kernel is configured as host or router.
1596 Range of ports used by TCP and UDP to choose the local port. Contains two
1597 numbers, the first number is the lowest port, the second number the highest
1598 local port. Default is 1024-4999. Should be changed to 32768-61000 for
1604 Global switch to turn path MTU discovery off. It can also be set on a per
1605 socket basis by the applications or on a per route basis.
1610 Enable/disable debugging of IP masquerading.
1612 IP fragmentation settings
1613 -------------------------
1615 ipfrag_high_trash and ipfrag_low_trash
1616 --------------------------------------
1618 Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1619 of memory is allocated for this purpose, the fragment handler will toss
1620 packets until ipfrag_low_thresh is reached.
1625 Time in seconds to keep an IP fragment in memory.
1633 This file controls the use of the ECN bit in the IPv4 headers. This is a new
1634 feature about Explicit Congestion Notification, but some routers and firewalls
1635 block traffic that has this bit set, so it could be necessary to echo 0 to
1636 /proc/sys/net/ipv4/tcp_ecn if you want to talk to these sites. For more info
1637 you could read RFC2481.
1639 tcp_retrans_collapse
1640 --------------------
1642 Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1643 larger packets to work around bugs in certain TCP stacks. Can be turned off by
1646 tcp_keepalive_probes
1647 --------------------
1649 Number of keep alive probes TCP sends out, until it decides that the
1650 connection is broken.
1655 How often TCP sends out keep alive messages, when keep alive is enabled. The
1661 Number of times initial SYNs for a TCP connection attempt will be
1662 retransmitted. Should not be higher than 255. This is only the timeout for
1663 outgoing connections, for incoming connections the number of retransmits is
1664 defined by tcp_retries1.
1669 Enable select acknowledgments after RFC2018.
1674 Enable timestamps as defined in RFC1323.
1679 Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1680 default is to use the BSD compatible interpretation of the urgent pointer
1681 pointing to the first byte after the urgent data. The RFC793 interpretation is
1682 to have it point to the last byte of urgent data. Enabling this option may
1683 lead to interoperability problems. Disabled by default.
1688 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1689 syncookies when the syn backlog queue of a socket overflows. This is to ward
1690 off the common 'syn flood attack'. Disabled by default.
1692 Note that the concept of a socket backlog is abandoned. This means the peer
1693 may not receive reliable error messages from an over loaded server with
1699 Enable window scaling as defined in RFC1323.
1704 The length of time in seconds it takes to receive a final FIN before the
1705 socket is always closed. This is strictly a violation of the TCP
1706 specification, but required to prevent denial-of-service attacks.
1711 Indicates how many keep alive probes are sent per slow timer run. Should not
1712 be set too high to prevent bursts.
1717 Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1718 in listen(2) only specifies the length of the backlog queue of already
1719 established sockets. When more connection requests arrive Linux starts to drop
1720 packets. When syncookies are enabled the packets are still answered and the
1721 maximum queue is effectively ignored.
1726 Defines how often an answer to a TCP connection request is retransmitted
1732 Defines how often a TCP packet is retransmitted before giving up.
1734 Interface specific settings
1735 ---------------------------
1737 In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1738 interface the system knows about and one directory calls all. Changes in the
1739 all subdirectory affect all interfaces, whereas changes in the other
1740 subdirectories affect only one interface. All directories have the same
1746 This switch decides if the kernel accepts ICMP redirect messages or not. The
1747 default is 'yes' if the kernel is configured for a regular host and 'no' for a
1748 router configuration.
1753 Should source routed packages be accepted or declined. The default is
1754 dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1760 Accept packets with source address 0.b.c.d with destinations not to this host
1761 as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1764 The default is 0, since this feature is not implemented yet (kernel version
1770 Enable or disable IP forwarding on this interface.
1775 Log packets with source addresses with no known route to kernel log.
1780 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1781 multicast routing daemon is required.
1786 Does (1) or does not (0) perform proxy ARP.
1791 Integer value determines if a source validation should be made. 1 means yes, 0
1792 means no. Disabled by default, but local/broadcast address spoofing is always
1795 If you set this to 1 on a router that is the only connection for a network to
1796 the net, it will prevent spoofing attacks against your internal networks
1797 (external addresses can still be spoofed), without the need for additional
1803 Accept ICMP redirect messages only for gateways, listed in default gateway
1804 list. Enabled by default.
1809 If it is not set the kernel does not assume that different subnets on this
1810 device can communicate directly. Default setting is 'yes'.
1815 Determines whether to send ICMP redirects to other hosts.
1820 The directory /proc/sys/net/ipv4/route contains several file to control
1823 error_burst and error_cost
1824 --------------------------
1826 These parameters are used to limit how many ICMP destination unreachable to
1827 send from the host in question. ICMP destination unreachable messages are
1828 sent when we cannot reach the next hop while trying to transmit a packet.
1829 It will also print some error messages to kernel logs if someone is ignoring
1830 our ICMP redirects. The higher the error_cost factor is, the fewer
1831 destination unreachable and error messages will be let through. Error_burst
1832 controls when destination unreachable messages and error messages will be
1833 dropped. The default settings limit warning messages to five every second.
1838 Writing to this file results in a flush of the routing cache.
1840 gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
1841 ---------------------------------------------------------------------
1843 Values to control the frequency and behavior of the garbage collection
1844 algorithm for the routing cache. gc_min_interval is deprecated and replaced
1845 by gc_min_interval_ms.
1851 Maximum size of the routing cache. Old entries will be purged once the cache
1852 reached has this size.
1854 max_delay, min_delay
1855 --------------------
1857 Delays for flushing the routing cache.
1859 redirect_load, redirect_number
1860 ------------------------------
1862 Factors which determine if more ICPM redirects should be sent to a specific
1863 host. No redirects will be sent once the load limit or the maximum number of
1864 redirects has been reached.
1869 Timeout for redirects. After this period redirects will be sent again, even if
1870 this has been stopped, because the load or number limit has been reached.
1872 Network Neighbor handling
1873 -------------------------
1875 Settings about how to handle connections with direct neighbors (nodes attached
1876 to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
1878 As we saw it in the conf directory, there is a default subdirectory which
1879 holds the default values, and one directory for each interface. The contents
1880 of the directories are identical, with the single exception that the default
1881 settings contain additional options to set garbage collection parameters.
1883 In the interface directories you'll find the following entries:
1885 base_reachable_time, base_reachable_time_ms
1886 -------------------------------------------
1888 A base value used for computing the random reachable time value as specified
1891 Expression of base_reachable_time, which is deprecated, is in seconds.
1892 Expression of base_reachable_time_ms is in milliseconds.
1894 retrans_time, retrans_time_ms
1895 -----------------------------
1897 The time between retransmitted Neighbor Solicitation messages.
1898 Used for address resolution and to determine if a neighbor is
1901 Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
1902 IPv4) or in jiffies (for IPv6).
1903 Expression of retrans_time_ms is in milliseconds.
1908 Maximum queue length for a pending arp request - the number of packets which
1909 are accepted from other layers while the ARP address is still resolved.
1914 Maximum for random delay of answers to neighbor solicitation messages in
1915 jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
1921 Maximum number of retries for unicast solicitation.
1926 Maximum number of retries for multicast solicitation.
1928 delay_first_probe_time
1929 ----------------------
1931 Delay for the first time probe if the neighbor is reachable. (see
1937 An ARP/neighbor entry is only replaced with a new one if the old is at least
1938 locktime old. This prevents ARP cache thrashing.
1943 Maximum time (real time is random [0..proxytime]) before answering to an ARP
1944 request for which we have an proxy ARP entry. In some cases, this is used to
1945 prevent network flooding.
1950 Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
1955 Determines the number of requests to send to the user level ARP daemon. Use 0
1961 Determines how often to check for stale ARP entries. After an ARP entry is
1962 stale it will be resolved again (which is useful when an IP address migrates
1963 to another machine). When ucast_solicit is greater than 0 it first tries to
1964 send an ARP packet directly to the known host When that fails and
1965 mcast_solicit is greater than 0, an ARP request is broadcasted.
1970 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
1971 when Appletalk is loaded. The configurable parameters are:
1976 The amount of time we keep an ARP entry before expiring it. Used to age out
1982 The amount of time we will spend trying to resolve an Appletalk address.
1984 aarp-retransmit-limit
1985 ---------------------
1987 The number of times we will retransmit a query before giving up.
1992 Controls the rate at which expires are checked.
1994 The directory /proc/net/appletalk holds the list of active Appletalk sockets
1997 The fields indicate the DDP type, the local address (in network:node format)
1998 the remote address, the size of the transmit pending queue, the size of the
1999 received queue (bytes waiting for applications to read) the state and the uid
2002 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
2003 shows the name of the interface, its Appletalk address, the network range on
2004 that address (or network number for phase 1 networks), and the status of the
2007 /proc/net/atalk_route lists each known network route. It lists the target
2008 (network) that the route leads to, the router (may be directly connected), the
2009 route flags, and the device the route is using.
2014 The IPX protocol has no tunable values in proc/sys/net.
2016 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
2017 socket giving the local and remote addresses in Novell format (that is
2018 network:node:port). In accordance with the strange Novell tradition,
2019 everything but the port is in hex. Not_Connected is displayed for sockets that
2020 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
2021 the number of bytes pending for transmission and reception. The state
2022 indicates the state the socket is in and the uid is the owning uid of the
2025 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
2026 it gives the network number, the node number, and indicates if the network is
2027 the primary network. It also indicates which device it is bound to (or
2028 Internal for internal networks) and the Frame Type if appropriate. Linux
2029 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
2032 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
2033 gives the destination network, the router node (or Directly) and the network
2034 address of the router (or Connected) for internal networks.
2036 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
2037 ----------------------------------------------------------
2039 The "mqueue" filesystem provides the necessary kernel features to enable the
2040 creation of a user space library that implements the POSIX message queues
2041 API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
2042 Interfaces specification.)
2044 The "mqueue" filesystem contains values for determining/setting the amount of
2045 resources used by the file system.
2047 /proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
2048 maximum number of message queues allowed on the system.
2050 /proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
2051 maximum number of messages in a queue value. In fact it is the limiting value
2052 for another (user) limit which is set in mq_open invocation. This attribute of
2053 a queue must be less or equal then msg_max.
2055 /proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
2056 maximum message size value (it is every message queue's attribute set during
2059 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
2060 ------------------------------------------------------
2062 This file can be used to adjust the score used to select which processes
2063 should be killed in an out-of-memory situation. Giving it a high score will
2064 increase the likelihood of this process being killed by the oom-killer. Valid
2065 values are in the range -16 to +15, plus the special value -17, which disables
2066 oom-killing altogether for this process.
2068 2.13 /proc/<pid>/oom_score - Display current oom-killer score
2069 -------------------------------------------------------------
2071 ------------------------------------------------------------------------------
2072 This file can be used to check the current score used by the oom-killer is for
2073 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
2074 process should be killed in an out-of-memory situation.
2076 ------------------------------------------------------------------------------
2078 ------------------------------------------------------------------------------
2079 Certain aspects of kernel behavior can be modified at runtime, without the
2080 need to recompile the kernel, or even to reboot the system. The files in the
2081 /proc/sys tree can not only be read, but also modified. You can use the echo
2082 command to write value into these files, thereby changing the default settings
2084 ------------------------------------------------------------------------------
2086 2.14 /proc/<pid>/io - Display the IO accounting fields
2087 -------------------------------------------------------
2089 This file contains IO statistics for each running process
2094 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
2097 test:/tmp # cat /proc/3828/io
2103 write_bytes: 323932160
2104 cancelled_write_bytes: 0
2113 I/O counter: chars read
2114 The number of bytes which this task has caused to be read from storage. This
2115 is simply the sum of bytes which this process passed to read() and pread().
2116 It includes things like tty IO and it is unaffected by whether or not actual
2117 physical disk IO was required (the read might have been satisfied from
2124 I/O counter: chars written
2125 The number of bytes which this task has caused, or shall cause to be written
2126 to disk. Similar caveats apply here as with rchar.
2132 I/O counter: read syscalls
2133 Attempt to count the number of read I/O operations, i.e. syscalls like read()
2140 I/O counter: write syscalls
2141 Attempt to count the number of write I/O operations, i.e. syscalls like
2142 write() and pwrite().
2148 I/O counter: bytes read
2149 Attempt to count the number of bytes which this process really did cause to
2150 be fetched from the storage layer. Done at the submit_bio() level, so it is
2151 accurate for block-backed filesystems. <please add status regarding NFS and
2152 CIFS at a later time>
2158 I/O counter: bytes written
2159 Attempt to count the number of bytes which this process caused to be sent to
2160 the storage layer. This is done at page-dirtying time.
2163 cancelled_write_bytes
2164 ---------------------
2166 The big inaccuracy here is truncate. If a process writes 1MB to a file and
2167 then deletes the file, it will in fact perform no writeout. But it will have
2168 been accounted as having caused 1MB of write.
2169 In other words: The number of bytes which this process caused to not happen,
2170 by truncating pagecache. A task can cause "negative" IO too. If this task
2171 truncates some dirty pagecache, some IO which another task has been accounted
2172 for (in it's write_bytes) will not be happening. We _could_ just subtract that
2173 from the truncating task's write_bytes, but there is information loss in doing
2180 At its current implementation state, this is a bit racy on 32-bit machines: if
2181 process A reads process B's /proc/pid/io while process B is updating one of
2182 those 64-bit counters, process A could see an intermediate result.
2185 More information about this can be found within the taskstats documentation in
2186 Documentation/accounting.
2188 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
2189 ---------------------------------------------------------------
2190 When a process is dumped, all anonymous memory is written to a core file as
2191 long as the size of the core file isn't limited. But sometimes we don't want
2192 to dump some memory segments, for example, huge shared memory. Conversely,
2193 sometimes we want to save file-backed memory segments into a core file, not
2194 only the individual files.
2196 /proc/<pid>/coredump_filter allows you to customize which memory segments
2197 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
2198 of memory types. If a bit of the bitmask is set, memory segments of the
2199 corresponding memory type are dumped, otherwise they are not dumped.
2201 The following 4 memory types are supported:
2202 - (bit 0) anonymous private memory
2203 - (bit 1) anonymous shared memory
2204 - (bit 2) file-backed private memory
2205 - (bit 3) file-backed shared memory
2207 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
2208 are always dumped regardless of the bitmask status.
2210 Default value of coredump_filter is 0x3; this means all anonymous memory
2211 segments are dumped.
2213 If you don't want to dump all shared memory segments attached to pid 1234,
2214 write 1 to the process's proc file.
2216 $ echo 0x1 > /proc/1234/coredump_filter
2218 When a new process is created, the process inherits the bitmask status from its
2219 parent. It is useful to set up coredump_filter before the program runs.
2222 $ echo 0x7 > /proc/self/coredump_filter
2225 ------------------------------------------------------------------------------
projects / linux-2.6 / blob