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
43 ------------------------------------------------------------------------------
45 ------------------------------------------------------------------------------
47 0.1 Introduction/Credits
48 ------------------------
50 This documentation is part of a soon (or so we hope) to be released book on
51 the SuSE Linux distribution. As there is no complete documentation for the
52 /proc file system and we've used many freely available sources to write these
53 chapters, it seems only fair to give the work back to the Linux community.
54 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
55 afraid it's still far from complete, but we hope it will be useful. As far as
56 we know, it is the first 'all-in-one' document about the /proc file system. It
57 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
58 SPARC, AXP, etc., features, you probably won't find what you are looking for.
59 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
60 additions and patches are welcome and will be added to this document if you
63 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
64 other people for help compiling this documentation. We'd also like to extend a
65 special thank you to Andi Kleen for documentation, which we relied on heavily
66 to create this document, as well as the additional information he provided.
67 Thanks to everybody else who contributed source or docs to the Linux kernel
68 and helped create a great piece of software... :)
70 If you have any comments, corrections or additions, please don't hesitate to
71 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
74 The latest version of this document is available online at
75 http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
77 If the above direction does not works for you, ypu could try the kernel
78 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
79 comandante@zaralinux.com.
84 We don't guarantee the correctness of this document, and if you come to us
85 complaining about how you screwed up your system because of incorrect
86 documentation, we won't feel responsible...
88 ------------------------------------------------------------------------------
89 CHAPTER 1: COLLECTING SYSTEM INFORMATION
90 ------------------------------------------------------------------------------
92 ------------------------------------------------------------------------------
94 ------------------------------------------------------------------------------
95 * Investigating the properties of the pseudo file system /proc and its
96 ability to provide information on the running Linux system
97 * Examining /proc's structure
98 * Uncovering various information about the kernel and the processes running
100 ------------------------------------------------------------------------------
103 The proc file system acts as an interface to internal data structures in the
104 kernel. It can be used to obtain information about the system and to change
105 certain kernel parameters at runtime (sysctl).
107 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
108 show you how you can use /proc/sys to change settings.
110 1.1 Process-Specific Subdirectories
111 -----------------------------------
113 The directory /proc contains (among other things) one subdirectory for each
114 process running on the system, which is named after the process ID (PID).
116 The link self points to the process reading the file system. Each process
117 subdirectory has the entries listed in Table 1-1.
120 Table 1-1: Process specific entries in /proc
121 ..............................................................................
123 cmdline Command line arguments
124 cpu Current and last cpu in wich it was executed (2.4)(smp)
125 cwd Link to the current working directory
126 environ Values of environment variables
127 exe Link to the executable of this process
128 fd Directory, which contains all file descriptors
129 maps Memory maps to executables and library files (2.4)
130 mem Memory held by this process
131 root Link to the root directory of this process
133 statm Process memory status information
134 status Process status in human readable form
135 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
136 ..............................................................................
138 For example, to get the status information of a process, all you have to do is
139 read the file /proc/PID/status:
141 >cat /proc/self/status
157 SigPnd: 0000000000000000
158 SigBlk: 0000000000000000
159 SigIgn: 0000000000000000
160 SigCgt: 0000000000000000
161 CapInh: 00000000fffffeff
162 CapPrm: 0000000000000000
163 CapEff: 0000000000000000
166 This shows you nearly the same information you would get if you viewed it with
167 the ps command. In fact, ps uses the proc file system to obtain its
168 information. The statm file contains more detailed information about the
169 process memory usage. Its seven fields are explained in Table 1-2.
172 Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
173 ..............................................................................
175 size total program size (pages) (same as VmSize in status)
176 resident size of memory portions (pages) (same as VmRSS in status)
177 shared number of pages that are shared (i.e. backed by a file)
178 trs number of pages that are 'code' (not including libs; broken,
179 includes data segment)
180 lrs number of pages of library (always 0 on 2.6)
181 drs number of pages of data/stack (including libs; broken,
182 includes library text)
183 dt number of dirty pages (always 0 on 2.6)
184 ..............................................................................
189 Similar to the process entries, the kernel data files give information about
190 the running kernel. The files used to obtain this information are contained in
191 /proc and are listed in Table 1-3. Not all of these will be present in your
192 system. It depends on the kernel configuration and the loaded modules, which
193 files are there, and which are missing.
195 Table 1-3: Kernel info in /proc
196 ..............................................................................
198 apm Advanced power management info
199 buddyinfo Kernel memory allocator information (see text) (2.5)
200 bus Directory containing bus specific information
201 cmdline Kernel command line
202 cpuinfo Info about the CPU
203 devices Available devices (block and character)
204 dma Used DMS channels
205 filesystems Supported filesystems
206 driver Various drivers grouped here, currently rtc (2.4)
207 execdomains Execdomains, related to security (2.4)
208 fb Frame Buffer devices (2.4)
209 fs File system parameters, currently nfs/exports (2.4)
210 ide Directory containing info about the IDE subsystem
211 interrupts Interrupt usage
212 iomem Memory map (2.4)
213 ioports I/O port usage
214 irq Masks for irq to cpu affinity (2.4)(smp?)
215 isapnp ISA PnP (Plug&Play) Info (2.4)
216 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
218 ksyms Kernel symbol table
219 loadavg Load average of last 1, 5 & 15 minutes
223 modules List of loaded modules
224 mounts Mounted filesystems
225 net Networking info (see text)
226 partitions Table of partitions known to the system
227 pci Depreciated info of PCI bus (new way -> /proc/bus/pci/,
228 decoupled by lspci (2.4)
230 scsi SCSI info (see text)
231 slabinfo Slab pool info
232 stat Overall statistics
233 swaps Swap space utilization
235 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
236 tty Info of tty drivers
238 version Kernel version
239 video bttv info of video resources (2.4)
240 ..............................................................................
242 You can, for example, check which interrupts are currently in use and what
243 they are used for by looking in the file /proc/interrupts:
245 > cat /proc/interrupts
247 0: 8728810 XT-PIC timer
248 1: 895 XT-PIC keyboard
250 3: 531695 XT-PIC aha152x
251 4: 2014133 XT-PIC serial
252 5: 44401 XT-PIC pcnet_cs
255 12: 182918 XT-PIC PS/2 Mouse
257 14: 1232265 XT-PIC ide0
261 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
262 output of a SMP machine):
264 > cat /proc/interrupts
267 0: 1243498 1214548 IO-APIC-edge timer
268 1: 8949 8958 IO-APIC-edge keyboard
269 2: 0 0 XT-PIC cascade
270 5: 11286 10161 IO-APIC-edge soundblaster
271 8: 1 0 IO-APIC-edge rtc
272 9: 27422 27407 IO-APIC-edge 3c503
273 12: 113645 113873 IO-APIC-edge PS/2 Mouse
275 14: 22491 24012 IO-APIC-edge ide0
276 15: 2183 2415 IO-APIC-edge ide1
277 17: 30564 30414 IO-APIC-level eth0
278 18: 177 164 IO-APIC-level bttv
283 NMI is incremented in this case because every timer interrupt generates a NMI
284 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
286 LOC is the local interrupt counter of the internal APIC of every CPU.
288 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
289 connects the CPUs in a SMP system. This means that an error has been detected,
290 the IO-APIC automatically retry the transmission, so it should not be a big
291 problem, but you should read the SMP-FAQ.
293 In this context it could be interesting to note the new irq directory in 2.4.
294 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
295 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
296 irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
300 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
301 1 11 13 15 17 19 3 5 7 9
305 The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
306 is the same by default:
308 > cat /proc/irq/0/smp_affinity
311 It's a bitmask, in wich you can specify wich CPUs can handle the IRQ, you can
314 > echo 1 > /proc/irq/prof_cpu_mask
316 This means that only the first CPU will handle the IRQ, but you can also echo 5
317 wich means that only the first and fourth CPU can handle the IRQ.
319 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
320 between all the CPUs which are allowed to handle it. As usual the kernel has
321 more info than you and does a better job than you, so the defaults are the
322 best choice for almost everyone.
324 There are three more important subdirectories in /proc: net, scsi, and sys.
325 The general rule is that the contents, or even the existence of these
326 directories, depend on your kernel configuration. If SCSI is not enabled, the
327 directory scsi may not exist. The same is true with the net, which is there
328 only when networking support is present in the running kernel.
330 The slabinfo file gives information about memory usage at the slab level.
331 Linux uses slab pools for memory management above page level in version 2.2.
332 Commonly used objects have their own slab pool (such as network buffers,
333 directory cache, and so on).
335 ..............................................................................
337 > cat /proc/buddyinfo
339 Node 0, zone DMA 0 4 5 4 4 3 ...
340 Node 0, zone Normal 1 0 0 1 101 8 ...
341 Node 0, zone HighMem 2 0 0 1 1 0 ...
343 Memory fragmentation is a problem under some workloads, and buddyinfo is a
344 useful tool for helping diagnose these problems. Buddyinfo will give you a
345 clue as to how big an area you can safely allocate, or why a previous
348 Each column represents the number of pages of a certain order which are
349 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
350 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
351 available in ZONE_NORMAL, etc...
354 1.3 IDE devices in /proc/ide
355 ----------------------------
357 The subdirectory /proc/ide contains information about all IDE devices of which
358 the kernel is aware. There is one subdirectory for each IDE controller, the
359 file drivers and a link for each IDE device, pointing to the device directory
360 in the controller specific subtree.
362 The file drivers contains general information about the drivers used for the
365 > cat /proc/ide/drivers
366 ide-cdrom version 4.53
367 ide-disk version 1.08
369 ..............................................................................
373 Provides information about distribution and utilization of memory. This
374 varies by architecture and compile options. The following is from a
375 16GB PIII, which has highmem enabled. You may not have all of these fields.
380 MemTotal: 16344972 kB
387 HighTotal: 15597528 kB
388 HighFree: 13629632 kB
397 Committed_AS: 1576424 kB
400 VmallocTotal: 112216 kB
402 VmallocChunk: 111088 kB
404 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
405 bits and the kernel binary code)
406 MemFree: The sum of LowFree+HighFree
407 Buffers: Relatively temporary storage for raw disk blocks
408 shouldn't get tremendously large (20MB or so)
409 Cached: in-memory cache for files read from the disk (the
410 pagecache). Doesn't include SwapCached
411 SwapCached: Memory that once was swapped out, is swapped back in but
412 still also is in the swapfile (if memory is needed it
413 doesn't need to be swapped out AGAIN because it is already
414 in the swapfile. This saves I/O)
415 Active: Memory that has been used more recently and usually not
416 reclaimed unless absolutely necessary.
417 Inactive: Memory which has been less recently used. It is more
418 eligible to be reclaimed for other purposes
420 HighFree: Highmem is all memory above ~860MB of physical memory
421 Highmem areas are for use by userspace programs, or
422 for the pagecache. The kernel must use tricks to access
423 this memory, making it slower to access than lowmem.
425 LowFree: Lowmem is memory which can be used for everything that
426 highmem can be used for, but it is also availble for the
427 kernel's use for its own data structures. Among many
428 other things, it is where everything from the Slab is
429 allocated. Bad things happen when you're out of lowmem.
430 SwapTotal: total amount of swap space available
431 SwapFree: Memory which has been evicted from RAM, and is temporarily
433 Dirty: Memory which is waiting to get written back to the disk
434 Writeback: Memory which is actively being written back to the disk
435 Mapped: files which have been mmaped, such as libraries
436 Slab: in-kernel data structures cache
437 Committed_AS: An estimate of how much RAM you would need to make a
438 99.99% guarantee that there never is OOM (out of memory)
439 for this workload. Normally the kernel will overcommit
440 memory. That means, say you do a 1GB malloc, nothing
441 happens, really. Only when you start USING that malloc
442 memory you will get real memory on demand, and just as
443 much as you use. So you sort of take a mortgage and hope
444 the bank doesn't go bust. Other cases might include when
445 you mmap a file that's shared only when you write to it
446 and you get a private copy of that data. While it normally
447 is shared between processes. The Committed_AS is a
448 guesstimate of how much RAM/swap you would need
450 PageTables: amount of memory dedicated to the lowest level of page
452 ReverseMaps: number of reverse mappings performed
453 VmallocTotal: total size of vmalloc memory area
454 VmallocUsed: amount of vmalloc area which is used
455 VmallocChunk: largest contigious block of vmalloc area which is free
457 More detailed information can be found in the controller specific
458 subdirectories. These are named ide0, ide1 and so on. Each of these
459 directories contains the files shown in table 1-4.
462 Table 1-4: IDE controller info in /proc/ide/ide?
463 ..............................................................................
465 channel IDE channel (0 or 1)
466 config Configuration (only for PCI/IDE bridge)
468 model Type/Chipset of IDE controller
469 ..............................................................................
471 Each device connected to a controller has a separate subdirectory in the
472 controllers directory. The files listed in table 1-5 are contained in these
476 Table 1-5: IDE device information
477 ..............................................................................
480 capacity Capacity of the medium (in 512Byte blocks)
481 driver driver and version
482 geometry physical and logical geometry
483 identify device identify block
485 model device identifier
486 settings device setup
487 smart_thresholds IDE disk management thresholds
488 smart_values IDE disk management values
489 ..............................................................................
491 The most interesting file is settings. This file contains a nice overview of
492 the drive parameters:
494 # cat /proc/ide/ide0/hda/settings
495 name value min max mode
496 ---- ----- --- --- ----
497 bios_cyl 526 0 65535 rw
498 bios_head 255 0 255 rw
500 breada_readahead 4 0 127 rw
502 file_readahead 72 0 2097151 rw
504 keepsettings 0 0 1 rw
505 max_kb_per_request 122 1 127 rw
509 pio_mode write-only 0 255 w
515 1.4 Networking info in /proc/net
516 --------------------------------
518 The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
519 additional values you get for IP version 6 if you configure the kernel to
520 support this. Table 1-7 lists the files and their meaning.
523 Table 1-6: IPv6 info in /proc/net
524 ..............................................................................
526 udp6 UDP sockets (IPv6)
527 tcp6 TCP sockets (IPv6)
528 raw6 Raw device statistics (IPv6)
529 igmp6 IP multicast addresses, which this host joined (IPv6)
530 if_inet6 List of IPv6 interface addresses
531 ipv6_route Kernel routing table for IPv6
532 rt6_stats Global IPv6 routing tables statistics
533 sockstat6 Socket statistics (IPv6)
534 snmp6 Snmp data (IPv6)
535 ..............................................................................
538 Table 1-7: Network info in /proc/net
539 ..............................................................................
542 dev network devices with statistics
543 dev_mcast the Layer2 multicast groups a device is listening too
544 (interface index, label, number of references, number of bound
546 dev_stat network device status
547 ip_fwchains Firewall chain linkage
548 ip_fwnames Firewall chain names
549 ip_masq Directory containing the masquerading tables
550 ip_masquerade Major masquerading table
551 netstat Network statistics
552 raw raw device statistics
553 route Kernel routing table
554 rpc Directory containing rpc info
555 rt_cache Routing cache
557 sockstat Socket statistics
559 tr_rif Token ring RIF routing table
561 unix UNIX domain sockets
562 wireless Wireless interface data (Wavelan etc)
563 igmp IP multicast addresses, which this host joined
564 psched Global packet scheduler parameters.
565 netlink List of PF_NETLINK sockets
566 ip_mr_vifs List of multicast virtual interfaces
567 ip_mr_cache List of multicast routing cache
568 ..............................................................................
570 You can use this information to see which network devices are available in
571 your system and how much traffic was routed over those devices:
575 face |bytes packets errs drop fifo frame compressed multicast|[...
576 lo: 908188 5596 0 0 0 0 0 0 [...
577 ppp0:15475140 20721 410 0 0 410 0 0 [...
578 eth0: 614530 7085 0 0 0 0 0 1 [...
581 ...] bytes packets errs drop fifo colls carrier compressed
582 ...] 908188 5596 0 0 0 0 0 0
583 ...] 1375103 17405 0 0 0 0 0 0
584 ...] 1703981 5535 0 0 0 3 0 0
586 In addition, each Channel Bond interface has it's own directory. For
587 example, the bond0 device will have a directory called /proc/net/bond0/.
588 It will contain information that is specific to that bond, such as the
589 current slaves of the bond, the link status of the slaves, and how
590 many times the slaves link has failed.
595 If you have a SCSI host adapter in your system, you'll find a subdirectory
596 named after the driver for this adapter in /proc/scsi. You'll also see a list
597 of all recognized SCSI devices in /proc/scsi:
601 Host: scsi0 Channel: 00 Id: 00 Lun: 00
602 Vendor: IBM Model: DGHS09U Rev: 03E0
603 Type: Direct-Access ANSI SCSI revision: 03
604 Host: scsi0 Channel: 00 Id: 06 Lun: 00
605 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
606 Type: CD-ROM ANSI SCSI revision: 02
609 The directory named after the driver has one file for each adapter found in
610 the system. These files contain information about the controller, including
611 the used IRQ and the IO address range. The amount of information shown is
612 dependent on the adapter you use. The example shows the output for an Adaptec
613 AHA-2940 SCSI adapter:
615 > cat /proc/scsi/aic7xxx/0
617 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
619 TCQ Enabled By Default : Disabled
620 AIC7XXX_PROC_STATS : Disabled
621 AIC7XXX_RESET_DELAY : 5
622 Adapter Configuration:
623 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
624 Ultra Wide Controller
625 PCI MMAPed I/O Base: 0xeb001000
626 Adapter SEEPROM Config: SEEPROM found and used.
627 Adaptec SCSI BIOS: Enabled
629 SCBs: Active 0, Max Active 2,
630 Allocated 15, HW 16, Page 255
632 BIOS Control Word: 0x18b6
633 Adapter Control Word: 0x005b
634 Extended Translation: Enabled
635 Disconnect Enable Flags: 0xffff
636 Ultra Enable Flags: 0x0001
637 Tag Queue Enable Flags: 0x0000
638 Ordered Queue Tag Flags: 0x0000
639 Default Tag Queue Depth: 8
640 Tagged Queue By Device array for aic7xxx host instance 0:
641 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
642 Actual queue depth per device for aic7xxx host instance 0:
643 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
646 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
647 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
648 Total transfers 160151 (74577 reads and 85574 writes)
650 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
651 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
652 Total transfers 0 (0 reads and 0 writes)
655 1.6 Parallel port info in /proc/parport
656 ---------------------------------------
658 The directory /proc/parport contains information about the parallel ports of
659 your system. It has one subdirectory for each port, named after the port
662 These directories contain the four files shown in Table 1-8.
665 Table 1-8: Files in /proc/parport
666 ..............................................................................
668 autoprobe Any IEEE-1284 device ID information that has been acquired.
669 devices list of the device drivers using that port. A + will appear by the
670 name of the device currently using the port (it might not appear
672 hardware Parallel port's base address, IRQ line and DMA channel.
673 irq IRQ that parport is using for that port. This is in a separate
674 file to allow you to alter it by writing a new value in (IRQ
676 ..............................................................................
678 1.7 TTY info in /proc/tty
679 -------------------------
681 Information about the available and actually used tty's can be found in the
682 directory /proc/tty.You'll find entries for drivers and line disciplines in
683 this directory, as shown in Table 1-9.
686 Table 1-9: Files in /proc/tty
687 ..............................................................................
689 drivers list of drivers and their usage
690 ldiscs registered line disciplines
691 driver/serial usage statistic and status of single tty lines
692 ..............................................................................
694 To see which tty's are currently in use, you can simply look into the file
697 > cat /proc/tty/drivers
698 pty_slave /dev/pts 136 0-255 pty:slave
699 pty_master /dev/ptm 128 0-255 pty:master
700 pty_slave /dev/ttyp 3 0-255 pty:slave
701 pty_master /dev/pty 2 0-255 pty:master
702 serial /dev/cua 5 64-67 serial:callout
703 serial /dev/ttyS 4 64-67 serial
704 /dev/tty0 /dev/tty0 4 0 system:vtmaster
705 /dev/ptmx /dev/ptmx 5 2 system
706 /dev/console /dev/console 5 1 system:console
707 /dev/tty /dev/tty 5 0 system:/dev/tty
708 unknown /dev/tty 4 1-63 console
711 1.8 Miscellaneous kernel statistics in /proc/stat
712 -------------------------------------------------
714 Various pieces of information about kernel activity are available in the
715 /proc/stat file. All of the numbers reported in this file are aggregates
716 since the system first booted. For a quick look, simply cat the file:
719 cpu 2255 34 2290 22625563 6290 127 456
720 cpu0 1132 34 1441 11311718 3675 127 438
721 cpu1 1123 0 849 11313845 2614 0 18
722 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
729 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
730 lines. These numbers identify the amount of time the CPU has spent performing
731 different kinds of work. Time units are in USER_HZ (typically hundredths of a
732 second). The meanings of the columns are as follows, from left to right:
734 - user: normal processes executing in user mode
735 - nice: niced processes executing in user mode
736 - system: processes executing in kernel mode
737 - idle: twiddling thumbs
738 - iowait: waiting for I/O to complete
739 - irq: servicing interrupts
740 - softirq: servicing softirqs
742 The "intr" line gives counts of interrupts serviced since boot time, for each
743 of the possible system interrupts. The first column is the total of all
744 interrupts serviced; each subsequent column is the total for that particular
747 The "ctxt" line gives the total number of context switches across all CPUs.
749 The "btime" line gives the time at which the system booted, in seconds since
752 The "processes" line gives the number of processes and threads created, which
753 includes (but is not limited to) those created by calls to the fork() and
754 clone() system calls.
756 The "procs_running" line gives the number of processes currently running on
759 The "procs_blocked" line gives the number of processes currently blocked,
760 waiting for I/O to complete.
763 ------------------------------------------------------------------------------
765 ------------------------------------------------------------------------------
766 The /proc file system serves information about the running system. It not only
767 allows access to process data but also allows you to request the kernel status
768 by reading files in the hierarchy.
770 The directory structure of /proc reflects the types of information and makes
771 it easy, if not obvious, where to look for specific data.
772 ------------------------------------------------------------------------------
774 ------------------------------------------------------------------------------
775 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
776 ------------------------------------------------------------------------------
778 ------------------------------------------------------------------------------
780 ------------------------------------------------------------------------------
781 * Modifying kernel parameters by writing into files found in /proc/sys
782 * Exploring the files which modify certain parameters
783 * Review of the /proc/sys file tree
784 ------------------------------------------------------------------------------
787 A very interesting part of /proc is the directory /proc/sys. This is not only
788 a source of information, it also allows you to change parameters within the
789 kernel. Be very careful when attempting this. You can optimize your system,
790 but you can also cause it to crash. Never alter kernel parameters on a
791 production system. Set up a development machine and test to make sure that
792 everything works the way you want it to. You may have no alternative but to
793 reboot the machine once an error has been made.
795 To change a value, simply echo the new value into the file. An example is
796 given below in the section on the file system data. You need to be root to do
797 this. You can create your own boot script to perform this every time your
800 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
801 general things in the operation of the Linux kernel. Since some of the files
802 can inadvertently disrupt your system, it is advisable to read both
803 documentation and source before actually making adjustments. In any case, be
804 very careful when writing to any of these files. The entries in /proc may
805 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
806 review the kernel documentation in the directory /usr/src/linux/Documentation.
807 This chapter is heavily based on the documentation included in the pre 2.2
808 kernels, and became part of it in version 2.2.1 of the Linux kernel.
810 2.1 /proc/sys/fs - File system data
811 -----------------------------------
813 This subdirectory contains specific file system, file handle, inode, dentry
814 and quota information.
816 Currently, these files are in /proc/sys/fs:
821 Status of the directory cache. Since directory entries are dynamically
822 allocated and deallocated, this file indicates the current status. It holds
823 six values, in which the last two are not used and are always zero. The others
824 are listed in table 2-1.
827 Table 2-1: Status files of the directory cache
828 ..............................................................................
830 nr_dentry Almost always zero
831 nr_unused Number of unused cache entries
833 in seconds after the entry may be reclaimed, when memory is short
834 want_pages internally
835 ..............................................................................
837 dquot-nr and dquot-max
838 ----------------------
840 The file dquot-max shows the maximum number of cached disk quota entries.
842 The file dquot-nr shows the number of allocated disk quota entries and the
843 number of free disk quota entries.
845 If the number of available cached disk quotas is very low and you have a large
846 number of simultaneous system users, you might want to raise the limit.
851 The kernel allocates file handles dynamically, but doesn't free them again at
854 The value in file-max denotes the maximum number of file handles that the
855 Linux kernel will allocate. When you get a lot of error messages about running
856 out of file handles, you might want to raise this limit. The default value is
857 10% of RAM in kilobytes. To change it, just write the new number into the
860 # cat /proc/sys/fs/file-max
862 # echo 8192 > /proc/sys/fs/file-max
863 # cat /proc/sys/fs/file-max
867 This method of revision is useful for all customizable parameters of the
868 kernel - simply echo the new value to the corresponding file.
870 Historically, the three values in file-nr denoted the number of allocated file
871 handles, the number of allocated but unused file handles, and the maximum
872 number of file handles. Linux 2.6 always reports 0 as the number of free file
873 handles -- this is not an error, it just means that the number of allocated
874 file handles exactly matches the number of used file handles.
876 Attempts to allocate more file descriptors than file-max are reported with
877 printk, look for "VFS: file-max limit <number> reached".
879 inode-state and inode-nr
880 ------------------------
882 The file inode-nr contains the first two items from inode-state, so we'll skip
885 inode-state contains two actual numbers and five dummy values. The numbers
886 are nr_inodes and nr_free_inodes (in order of appearance).
891 Denotes the number of inodes the system has allocated. This number will
892 grow and shrink dynamically.
897 Represents the number of free inodes. Ie. The number of inuse inodes is
898 (nr_inodes - nr_free_inodes).
900 super-nr and super-max
901 ----------------------
903 Again, super block structures are allocated by the kernel, but not freed. The
904 file super-max contains the maximum number of super block handlers, where
905 super-nr shows the number of currently allocated ones.
907 Every mounted file system needs a super block, so if you plan to mount lots of
908 file systems, you may want to increase these numbers.
910 aio-nr and aio-max-nr
911 ---------------------
913 aio-nr is the running total of the number of events specified on the
914 io_setup system call for all currently active aio contexts. If aio-nr
915 reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
916 raising aio-max-nr does not result in the pre-allocation or re-sizing
917 of any kernel data structures.
919 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
920 -----------------------------------------------------------
922 Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
923 handles the kernel support for miscellaneous binary formats.
925 Binfmt_misc provides the ability to register additional binary formats to the
926 Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
927 needs to know magic numbers at the beginning or the filename extension of the
930 It works by maintaining a linked list of structs that contain a description of
931 a binary format, including a magic with size (or the filename extension),
932 offset and mask, and the interpreter name. On request it invokes the given
933 interpreter with the original program as argument, as binfmt_java and
934 binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
935 binary-formats, you have to register an additional binary-format.
937 There are two general files in binfmt_misc and one file per registered format.
938 The two general files are register and status.
940 Registering a new binary format
941 -------------------------------
943 To register a new binary format you have to issue the command
945 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
949 with appropriate name (the name for the /proc-dir entry), offset (defaults to
950 0, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
951 last but not least, the interpreter that is to be invoked (for example and
952 testing /bin/echo). Type can be M for usual magic matching or E for filename
953 extension matching (give extension in place of magic).
955 Check or reset the status of the binary format handler
956 ------------------------------------------------------
958 If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
959 current status (enabled/disabled) of binfmt_misc. Change the status by echoing
960 0 (disables) or 1 (enables) or -1 (caution: this clears all previously
961 registered binary formats) to status. For example echo 0 > status to disable
962 binfmt_misc (temporarily).
964 Status of a single handler
965 --------------------------
967 Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
968 perform the same function as status, but their scope is limited to the actual
969 binary format. By cating this file, you also receive all related information
970 about the interpreter/magic of the binfmt.
972 Example usage of binfmt_misc (emulate binfmt_java)
973 --------------------------------------------------
975 cd /proc/sys/fs/binfmt_misc
976 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
977 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
978 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
979 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
982 These four lines add support for Java executables and Java applets (like
983 binfmt_java, additionally recognizing the .html extension with no need to put
984 <!--applet> to every applet file). You have to install the JDK and the
985 shell-script /usr/local/java/bin/javawrapper too. It works around the
986 brokenness of the Java filename handling. To add a Java binary, just create a
987 link to the class-file somewhere in the path.
989 2.3 /proc/sys/kernel - general kernel parameters
990 ------------------------------------------------
992 This directory reflects general kernel behaviors. As I've said before, the
993 contents depend on your configuration. Here you'll find the most important
994 files, along with descriptions of what they mean and how to use them.
999 The file contains three values; highwater, lowwater, and frequency.
1001 It exists only when BSD-style process accounting is enabled. These values
1002 control its behavior. If the free space on the file system where the log lives
1003 goes below lowwater percentage, accounting suspends. If it goes above
1004 highwater percentage, accounting resumes. Frequency determines how often you
1005 check the amount of free space (value is in seconds). Default settings are: 4,
1006 2, and 30. That is, suspend accounting if there is less than 2 percent free;
1007 resume it if we have a value of 3 or more percent; consider information about
1008 the amount of free space valid for 30 seconds
1013 When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1014 program to handle a graceful restart. However, when the value is greater that
1015 zero, Linux's reaction to this key combination will be an immediate reboot,
1016 without syncing its dirty buffers.
1019 When a program (like dosemu) has the keyboard in raw mode, the
1020 ctrl-alt-del is intercepted by the program before it ever reaches the
1021 kernel tty layer, and it is up to the program to decide what to do with
1024 domainname and hostname
1025 -----------------------
1027 These files can be controlled to set the NIS domainname and hostname of your
1028 box. For the classic darkstar.frop.org a simple:
1030 # echo "darkstar" > /proc/sys/kernel/hostname
1031 # echo "frop.org" > /proc/sys/kernel/domainname
1034 would suffice to set your hostname and NIS domainname.
1036 osrelease, ostype and version
1037 -----------------------------
1039 The names make it pretty obvious what these fields contain:
1041 > cat /proc/sys/kernel/osrelease
1044 > cat /proc/sys/kernel/ostype
1047 > cat /proc/sys/kernel/version
1048 #4 Fri Oct 1 12:41:14 PDT 1999
1051 The files osrelease and ostype should be clear enough. Version needs a little
1052 more clarification. The #4 means that this is the 4th kernel built from this
1053 source base and the date after it indicates the time the kernel was built. The
1054 only way to tune these values is to rebuild the kernel.
1059 The value in this file represents the number of seconds the kernel waits
1060 before rebooting on a panic. When you use the software watchdog, the
1061 recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1062 is disabled, which is the default setting.
1067 The four values in printk denote
1069 * default_message_loglevel,
1070 * minimum_console_loglevel and
1071 * default_console_loglevel
1074 These values influence printk() behavior when printing or logging error
1075 messages, which come from inside the kernel. See syslog(2) for more
1076 information on the different log levels.
1081 Messages with a higher priority than this will be printed to the console.
1083 default_message_level
1084 ---------------------
1086 Messages without an explicit priority will be printed with this priority.
1088 minimum_console_loglevel
1089 ------------------------
1091 Minimum (highest) value to which the console_loglevel can be set.
1093 default_console_loglevel
1094 ------------------------
1096 Default value for console_loglevel.
1101 This file shows the size of the generic SCSI (sg) buffer. At this point, you
1102 can't tune it yet, but you can change it at compile time by editing
1103 include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1105 If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1106 this to a higher value. Refer to the SANE documentation on this issue.
1111 The location where the modprobe binary is located. The kernel uses this
1112 program to load modules on demand.
1117 The value in this file affects behavior of handling NMI. When the value is
1118 non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1119 debugging information is displayed on console.
1121 NMI switch that most IA32 servers have fires unknown NMI up, for example.
1122 If a system hangs up, try pressing the NMI switch.
1125 This function and oprofile share a NMI callback. Therefore this function
1126 cannot be enabled when oprofile is activated.
1127 And NMI watchdog will be disabled when the value in this file is set to
1131 2.4 /proc/sys/vm - The virtual memory subsystem
1132 -----------------------------------------------
1134 The files in this directory can be used to tune the operation of the virtual
1135 memory (VM) subsystem of the Linux kernel.
1140 Controls the tendency of the kernel to reclaim the memory which is used for
1141 caching of directory and inode objects.
1143 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1144 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1145 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1146 to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1147 causes the kernel to prefer to reclaim dentries and inodes.
1149 dirty_background_ratio
1150 ----------------------
1152 Contains, as a percentage of total system memory, the number of pages at which
1153 the pdflush background writeback daemon will start writing out dirty data.
1158 Contains, as a percentage of total system memory, the number of pages at which
1159 a process which is generating disk writes will itself start writing out dirty
1162 dirty_writeback_centisecs
1163 -------------------------
1165 The pdflush writeback daemons will periodically wake up and write `old' data
1166 out to disk. This tunable expresses the interval between those wakeups, in
1167 100'ths of a second.
1169 Setting this to zero disables periodic writeback altogether.
1171 dirty_expire_centisecs
1172 ----------------------
1174 This tunable is used to define when dirty data is old enough to be eligible
1175 for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1176 Data which has been dirty in-memory for longer than this interval will be
1177 written out next time a pdflush daemon wakes up.
1182 If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1183 will use the legacy (2.4) layout for all processes.
1185 lower_zone_protection
1186 ---------------------
1188 For some specialised workloads on highmem machines it is dangerous for
1189 the kernel to allow process memory to be allocated from the "lowmem"
1190 zone. This is because that memory could then be pinned via the mlock()
1191 system call, or by unavailability of swapspace.
1193 And on large highmem machines this lack of reclaimable lowmem memory
1196 So the Linux page allocator has a mechanism which prevents allocations
1197 which _could_ use highmem from using too much lowmem. This means that
1198 a certain amount of lowmem is defended from the possibility of being
1199 captured into pinned user memory.
1201 (The same argument applies to the old 16 megabyte ISA DMA region. This
1202 mechanism will also defend that region from allocations which could use
1205 The `lower_zone_protection' tunable determines how aggressive the kernel is
1206 in defending these lower zones. The default value is zero - no
1209 If you have a machine which uses highmem or ISA DMA and your
1210 applications are using mlock(), or if you are running with no swap then
1211 you probably should increase the lower_zone_protection setting.
1213 The units of this tunable are fairly vague. It is approximately equal
1214 to "megabytes". So setting lower_zone_protection=100 will protect around 100
1215 megabytes of the lowmem zone from user allocations. It will also make
1216 those 100 megabytes unavaliable for use by applications and by
1217 pagecache, so there is a cost.
1219 The effects of this tunable may be observed by monitoring
1220 /proc/meminfo:LowFree. Write a single huge file and observe the point
1221 at which LowFree ceases to fall.
1223 A reasonable value for lower_zone_protection is 100.
1228 page-cluster controls the number of pages which are written to swap in
1229 a single attempt. The swap I/O size.
1231 It is a logarithmic value - setting it to zero means "1 page", setting
1232 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1234 The default value is three (eight pages at a time). There may be some
1235 small benefits in tuning this to a different value if your workload is
1241 This file contains one value. The following algorithm is used to decide if
1242 there's enough memory: if the value of overcommit_memory is positive, then
1243 there's always enough memory. This is a useful feature, since programs often
1244 malloc() huge amounts of memory 'just in case', while they only use a small
1245 part of it. Leaving this value at 0 will lead to the failure of such a huge
1246 malloc(), when in fact the system has enough memory for the program to run.
1248 On the other hand, enabling this feature can cause you to run out of memory
1249 and thrash the system to death, so large and/or important servers will want to
1250 set this value to 0.
1252 nr_hugepages and hugetlb_shm_group
1253 ----------------------------------
1255 nr_hugepages configures number of hugetlb page reserved for the system.
1257 hugetlb_shm_group contains group id that is allowed to create SysV shared
1258 memory segment using hugetlb page.
1263 laptop_mode is a knob that controls "laptop mode". All the things that are
1264 controlled by this knob are discussed in Documentation/laptop-mode.txt.
1269 block_dump enables block I/O debugging when set to a nonzero value. More
1270 information on block I/O debugging is in Documentation/laptop-mode.txt.
1272 2.5 /proc/sys/dev - Device specific parameters
1273 ----------------------------------------------
1275 Currently there is only support for CDROM drives, and for those, there is only
1276 one read-only file containing information about the CD-ROM drives attached to
1279 >cat /proc/sys/dev/cdrom/info
1280 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1284 drive # of slots: 1 0
1288 Can change speed: 1 1
1289 Can select disk: 0 1
1290 Can read multisession: 1 1
1292 Reports media changed: 1 1
1296 You see two drives, sr0 and hdb, along with a list of their features.
1298 2.6 /proc/sys/sunrpc - Remote procedure calls
1299 ---------------------------------------------
1301 This directory contains four files, which enable or disable debugging for the
1302 RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1303 be set to one to turn debugging on. (The default value is 0 for each)
1305 2.7 /proc/sys/net - Networking stuff
1306 ------------------------------------
1308 The interface to the networking parts of the kernel is located in
1309 /proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1310 some of them, depending on your kernel's configuration.
1313 Table 2-3: Subdirectories in /proc/sys/net
1314 ..............................................................................
1315 Directory Content Directory Content
1316 core General parameter appletalk Appletalk protocol
1317 unix Unix domain sockets netrom NET/ROM
1318 802 E802 protocol ax25 AX25
1319 ethernet Ethernet protocol rose X.25 PLP layer
1320 ipv4 IP version 4 x25 X.25 protocol
1321 ipx IPX token-ring IBM token ring
1322 bridge Bridging decnet DEC net
1324 ..............................................................................
1326 We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1327 only minor players in the Linux world, we'll skip them in this chapter. You'll
1328 find some short info on Appletalk and IPX further on in this chapter. Review
1329 the online documentation and the kernel source to get a detailed view of the
1330 parameters for those protocols. In this section we'll discuss the
1331 subdirectories printed in bold letters in the table above. As default values
1332 are suitable for most needs, there is no need to change these values.
1334 /proc/sys/net/core - Network core options
1335 -----------------------------------------
1340 The default setting of the socket receive buffer in bytes.
1345 The maximum receive socket buffer size in bytes.
1350 The default setting (in bytes) of the socket send buffer.
1355 The maximum send socket buffer size in bytes.
1357 message_burst and message_cost
1358 ------------------------------
1360 These parameters are used to limit the warning messages written to the kernel
1361 log from the networking code. They enforce a rate limit to make a
1362 denial-of-service attack impossible. A higher message_cost factor, results in
1363 fewer messages that will be written. Message_burst controls when messages will
1364 be dropped. The default settings limit warning messages to one every five
1370 Maximum number of packets, queued on the INPUT side, when the interface
1371 receives packets faster than kernel can process them.
1376 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1377 of struct cmsghdr structures with appended data.
1379 /proc/sys/net/unix - Parameters for Unix domain sockets
1380 -------------------------------------------------------
1382 There are only two files in this subdirectory. They control the delays for
1383 deleting and destroying socket descriptors.
1385 2.8 /proc/sys/net/ipv4 - IPV4 settings
1386 --------------------------------------
1388 IP version 4 is still the most used protocol in Unix networking. It will be
1389 replaced by IP version 6 in the next couple of years, but for the moment it's
1390 the de facto standard for the internet and is used in most networking
1391 environments around the world. Because of the importance of this protocol,
1392 we'll have a deeper look into the subtree controlling the behavior of the IPv4
1393 subsystem of the Linux kernel.
1395 Let's start with the entries in /proc/sys/net/ipv4.
1400 icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1401 ----------------------------------------------------
1403 Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1404 just those to broadcast and multicast addresses.
1406 Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1407 destination address your network may be used as an exploder for denial of
1408 service packet flooding attacks to other hosts.
1410 icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1411 ---------------------------------------------------------------------------------------
1413 Sets limits for sending ICMP packets to specific targets. A value of zero
1414 disables all limiting. Any positive value sets the maximum package rate in
1415 hundredth of a second (on Intel systems).
1423 This file contains the number one if the host received its IP configuration by
1424 RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1429 TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1430 hops a packet may travel.
1435 Enable dynamic socket address rewriting on interface address change. This is
1436 useful for dialup interface with changing IP addresses.
1441 Enable or disable forwarding of IP packages between interfaces. Changing this
1442 value resets all other parameters to their default values. They differ if the
1443 kernel is configured as host or router.
1448 Range of ports used by TCP and UDP to choose the local port. Contains two
1449 numbers, the first number is the lowest port, the second number the highest
1450 local port. Default is 1024-4999. Should be changed to 32768-61000 for
1456 Global switch to turn path MTU discovery off. It can also be set on a per
1457 socket basis by the applications or on a per route basis.
1462 Enable/disable debugging of IP masquerading.
1464 IP fragmentation settings
1465 -------------------------
1467 ipfrag_high_trash and ipfrag_low_trash
1468 --------------------------------------
1470 Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1471 of memory is allocated for this purpose, the fragment handler will toss
1472 packets until ipfrag_low_thresh is reached.
1477 Time in seconds to keep an IP fragment in memory.
1485 This file controls the use of the ECN bit in the IPv4 headers, this is a new
1486 feature about Explicit Congestion Notification, but some routers and firewalls
1487 block trafic that has this bit set, so it could be necessary to echo 0 to
1488 /proc/sys/net/ipv4/tcp_ecn, if you want to talk to this sites. For more info
1489 you could read RFC2481.
1491 tcp_retrans_collapse
1492 --------------------
1494 Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1495 larger packets to work around bugs in certain TCP stacks. Can be turned off by
1498 tcp_keepalive_probes
1499 --------------------
1501 Number of keep alive probes TCP sends out, until it decides that the
1502 connection is broken.
1507 How often TCP sends out keep alive messages, when keep alive is enabled. The
1513 Number of times initial SYNs for a TCP connection attempt will be
1514 retransmitted. Should not be higher than 255. This is only the timeout for
1515 outgoing connections, for incoming connections the number of retransmits is
1516 defined by tcp_retries1.
1521 Enable select acknowledgments after RFC2018.
1526 Enable timestamps as defined in RFC1323.
1531 Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1532 default is to use the BSD compatible interpretation of the urgent pointer
1533 pointing to the first byte after the urgent data. The RFC793 interpretation is
1534 to have it point to the last byte of urgent data. Enabling this option may
1535 lead to interoperatibility problems. Disabled by default.
1540 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1541 syncookies when the syn backlog queue of a socket overflows. This is to ward
1542 off the common 'syn flood attack'. Disabled by default.
1544 Note that the concept of a socket backlog is abandoned. This means the peer
1545 may not receive reliable error messages from an over loaded server with
1551 Enable window scaling as defined in RFC1323.
1556 The length of time in seconds it takes to receive a final FIN before the
1557 socket is always closed. This is strictly a violation of the TCP
1558 specification, but required to prevent denial-of-service attacks.
1563 Indicates how many keep alive probes are sent per slow timer run. Should not
1564 be set too high to prevent bursts.
1569 Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1570 in listen(2) only specifies the length of the backlog queue of already
1571 established sockets. When more connection requests arrive Linux starts to drop
1572 packets. When syncookies are enabled the packets are still answered and the
1573 maximum queue is effectively ignored.
1578 Defines how often an answer to a TCP connection request is retransmitted
1584 Defines how often a TCP packet is retransmitted before giving up.
1586 Interface specific settings
1587 ---------------------------
1589 In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1590 interface the system knows about and one directory calls all. Changes in the
1591 all subdirectory affect all interfaces, whereas changes in the other
1592 subdirectories affect only one interface. All directories have the same
1598 This switch decides if the kernel accepts ICMP redirect messages or not. The
1599 default is 'yes' if the kernel is configured for a regular host and 'no' for a
1600 router configuration.
1605 Should source routed packages be accepted or declined. The default is
1606 dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1612 Accept packets with source address 0.b.c.d with destinations not to this host
1613 as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1616 The default is 0, since this feature is not implemented yet (kernel version
1622 Enable or disable IP forwarding on this interface.
1627 Log packets with source addresses with no known route to kernel log.
1632 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1633 multicast routing daemon is required.
1638 Does (1) or does not (0) perform proxy ARP.
1643 Integer value determines if a source validation should be made. 1 means yes, 0
1644 means no. Disabled by default, but local/broadcast address spoofing is always
1647 If you set this to 1 on a router that is the only connection for a network to
1648 the net, it will prevent spoofing attacks against your internal networks
1649 (external addresses can still be spoofed), without the need for additional
1655 Accept ICMP redirect messages only for gateways, listed in default gateway
1656 list. Enabled by default.
1661 If it is not set the kernel does not assume that different subnets on this
1662 device can communicate directly. Default setting is 'yes'.
1667 Determines whether to send ICMP redirects to other hosts.
1672 The directory /proc/sys/net/ipv4/route contains several file to control
1675 error_burst and error_cost
1676 --------------------------
1678 These parameters are used to limit how many ICMP destination unreachable to
1679 send from the host in question. ICMP destination unreachable messages are
1680 sent when we can not reach the next hop, while trying to transmit a packet.
1681 It will also print some error messages to kernel logs if someone is ignoring
1682 our ICMP redirects. The higher the error_cost factor is, the fewer
1683 destination unreachable and error messages will be let through. Error_burst
1684 controls when destination unreachable messages and error messages will be
1685 dropped. The default settings limit warning messages to five every second.
1690 Writing to this file results in a flush of the routing cache.
1692 gc_elasticity, gc_interval, gc_min_interval, gc_tresh, gc_timeout,
1693 gc_thresh, gc_thresh1, gc_thresh2, gc_thresh3
1694 --------------------------------------------------------------
1696 Values to control the frequency and behavior of the garbage collection
1697 algorithm for the routing cache.
1702 Maximum size of the routing cache. Old entries will be purged once the cache
1703 reached has this size.
1705 max_delay, min_delay
1706 --------------------
1708 Delays for flushing the routing cache.
1710 redirect_load, redirect_number
1711 ------------------------------
1713 Factors which determine if more ICPM redirects should be sent to a specific
1714 host. No redirects will be sent once the load limit or the maximum number of
1715 redirects has been reached.
1720 Timeout for redirects. After this period redirects will be sent again, even if
1721 this has been stopped, because the load or number limit has been reached.
1723 Network Neighbor handling
1724 -------------------------
1726 Settings about how to handle connections with direct neighbors (nodes attached
1727 to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
1729 As we saw it in the conf directory, there is a default subdirectory which
1730 holds the default values, and one directory for each interface. The contents
1731 of the directories are identical, with the single exception that the default
1732 settings contain additional options to set garbage collection parameters.
1734 In the interface directories you'll find the following entries:
1739 A base value used for computing the random reachable time value as specified
1745 The time, expressed in jiffies (1/100 sec), between retransmitted Neighbor
1746 Solicitation messages. Used for address resolution and to determine if a
1747 neighbor is unreachable.
1752 Maximum queue length for a pending arp request - the number of packets which
1753 are accepted from other layers while the ARP address is still resolved.
1758 Maximum for random delay of answers to neighbor solicitation messages in
1759 jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
1765 Maximum number of retries for unicast solicitation.
1770 Maximum number of retries for multicast solicitation.
1772 delay_first_probe_time
1773 ----------------------
1775 Delay for the first time probe if the neighbor is reachable. (see
1781 An ARP/neighbor entry is only replaced with a new one if the old is at least
1782 locktime old. This prevents ARP cache thrashing.
1787 Maximum time (real time is random [0..proxytime]) before answering to an ARP
1788 request for which we have an proxy ARP entry. In some cases, this is used to
1789 prevent network flooding.
1794 Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
1799 Determines the number of requests to send to the user level ARP daemon. Use 0
1805 Determines how often to check for stale ARP entries. After an ARP entry is
1806 stale it will be resolved again (which is useful when an IP address migrates
1807 to another machine). When ucast_solicit is greater than 0 it first tries to
1808 send an ARP packet directly to the known host When that fails and
1809 mcast_solicit is greater than 0, an ARP request is broadcasted.
1814 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
1815 when Appletalk is loaded. The configurable parameters are:
1820 The amount of time we keep an ARP entry before expiring it. Used to age out
1826 The amount of time we will spend trying to resolve an Appletalk address.
1828 aarp-retransmit-limit
1829 ---------------------
1831 The number of times we will retransmit a query before giving up.
1836 Controls the rate at which expires are checked.
1838 The directory /proc/net/appletalk holds the list of active Appletalk sockets
1841 The fields indicate the DDP type, the local address (in network:node format)
1842 the remote address, the size of the transmit pending queue, the size of the
1843 received queue (bytes waiting for applications to read) the state and the uid
1846 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
1847 shows the name of the interface, its Appletalk address, the network range on
1848 that address (or network number for phase 1 networks), and the status of the
1851 /proc/net/atalk_route lists each known network route. It lists the target
1852 (network) that the route leads to, the router (may be directly connected), the
1853 route flags, and the device the route is using.
1858 The IPX protocol has no tunable values in proc/sys/net.
1860 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
1861 socket giving the local and remote addresses in Novell format (that is
1862 network:node:port). In accordance with the strange Novell tradition,
1863 everything but the port is in hex. Not_Connected is displayed for sockets that
1864 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
1865 the number of bytes pending for transmission and reception. The state
1866 indicates the state the socket is in and the uid is the owning uid of the
1869 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
1870 it gives the network number, the node number, and indicates if the network is
1871 the primary network. It also indicates which device it is bound to (or
1872 Internal for internal networks) and the Frame Type if appropriate. Linux
1873 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
1876 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
1877 gives the destination network, the router node (or Directly) and the network
1878 address of the router (or Connected) for internal networks.
1880 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
1881 ----------------------------------------------------------
1883 The "mqueue" filesystem provides the necessary kernel features to enable the
1884 creation of a user space library that implements the POSIX message queues
1885 API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
1886 Interfaces specification.)
1888 The "mqueue" filesystem contains values for determining/setting the amount of
1889 resources used by the file system.
1891 /proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
1892 maximum number of message queues allowed on the system.
1894 /proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
1895 maximum number of messages in a queue value. In fact it is the limiting value
1896 for another (user) limit which is set in mq_open invocation. This attribute of
1897 a queue must be less or equal then msg_max.
1899 /proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
1900 maximum message size value (it is every message queue's attribute set during
1904 ------------------------------------------------------------------------------
1906 ------------------------------------------------------------------------------
1907 Certain aspects of kernel behavior can be modified at runtime, without the
1908 need to recompile the kernel, or even to reboot the system. The files in the
1909 /proc/sys tree can not only be read, but also modified. You can use the echo
1910 command to write value into these files, thereby changing the default settings
1912 ------------------------------------------------------------------------------