proc.txt 81 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050105110521053105410551056105710581059106010611062106310641065106610671068106910701071107210731074107510761077107810791080108110821083108410851086108710881089109010911092109310941095109610971098109911001101110211031104110511061107110811091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239124012411242124312441245124612471248124912501251125212531254125512561257125812591260126112621263126412651266126712681269127012711272127312741275127612771278127912801281128212831284128512861287128812891290129112921293129412951296129712981299130013011302130313041305130613071308130913101311131213131314131513161317131813191320132113221323132413251326132713281329133013311332133313341335133613371338133913401341134213431344134513461347134813491350135113521353135413551356135713581359136013611362136313641365136613671368136913701371137213731374137513761377137813791380138113821383138413851386138713881389139013911392139313941395139613971398139914001401140214031404140514061407140814091410141114121413141414151416141714181419142014211422142314241425142614271428142914301431143214331434143514361437143814391440144114421443144414451446144714481449145014511452145314541455145614571458145914601461146214631464146514661467146814691470147114721473147414751476147714781479148014811482148314841485148614871488148914901491149214931494149514961497149814991500150115021503150415051506150715081509151015111512151315141515151615171518151915201521152215231524152515261527152815291530153115321533153415351536153715381539154015411542154315441545154615471548154915501551155215531554155515561557155815591560156115621563156415651566156715681569157015711572157315741575157615771578157915801581158215831584158515861587158815891590159115921593159415951596159715981599160016011602160316041605160616071608160916101611161216131614161516161617161816191620162116221623162416251626162716281629163016311632163316341635163616371638163916401641164216431644164516461647164816491650165116521653165416551656165716581659166016611662166316641665166616671668166916701671167216731674167516761677167816791680168116821683168416851686168716881689169016911692169316941695169616971698169917001701170217031704170517061707170817091710171117121713171417151716171717181719172017211722172317241725172617271728172917301731173217331734173517361737173817391740174117421743174417451746174717481749175017511752175317541755175617571758175917601761176217631764176517661767176817691770177117721773177417751776177717781779178017811782178317841785178617871788178917901791179217931794179517961797179817991800180118021803180418051806180718081809181018111812181318141815181618171818181918201821182218231824182518261827182818291830183118321833183418351836183718381839184018411842184318441845184618471848184918501851185218531854185518561857185818591860186118621863186418651866186718681869187018711872187318741875187618771878187918801881188218831884188518861887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911191219131914191519161917191819191920192119221923192419251926192719281929193019311932193319341935193619371938193919401941194219431944194519461947194819491950195119521953195419551956195719581959196019611962196319641965196619671968196919701971197219731974197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010201120122013201420152016201720182019202020212022202320242025202620272028202920302031203220332034203520362037203820392040204120422043204420452046204720482049205020512052205320542055205620572058205920602061206220632064206520662067206820692070207120722073207420752076207720782079208020812082208320842085208620872088208920902091209220932094209520962097
  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>
  6. 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
  7. ------------------------------------------------------------------------------
  8. Version 1.3 Kernel version 2.2.12
  9. Kernel version 2.4.0-test11-pre4
  10. ------------------------------------------------------------------------------
  11. Table of Contents
  12. -----------------
  13. 0 Preface
  14. 0.1 Introduction/Credits
  15. 0.2 Legal Stuff
  16. 1 Collecting System Information
  17. 1.1 Process-Specific Subdirectories
  18. 1.2 Kernel data
  19. 1.3 IDE devices in /proc/ide
  20. 1.4 Networking info in /proc/net
  21. 1.5 SCSI info
  22. 1.6 Parallel port info in /proc/parport
  23. 1.7 TTY info in /proc/tty
  24. 1.8 Miscellaneous kernel statistics in /proc/stat
  25. 2 Modifying System Parameters
  26. 2.1 /proc/sys/fs - File system data
  27. 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
  28. 2.3 /proc/sys/kernel - general kernel parameters
  29. 2.4 /proc/sys/vm - The virtual memory subsystem
  30. 2.5 /proc/sys/dev - Device specific parameters
  31. 2.6 /proc/sys/sunrpc - Remote procedure calls
  32. 2.7 /proc/sys/net - Networking stuff
  33. 2.8 /proc/sys/net/ipv4 - IPV4 settings
  34. 2.9 Appletalk
  35. 2.10 IPX
  36. 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
  37. 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
  38. 2.13 /proc/<pid>/oom_score - Display current oom-killer score
  39. 2.14 /proc/<pid>/io - Display the IO accounting fields
  40. ------------------------------------------------------------------------------
  41. Preface
  42. ------------------------------------------------------------------------------
  43. 0.1 Introduction/Credits
  44. ------------------------
  45. This documentation is part of a soon (or so we hope) to be released book on
  46. the SuSE Linux distribution. As there is no complete documentation for the
  47. /proc file system and we've used many freely available sources to write these
  48. chapters, it seems only fair to give the work back to the Linux community.
  49. This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
  50. afraid it's still far from complete, but we hope it will be useful. As far as
  51. we know, it is the first 'all-in-one' document about the /proc file system. It
  52. is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
  53. SPARC, AXP, etc., features, you probably won't find what you are looking for.
  54. It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
  55. additions and patches are welcome and will be added to this document if you
  56. mail them to Bodo.
  57. We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
  58. other people for help compiling this documentation. We'd also like to extend a
  59. special thank you to Andi Kleen for documentation, which we relied on heavily
  60. to create this document, as well as the additional information he provided.
  61. Thanks to everybody else who contributed source or docs to the Linux kernel
  62. and helped create a great piece of software... :)
  63. If you have any comments, corrections or additions, please don't hesitate to
  64. contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
  65. document.
  66. The latest version of this document is available online at
  67. http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
  68. If the above direction does not works for you, ypu could try the kernel
  69. mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
  70. comandante@zaralinux.com.
  71. 0.2 Legal Stuff
  72. ---------------
  73. We don't guarantee the correctness of this document, and if you come to us
  74. complaining about how you screwed up your system because of incorrect
  75. documentation, we won't feel responsible...
  76. ------------------------------------------------------------------------------
  77. CHAPTER 1: COLLECTING SYSTEM INFORMATION
  78. ------------------------------------------------------------------------------
  79. ------------------------------------------------------------------------------
  80. In This Chapter
  81. ------------------------------------------------------------------------------
  82. * Investigating the properties of the pseudo file system /proc and its
  83. ability to provide information on the running Linux system
  84. * Examining /proc's structure
  85. * Uncovering various information about the kernel and the processes running
  86. on the system
  87. ------------------------------------------------------------------------------
  88. The proc file system acts as an interface to internal data structures in the
  89. kernel. It can be used to obtain information about the system and to change
  90. certain kernel parameters at runtime (sysctl).
  91. First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
  92. show you how you can use /proc/sys to change settings.
  93. 1.1 Process-Specific Subdirectories
  94. -----------------------------------
  95. The directory /proc contains (among other things) one subdirectory for each
  96. process running on the system, which is named after the process ID (PID).
  97. The link self points to the process reading the file system. Each process
  98. subdirectory has the entries listed in Table 1-1.
  99. Table 1-1: Process specific entries in /proc
  100. ..............................................................................
  101. File Content
  102. cmdline Command line arguments
  103. cpu Current and last cpu in which it was executed (2.4)(smp)
  104. cwd Link to the current working directory
  105. environ Values of environment variables
  106. exe Link to the executable of this process
  107. fd Directory, which contains all file descriptors
  108. maps Memory maps to executables and library files (2.4)
  109. mem Memory held by this process
  110. root Link to the root directory of this process
  111. stat Process status
  112. statm Process memory status information
  113. status Process status in human readable form
  114. wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
  115. smaps Extension based on maps, presenting the rss size for each mapped file
  116. ..............................................................................
  117. For example, to get the status information of a process, all you have to do is
  118. read the file /proc/PID/status:
  119. >cat /proc/self/status
  120. Name: cat
  121. State: R (running)
  122. Pid: 5452
  123. PPid: 743
  124. TracerPid: 0 (2.4)
  125. Uid: 501 501 501 501
  126. Gid: 100 100 100 100
  127. Groups: 100 14 16
  128. VmSize: 1112 kB
  129. VmLck: 0 kB
  130. VmRSS: 348 kB
  131. VmData: 24 kB
  132. VmStk: 12 kB
  133. VmExe: 8 kB
  134. VmLib: 1044 kB
  135. SigPnd: 0000000000000000
  136. SigBlk: 0000000000000000
  137. SigIgn: 0000000000000000
  138. SigCgt: 0000000000000000
  139. CapInh: 00000000fffffeff
  140. CapPrm: 0000000000000000
  141. CapEff: 0000000000000000
  142. This shows you nearly the same information you would get if you viewed it with
  143. the ps command. In fact, ps uses the proc file system to obtain its
  144. information. The statm file contains more detailed information about the
  145. process memory usage. Its seven fields are explained in Table 1-2.
  146. Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
  147. ..............................................................................
  148. Field Content
  149. size total program size (pages) (same as VmSize in status)
  150. resident size of memory portions (pages) (same as VmRSS in status)
  151. shared number of pages that are shared (i.e. backed by a file)
  152. trs number of pages that are 'code' (not including libs; broken,
  153. includes data segment)
  154. lrs number of pages of library (always 0 on 2.6)
  155. drs number of pages of data/stack (including libs; broken,
  156. includes library text)
  157. dt number of dirty pages (always 0 on 2.6)
  158. ..............................................................................
  159. 1.2 Kernel data
  160. ---------------
  161. Similar to the process entries, the kernel data files give information about
  162. the running kernel. The files used to obtain this information are contained in
  163. /proc and are listed in Table 1-3. Not all of these will be present in your
  164. system. It depends on the kernel configuration and the loaded modules, which
  165. files are there, and which are missing.
  166. Table 1-3: Kernel info in /proc
  167. ..............................................................................
  168. File Content
  169. apm Advanced power management info
  170. buddyinfo Kernel memory allocator information (see text) (2.5)
  171. bus Directory containing bus specific information
  172. cmdline Kernel command line
  173. cpuinfo Info about the CPU
  174. devices Available devices (block and character)
  175. dma Used DMS channels
  176. filesystems Supported filesystems
  177. driver Various drivers grouped here, currently rtc (2.4)
  178. execdomains Execdomains, related to security (2.4)
  179. fb Frame Buffer devices (2.4)
  180. fs File system parameters, currently nfs/exports (2.4)
  181. ide Directory containing info about the IDE subsystem
  182. interrupts Interrupt usage
  183. iomem Memory map (2.4)
  184. ioports I/O port usage
  185. irq Masks for irq to cpu affinity (2.4)(smp?)
  186. isapnp ISA PnP (Plug&Play) Info (2.4)
  187. kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
  188. kmsg Kernel messages
  189. ksyms Kernel symbol table
  190. loadavg Load average of last 1, 5 & 15 minutes
  191. locks Kernel locks
  192. meminfo Memory info
  193. misc Miscellaneous
  194. modules List of loaded modules
  195. mounts Mounted filesystems
  196. net Networking info (see text)
  197. partitions Table of partitions known to the system
  198. pci Depreciated info of PCI bus (new way -> /proc/bus/pci/,
  199. decoupled by lspci (2.4)
  200. rtc Real time clock
  201. scsi SCSI info (see text)
  202. slabinfo Slab pool info
  203. stat Overall statistics
  204. swaps Swap space utilization
  205. sys See chapter 2
  206. sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
  207. tty Info of tty drivers
  208. uptime System uptime
  209. version Kernel version
  210. video bttv info of video resources (2.4)
  211. ..............................................................................
  212. You can, for example, check which interrupts are currently in use and what
  213. they are used for by looking in the file /proc/interrupts:
  214. > cat /proc/interrupts
  215. CPU0
  216. 0: 8728810 XT-PIC timer
  217. 1: 895 XT-PIC keyboard
  218. 2: 0 XT-PIC cascade
  219. 3: 531695 XT-PIC aha152x
  220. 4: 2014133 XT-PIC serial
  221. 5: 44401 XT-PIC pcnet_cs
  222. 8: 2 XT-PIC rtc
  223. 11: 8 XT-PIC i82365
  224. 12: 182918 XT-PIC PS/2 Mouse
  225. 13: 1 XT-PIC fpu
  226. 14: 1232265 XT-PIC ide0
  227. 15: 7 XT-PIC ide1
  228. NMI: 0
  229. In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
  230. output of a SMP machine):
  231. > cat /proc/interrupts
  232. CPU0 CPU1
  233. 0: 1243498 1214548 IO-APIC-edge timer
  234. 1: 8949 8958 IO-APIC-edge keyboard
  235. 2: 0 0 XT-PIC cascade
  236. 5: 11286 10161 IO-APIC-edge soundblaster
  237. 8: 1 0 IO-APIC-edge rtc
  238. 9: 27422 27407 IO-APIC-edge 3c503
  239. 12: 113645 113873 IO-APIC-edge PS/2 Mouse
  240. 13: 0 0 XT-PIC fpu
  241. 14: 22491 24012 IO-APIC-edge ide0
  242. 15: 2183 2415 IO-APIC-edge ide1
  243. 17: 30564 30414 IO-APIC-level eth0
  244. 18: 177 164 IO-APIC-level bttv
  245. NMI: 2457961 2457959
  246. LOC: 2457882 2457881
  247. ERR: 2155
  248. NMI is incremented in this case because every timer interrupt generates a NMI
  249. (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
  250. LOC is the local interrupt counter of the internal APIC of every CPU.
  251. ERR is incremented in the case of errors in the IO-APIC bus (the bus that
  252. connects the CPUs in a SMP system. This means that an error has been detected,
  253. the IO-APIC automatically retry the transmission, so it should not be a big
  254. problem, but you should read the SMP-FAQ.
  255. In this context it could be interesting to note the new irq directory in 2.4.
  256. It could be used to set IRQ to CPU affinity, this means that you can "hook" an
  257. IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
  258. irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
  259. For example
  260. > ls /proc/irq/
  261. 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
  262. 1 11 13 15 17 19 3 5 7 9
  263. > ls /proc/irq/0/
  264. smp_affinity
  265. The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
  266. is the same by default:
  267. > cat /proc/irq/0/smp_affinity
  268. ffffffff
  269. It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can
  270. set it by doing:
  271. > echo 1 > /proc/irq/prof_cpu_mask
  272. This means that only the first CPU will handle the IRQ, but you can also echo 5
  273. which means that only the first and fourth CPU can handle the IRQ.
  274. The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
  275. between all the CPUs which are allowed to handle it. As usual the kernel has
  276. more info than you and does a better job than you, so the defaults are the
  277. best choice for almost everyone.
  278. There are three more important subdirectories in /proc: net, scsi, and sys.
  279. The general rule is that the contents, or even the existence of these
  280. directories, depend on your kernel configuration. If SCSI is not enabled, the
  281. directory scsi may not exist. The same is true with the net, which is there
  282. only when networking support is present in the running kernel.
  283. The slabinfo file gives information about memory usage at the slab level.
  284. Linux uses slab pools for memory management above page level in version 2.2.
  285. Commonly used objects have their own slab pool (such as network buffers,
  286. directory cache, and so on).
  287. ..............................................................................
  288. > cat /proc/buddyinfo
  289. Node 0, zone DMA 0 4 5 4 4 3 ...
  290. Node 0, zone Normal 1 0 0 1 101 8 ...
  291. Node 0, zone HighMem 2 0 0 1 1 0 ...
  292. Memory fragmentation is a problem under some workloads, and buddyinfo is a
  293. useful tool for helping diagnose these problems. Buddyinfo will give you a
  294. clue as to how big an area you can safely allocate, or why a previous
  295. allocation failed.
  296. Each column represents the number of pages of a certain order which are
  297. available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
  298. ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
  299. available in ZONE_NORMAL, etc...
  300. ..............................................................................
  301. meminfo:
  302. Provides information about distribution and utilization of memory. This
  303. varies by architecture and compile options. The following is from a
  304. 16GB PIII, which has highmem enabled. You may not have all of these fields.
  305. > cat /proc/meminfo
  306. MemTotal: 16344972 kB
  307. MemFree: 13634064 kB
  308. Buffers: 3656 kB
  309. Cached: 1195708 kB
  310. SwapCached: 0 kB
  311. Active: 891636 kB
  312. Inactive: 1077224 kB
  313. HighTotal: 15597528 kB
  314. HighFree: 13629632 kB
  315. LowTotal: 747444 kB
  316. LowFree: 4432 kB
  317. SwapTotal: 0 kB
  318. SwapFree: 0 kB
  319. Dirty: 968 kB
  320. Writeback: 0 kB
  321. Mapped: 280372 kB
  322. Slab: 684068 kB
  323. CommitLimit: 7669796 kB
  324. Committed_AS: 100056 kB
  325. PageTables: 24448 kB
  326. VmallocTotal: 112216 kB
  327. VmallocUsed: 428 kB
  328. VmallocChunk: 111088 kB
  329. MemTotal: Total usable ram (i.e. physical ram minus a few reserved
  330. bits and the kernel binary code)
  331. MemFree: The sum of LowFree+HighFree
  332. Buffers: Relatively temporary storage for raw disk blocks
  333. shouldn't get tremendously large (20MB or so)
  334. Cached: in-memory cache for files read from the disk (the
  335. pagecache). Doesn't include SwapCached
  336. SwapCached: Memory that once was swapped out, is swapped back in but
  337. still also is in the swapfile (if memory is needed it
  338. doesn't need to be swapped out AGAIN because it is already
  339. in the swapfile. This saves I/O)
  340. Active: Memory that has been used more recently and usually not
  341. reclaimed unless absolutely necessary.
  342. Inactive: Memory which has been less recently used. It is more
  343. eligible to be reclaimed for other purposes
  344. HighTotal:
  345. HighFree: Highmem is all memory above ~860MB of physical memory
  346. Highmem areas are for use by userspace programs, or
  347. for the pagecache. The kernel must use tricks to access
  348. this memory, making it slower to access than lowmem.
  349. LowTotal:
  350. LowFree: Lowmem is memory which can be used for everything that
  351. highmem can be used for, but it is also available for the
  352. kernel's use for its own data structures. Among many
  353. other things, it is where everything from the Slab is
  354. allocated. Bad things happen when you're out of lowmem.
  355. SwapTotal: total amount of swap space available
  356. SwapFree: Memory which has been evicted from RAM, and is temporarily
  357. on the disk
  358. Dirty: Memory which is waiting to get written back to the disk
  359. Writeback: Memory which is actively being written back to the disk
  360. Mapped: files which have been mmaped, such as libraries
  361. Slab: in-kernel data structures cache
  362. CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
  363. this is the total amount of memory currently available to
  364. be allocated on the system. This limit is only adhered to
  365. if strict overcommit accounting is enabled (mode 2 in
  366. 'vm.overcommit_memory').
  367. The CommitLimit is calculated with the following formula:
  368. CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
  369. For example, on a system with 1G of physical RAM and 7G
  370. of swap with a `vm.overcommit_ratio` of 30 it would
  371. yield a CommitLimit of 7.3G.
  372. For more details, see the memory overcommit documentation
  373. in vm/overcommit-accounting.
  374. Committed_AS: The amount of memory presently allocated on the system.
  375. The committed memory is a sum of all of the memory which
  376. has been allocated by processes, even if it has not been
  377. "used" by them as of yet. A process which malloc()'s 1G
  378. of memory, but only touches 300M of it will only show up
  379. as using 300M of memory even if it has the address space
  380. allocated for the entire 1G. This 1G is memory which has
  381. been "committed" to by the VM and can be used at any time
  382. by the allocating application. With strict overcommit
  383. enabled on the system (mode 2 in 'vm.overcommit_memory'),
  384. allocations which would exceed the CommitLimit (detailed
  385. above) will not be permitted. This is useful if one needs
  386. to guarantee that processes will not fail due to lack of
  387. memory once that memory has been successfully allocated.
  388. PageTables: amount of memory dedicated to the lowest level of page
  389. tables.
  390. VmallocTotal: total size of vmalloc memory area
  391. VmallocUsed: amount of vmalloc area which is used
  392. VmallocChunk: largest contigious block of vmalloc area which is free
  393. 1.3 IDE devices in /proc/ide
  394. ----------------------------
  395. The subdirectory /proc/ide contains information about all IDE devices of which
  396. the kernel is aware. There is one subdirectory for each IDE controller, the
  397. file drivers and a link for each IDE device, pointing to the device directory
  398. in the controller specific subtree.
  399. The file drivers contains general information about the drivers used for the
  400. IDE devices:
  401. > cat /proc/ide/drivers
  402. ide-cdrom version 4.53
  403. ide-disk version 1.08
  404. More detailed information can be found in the controller specific
  405. subdirectories. These are named ide0, ide1 and so on. Each of these
  406. directories contains the files shown in table 1-4.
  407. Table 1-4: IDE controller info in /proc/ide/ide?
  408. ..............................................................................
  409. File Content
  410. channel IDE channel (0 or 1)
  411. config Configuration (only for PCI/IDE bridge)
  412. mate Mate name
  413. model Type/Chipset of IDE controller
  414. ..............................................................................
  415. Each device connected to a controller has a separate subdirectory in the
  416. controllers directory. The files listed in table 1-5 are contained in these
  417. directories.
  418. Table 1-5: IDE device information
  419. ..............................................................................
  420. File Content
  421. cache The cache
  422. capacity Capacity of the medium (in 512Byte blocks)
  423. driver driver and version
  424. geometry physical and logical geometry
  425. identify device identify block
  426. media media type
  427. model device identifier
  428. settings device setup
  429. smart_thresholds IDE disk management thresholds
  430. smart_values IDE disk management values
  431. ..............................................................................
  432. The most interesting file is settings. This file contains a nice overview of
  433. the drive parameters:
  434. # cat /proc/ide/ide0/hda/settings
  435. name value min max mode
  436. ---- ----- --- --- ----
  437. bios_cyl 526 0 65535 rw
  438. bios_head 255 0 255 rw
  439. bios_sect 63 0 63 rw
  440. breada_readahead 4 0 127 rw
  441. bswap 0 0 1 r
  442. file_readahead 72 0 2097151 rw
  443. io_32bit 0 0 3 rw
  444. keepsettings 0 0 1 rw
  445. max_kb_per_request 122 1 127 rw
  446. multcount 0 0 8 rw
  447. nice1 1 0 1 rw
  448. nowerr 0 0 1 rw
  449. pio_mode write-only 0 255 w
  450. slow 0 0 1 rw
  451. unmaskirq 0 0 1 rw
  452. using_dma 0 0 1 rw
  453. 1.4 Networking info in /proc/net
  454. --------------------------------
  455. The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
  456. additional values you get for IP version 6 if you configure the kernel to
  457. support this. Table 1-7 lists the files and their meaning.
  458. Table 1-6: IPv6 info in /proc/net
  459. ..............................................................................
  460. File Content
  461. udp6 UDP sockets (IPv6)
  462. tcp6 TCP sockets (IPv6)
  463. raw6 Raw device statistics (IPv6)
  464. igmp6 IP multicast addresses, which this host joined (IPv6)
  465. if_inet6 List of IPv6 interface addresses
  466. ipv6_route Kernel routing table for IPv6
  467. rt6_stats Global IPv6 routing tables statistics
  468. sockstat6 Socket statistics (IPv6)
  469. snmp6 Snmp data (IPv6)
  470. ..............................................................................
  471. Table 1-7: Network info in /proc/net
  472. ..............................................................................
  473. File Content
  474. arp Kernel ARP table
  475. dev network devices with statistics
  476. dev_mcast the Layer2 multicast groups a device is listening too
  477. (interface index, label, number of references, number of bound
  478. addresses).
  479. dev_stat network device status
  480. ip_fwchains Firewall chain linkage
  481. ip_fwnames Firewall chain names
  482. ip_masq Directory containing the masquerading tables
  483. ip_masquerade Major masquerading table
  484. netstat Network statistics
  485. raw raw device statistics
  486. route Kernel routing table
  487. rpc Directory containing rpc info
  488. rt_cache Routing cache
  489. snmp SNMP data
  490. sockstat Socket statistics
  491. tcp TCP sockets
  492. tr_rif Token ring RIF routing table
  493. udp UDP sockets
  494. unix UNIX domain sockets
  495. wireless Wireless interface data (Wavelan etc)
  496. igmp IP multicast addresses, which this host joined
  497. psched Global packet scheduler parameters.
  498. netlink List of PF_NETLINK sockets
  499. ip_mr_vifs List of multicast virtual interfaces
  500. ip_mr_cache List of multicast routing cache
  501. ..............................................................................
  502. You can use this information to see which network devices are available in
  503. your system and how much traffic was routed over those devices:
  504. > cat /proc/net/dev
  505. Inter-|Receive |[...
  506. face |bytes packets errs drop fifo frame compressed multicast|[...
  507. lo: 908188 5596 0 0 0 0 0 0 [...
  508. ppp0:15475140 20721 410 0 0 410 0 0 [...
  509. eth0: 614530 7085 0 0 0 0 0 1 [...
  510. ...] Transmit
  511. ...] bytes packets errs drop fifo colls carrier compressed
  512. ...] 908188 5596 0 0 0 0 0 0
  513. ...] 1375103 17405 0 0 0 0 0 0
  514. ...] 1703981 5535 0 0 0 3 0 0
  515. In addition, each Channel Bond interface has it's own directory. For
  516. example, the bond0 device will have a directory called /proc/net/bond0/.
  517. It will contain information that is specific to that bond, such as the
  518. current slaves of the bond, the link status of the slaves, and how
  519. many times the slaves link has failed.
  520. 1.5 SCSI info
  521. -------------
  522. If you have a SCSI host adapter in your system, you'll find a subdirectory
  523. named after the driver for this adapter in /proc/scsi. You'll also see a list
  524. of all recognized SCSI devices in /proc/scsi:
  525. >cat /proc/scsi/scsi
  526. Attached devices:
  527. Host: scsi0 Channel: 00 Id: 00 Lun: 00
  528. Vendor: IBM Model: DGHS09U Rev: 03E0
  529. Type: Direct-Access ANSI SCSI revision: 03
  530. Host: scsi0 Channel: 00 Id: 06 Lun: 00
  531. Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
  532. Type: CD-ROM ANSI SCSI revision: 02
  533. The directory named after the driver has one file for each adapter found in
  534. the system. These files contain information about the controller, including
  535. the used IRQ and the IO address range. The amount of information shown is
  536. dependent on the adapter you use. The example shows the output for an Adaptec
  537. AHA-2940 SCSI adapter:
  538. > cat /proc/scsi/aic7xxx/0
  539. Adaptec AIC7xxx driver version: 5.1.19/3.2.4
  540. Compile Options:
  541. TCQ Enabled By Default : Disabled
  542. AIC7XXX_PROC_STATS : Disabled
  543. AIC7XXX_RESET_DELAY : 5
  544. Adapter Configuration:
  545. SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
  546. Ultra Wide Controller
  547. PCI MMAPed I/O Base: 0xeb001000
  548. Adapter SEEPROM Config: SEEPROM found and used.
  549. Adaptec SCSI BIOS: Enabled
  550. IRQ: 10
  551. SCBs: Active 0, Max Active 2,
  552. Allocated 15, HW 16, Page 255
  553. Interrupts: 160328
  554. BIOS Control Word: 0x18b6
  555. Adapter Control Word: 0x005b
  556. Extended Translation: Enabled
  557. Disconnect Enable Flags: 0xffff
  558. Ultra Enable Flags: 0x0001
  559. Tag Queue Enable Flags: 0x0000
  560. Ordered Queue Tag Flags: 0x0000
  561. Default Tag Queue Depth: 8
  562. Tagged Queue By Device array for aic7xxx host instance 0:
  563. {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
  564. Actual queue depth per device for aic7xxx host instance 0:
  565. {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
  566. Statistics:
  567. (scsi0:0:0:0)
  568. Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
  569. Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
  570. Total transfers 160151 (74577 reads and 85574 writes)
  571. (scsi0:0:6:0)
  572. Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
  573. Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
  574. Total transfers 0 (0 reads and 0 writes)
  575. 1.6 Parallel port info in /proc/parport
  576. ---------------------------------------
  577. The directory /proc/parport contains information about the parallel ports of
  578. your system. It has one subdirectory for each port, named after the port
  579. number (0,1,2,...).
  580. These directories contain the four files shown in Table 1-8.
  581. Table 1-8: Files in /proc/parport
  582. ..............................................................................
  583. File Content
  584. autoprobe Any IEEE-1284 device ID information that has been acquired.
  585. devices list of the device drivers using that port. A + will appear by the
  586. name of the device currently using the port (it might not appear
  587. against any).
  588. hardware Parallel port's base address, IRQ line and DMA channel.
  589. irq IRQ that parport is using for that port. This is in a separate
  590. file to allow you to alter it by writing a new value in (IRQ
  591. number or none).
  592. ..............................................................................
  593. 1.7 TTY info in /proc/tty
  594. -------------------------
  595. Information about the available and actually used tty's can be found in the
  596. directory /proc/tty.You'll find entries for drivers and line disciplines in
  597. this directory, as shown in Table 1-9.
  598. Table 1-9: Files in /proc/tty
  599. ..............................................................................
  600. File Content
  601. drivers list of drivers and their usage
  602. ldiscs registered line disciplines
  603. driver/serial usage statistic and status of single tty lines
  604. ..............................................................................
  605. To see which tty's are currently in use, you can simply look into the file
  606. /proc/tty/drivers:
  607. > cat /proc/tty/drivers
  608. pty_slave /dev/pts 136 0-255 pty:slave
  609. pty_master /dev/ptm 128 0-255 pty:master
  610. pty_slave /dev/ttyp 3 0-255 pty:slave
  611. pty_master /dev/pty 2 0-255 pty:master
  612. serial /dev/cua 5 64-67 serial:callout
  613. serial /dev/ttyS 4 64-67 serial
  614. /dev/tty0 /dev/tty0 4 0 system:vtmaster
  615. /dev/ptmx /dev/ptmx 5 2 system
  616. /dev/console /dev/console 5 1 system:console
  617. /dev/tty /dev/tty 5 0 system:/dev/tty
  618. unknown /dev/tty 4 1-63 console
  619. 1.8 Miscellaneous kernel statistics in /proc/stat
  620. -------------------------------------------------
  621. Various pieces of information about kernel activity are available in the
  622. /proc/stat file. All of the numbers reported in this file are aggregates
  623. since the system first booted. For a quick look, simply cat the file:
  624. > cat /proc/stat
  625. cpu 2255 34 2290 22625563 6290 127 456
  626. cpu0 1132 34 1441 11311718 3675 127 438
  627. cpu1 1123 0 849 11313845 2614 0 18
  628. intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
  629. ctxt 1990473
  630. btime 1062191376
  631. processes 2915
  632. procs_running 1
  633. procs_blocked 0
  634. The very first "cpu" line aggregates the numbers in all of the other "cpuN"
  635. lines. These numbers identify the amount of time the CPU has spent performing
  636. different kinds of work. Time units are in USER_HZ (typically hundredths of a
  637. second). The meanings of the columns are as follows, from left to right:
  638. - user: normal processes executing in user mode
  639. - nice: niced processes executing in user mode
  640. - system: processes executing in kernel mode
  641. - idle: twiddling thumbs
  642. - iowait: waiting for I/O to complete
  643. - irq: servicing interrupts
  644. - softirq: servicing softirqs
  645. The "intr" line gives counts of interrupts serviced since boot time, for each
  646. of the possible system interrupts. The first column is the total of all
  647. interrupts serviced; each subsequent column is the total for that particular
  648. interrupt.
  649. The "ctxt" line gives the total number of context switches across all CPUs.
  650. The "btime" line gives the time at which the system booted, in seconds since
  651. the Unix epoch.
  652. The "processes" line gives the number of processes and threads created, which
  653. includes (but is not limited to) those created by calls to the fork() and
  654. clone() system calls.
  655. The "procs_running" line gives the number of processes currently running on
  656. CPUs.
  657. The "procs_blocked" line gives the number of processes currently blocked,
  658. waiting for I/O to complete.
  659. ------------------------------------------------------------------------------
  660. Summary
  661. ------------------------------------------------------------------------------
  662. The /proc file system serves information about the running system. It not only
  663. allows access to process data but also allows you to request the kernel status
  664. by reading files in the hierarchy.
  665. The directory structure of /proc reflects the types of information and makes
  666. it easy, if not obvious, where to look for specific data.
  667. ------------------------------------------------------------------------------
  668. ------------------------------------------------------------------------------
  669. CHAPTER 2: MODIFYING SYSTEM PARAMETERS
  670. ------------------------------------------------------------------------------
  671. ------------------------------------------------------------------------------
  672. In This Chapter
  673. ------------------------------------------------------------------------------
  674. * Modifying kernel parameters by writing into files found in /proc/sys
  675. * Exploring the files which modify certain parameters
  676. * Review of the /proc/sys file tree
  677. ------------------------------------------------------------------------------
  678. A very interesting part of /proc is the directory /proc/sys. This is not only
  679. a source of information, it also allows you to change parameters within the
  680. kernel. Be very careful when attempting this. You can optimize your system,
  681. but you can also cause it to crash. Never alter kernel parameters on a
  682. production system. Set up a development machine and test to make sure that
  683. everything works the way you want it to. You may have no alternative but to
  684. reboot the machine once an error has been made.
  685. To change a value, simply echo the new value into the file. An example is
  686. given below in the section on the file system data. You need to be root to do
  687. this. You can create your own boot script to perform this every time your
  688. system boots.
  689. The files in /proc/sys can be used to fine tune and monitor miscellaneous and
  690. general things in the operation of the Linux kernel. Since some of the files
  691. can inadvertently disrupt your system, it is advisable to read both
  692. documentation and source before actually making adjustments. In any case, be
  693. very careful when writing to any of these files. The entries in /proc may
  694. change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
  695. review the kernel documentation in the directory /usr/src/linux/Documentation.
  696. This chapter is heavily based on the documentation included in the pre 2.2
  697. kernels, and became part of it in version 2.2.1 of the Linux kernel.
  698. 2.1 /proc/sys/fs - File system data
  699. -----------------------------------
  700. This subdirectory contains specific file system, file handle, inode, dentry
  701. and quota information.
  702. Currently, these files are in /proc/sys/fs:
  703. dentry-state
  704. ------------
  705. Status of the directory cache. Since directory entries are dynamically
  706. allocated and deallocated, this file indicates the current status. It holds
  707. six values, in which the last two are not used and are always zero. The others
  708. are listed in table 2-1.
  709. Table 2-1: Status files of the directory cache
  710. ..............................................................................
  711. File Content
  712. nr_dentry Almost always zero
  713. nr_unused Number of unused cache entries
  714. age_limit
  715. in seconds after the entry may be reclaimed, when memory is short
  716. want_pages internally
  717. ..............................................................................
  718. dquot-nr and dquot-max
  719. ----------------------
  720. The file dquot-max shows the maximum number of cached disk quota entries.
  721. The file dquot-nr shows the number of allocated disk quota entries and the
  722. number of free disk quota entries.
  723. If the number of available cached disk quotas is very low and you have a large
  724. number of simultaneous system users, you might want to raise the limit.
  725. file-nr and file-max
  726. --------------------
  727. The kernel allocates file handles dynamically, but doesn't free them again at
  728. this time.
  729. The value in file-max denotes the maximum number of file handles that the
  730. Linux kernel will allocate. When you get a lot of error messages about running
  731. out of file handles, you might want to raise this limit. The default value is
  732. 10% of RAM in kilobytes. To change it, just write the new number into the
  733. file:
  734. # cat /proc/sys/fs/file-max
  735. 4096
  736. # echo 8192 > /proc/sys/fs/file-max
  737. # cat /proc/sys/fs/file-max
  738. 8192
  739. This method of revision is useful for all customizable parameters of the
  740. kernel - simply echo the new value to the corresponding file.
  741. Historically, the three values in file-nr denoted the number of allocated file
  742. handles, the number of allocated but unused file handles, and the maximum
  743. number of file handles. Linux 2.6 always reports 0 as the number of free file
  744. handles -- this is not an error, it just means that the number of allocated
  745. file handles exactly matches the number of used file handles.
  746. Attempts to allocate more file descriptors than file-max are reported with
  747. printk, look for "VFS: file-max limit <number> reached".
  748. inode-state and inode-nr
  749. ------------------------
  750. The file inode-nr contains the first two items from inode-state, so we'll skip
  751. to that file...
  752. inode-state contains two actual numbers and five dummy values. The numbers
  753. are nr_inodes and nr_free_inodes (in order of appearance).
  754. nr_inodes
  755. ~~~~~~~~~
  756. Denotes the number of inodes the system has allocated. This number will
  757. grow and shrink dynamically.
  758. nr_free_inodes
  759. --------------
  760. Represents the number of free inodes. Ie. The number of inuse inodes is
  761. (nr_inodes - nr_free_inodes).
  762. aio-nr and aio-max-nr
  763. ---------------------
  764. aio-nr is the running total of the number of events specified on the
  765. io_setup system call for all currently active aio contexts. If aio-nr
  766. reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
  767. raising aio-max-nr does not result in the pre-allocation or re-sizing
  768. of any kernel data structures.
  769. 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
  770. -----------------------------------------------------------
  771. Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
  772. handles the kernel support for miscellaneous binary formats.
  773. Binfmt_misc provides the ability to register additional binary formats to the
  774. Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
  775. needs to know magic numbers at the beginning or the filename extension of the
  776. binary.
  777. It works by maintaining a linked list of structs that contain a description of
  778. a binary format, including a magic with size (or the filename extension),
  779. offset and mask, and the interpreter name. On request it invokes the given
  780. interpreter with the original program as argument, as binfmt_java and
  781. binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
  782. binary-formats, you have to register an additional binary-format.
  783. There are two general files in binfmt_misc and one file per registered format.
  784. The two general files are register and status.
  785. Registering a new binary format
  786. -------------------------------
  787. To register a new binary format you have to issue the command
  788. echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
  789. with appropriate name (the name for the /proc-dir entry), offset (defaults to
  790. 0, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
  791. last but not least, the interpreter that is to be invoked (for example and
  792. testing /bin/echo). Type can be M for usual magic matching or E for filename
  793. extension matching (give extension in place of magic).
  794. Check or reset the status of the binary format handler
  795. ------------------------------------------------------
  796. If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
  797. current status (enabled/disabled) of binfmt_misc. Change the status by echoing
  798. 0 (disables) or 1 (enables) or -1 (caution: this clears all previously
  799. registered binary formats) to status. For example echo 0 > status to disable
  800. binfmt_misc (temporarily).
  801. Status of a single handler
  802. --------------------------
  803. Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
  804. perform the same function as status, but their scope is limited to the actual
  805. binary format. By cating this file, you also receive all related information
  806. about the interpreter/magic of the binfmt.
  807. Example usage of binfmt_misc (emulate binfmt_java)
  808. --------------------------------------------------
  809. cd /proc/sys/fs/binfmt_misc
  810. echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
  811. echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
  812. echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
  813. echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
  814. These four lines add support for Java executables and Java applets (like
  815. binfmt_java, additionally recognizing the .html extension with no need to put
  816. <!--applet> to every applet file). You have to install the JDK and the
  817. shell-script /usr/local/java/bin/javawrapper too. It works around the
  818. brokenness of the Java filename handling. To add a Java binary, just create a
  819. link to the class-file somewhere in the path.
  820. 2.3 /proc/sys/kernel - general kernel parameters
  821. ------------------------------------------------
  822. This directory reflects general kernel behaviors. As I've said before, the
  823. contents depend on your configuration. Here you'll find the most important
  824. files, along with descriptions of what they mean and how to use them.
  825. acct
  826. ----
  827. The file contains three values; highwater, lowwater, and frequency.
  828. It exists only when BSD-style process accounting is enabled. These values
  829. control its behavior. If the free space on the file system where the log lives
  830. goes below lowwater percentage, accounting suspends. If it goes above
  831. highwater percentage, accounting resumes. Frequency determines how often you
  832. check the amount of free space (value is in seconds). Default settings are: 4,
  833. 2, and 30. That is, suspend accounting if there is less than 2 percent free;
  834. resume it if we have a value of 3 or more percent; consider information about
  835. the amount of free space valid for 30 seconds
  836. ctrl-alt-del
  837. ------------
  838. When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
  839. program to handle a graceful restart. However, when the value is greater that
  840. zero, Linux's reaction to this key combination will be an immediate reboot,
  841. without syncing its dirty buffers.
  842. [NOTE]
  843. When a program (like dosemu) has the keyboard in raw mode, the
  844. ctrl-alt-del is intercepted by the program before it ever reaches the
  845. kernel tty layer, and it is up to the program to decide what to do with
  846. it.
  847. domainname and hostname
  848. -----------------------
  849. These files can be controlled to set the NIS domainname and hostname of your
  850. box. For the classic darkstar.frop.org a simple:
  851. # echo "darkstar" > /proc/sys/kernel/hostname
  852. # echo "frop.org" > /proc/sys/kernel/domainname
  853. would suffice to set your hostname and NIS domainname.
  854. osrelease, ostype and version
  855. -----------------------------
  856. The names make it pretty obvious what these fields contain:
  857. > cat /proc/sys/kernel/osrelease
  858. 2.2.12
  859. > cat /proc/sys/kernel/ostype
  860. Linux
  861. > cat /proc/sys/kernel/version
  862. #4 Fri Oct 1 12:41:14 PDT 1999
  863. The files osrelease and ostype should be clear enough. Version needs a little
  864. more clarification. The #4 means that this is the 4th kernel built from this
  865. source base and the date after it indicates the time the kernel was built. The
  866. only way to tune these values is to rebuild the kernel.
  867. panic
  868. -----
  869. The value in this file represents the number of seconds the kernel waits
  870. before rebooting on a panic. When you use the software watchdog, the
  871. recommended setting is 60. If set to 0, the auto reboot after a kernel panic
  872. is disabled, which is the default setting.
  873. printk
  874. ------
  875. The four values in printk denote
  876. * console_loglevel,
  877. * default_message_loglevel,
  878. * minimum_console_loglevel and
  879. * default_console_loglevel
  880. respectively.
  881. These values influence printk() behavior when printing or logging error
  882. messages, which come from inside the kernel. See syslog(2) for more
  883. information on the different log levels.
  884. console_loglevel
  885. ----------------
  886. Messages with a higher priority than this will be printed to the console.
  887. default_message_level
  888. ---------------------
  889. Messages without an explicit priority will be printed with this priority.
  890. minimum_console_loglevel
  891. ------------------------
  892. Minimum (highest) value to which the console_loglevel can be set.
  893. default_console_loglevel
  894. ------------------------
  895. Default value for console_loglevel.
  896. sg-big-buff
  897. -----------
  898. This file shows the size of the generic SCSI (sg) buffer. At this point, you
  899. can't tune it yet, but you can change it at compile time by editing
  900. include/scsi/sg.h and changing the value of SG_BIG_BUFF.
  901. If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
  902. this to a higher value. Refer to the SANE documentation on this issue.
  903. modprobe
  904. --------
  905. The location where the modprobe binary is located. The kernel uses this
  906. program to load modules on demand.
  907. unknown_nmi_panic
  908. -----------------
  909. The value in this file affects behavior of handling NMI. When the value is
  910. non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
  911. debugging information is displayed on console.
  912. NMI switch that most IA32 servers have fires unknown NMI up, for example.
  913. If a system hangs up, try pressing the NMI switch.
  914. nmi_watchdog
  915. ------------
  916. Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
  917. the NMI watchdog is enabled and will continuously test all online cpus to
  918. determine whether or not they are still functioning properly.
  919. Because the NMI watchdog shares registers with oprofile, by disabling the NMI
  920. watchdog, oprofile may have more registers to utilize.
  921. 2.4 /proc/sys/vm - The virtual memory subsystem
  922. -----------------------------------------------
  923. The files in this directory can be used to tune the operation of the virtual
  924. memory (VM) subsystem of the Linux kernel.
  925. vfs_cache_pressure
  926. ------------------
  927. Controls the tendency of the kernel to reclaim the memory which is used for
  928. caching of directory and inode objects.
  929. At the default value of vfs_cache_pressure=100 the kernel will attempt to
  930. reclaim dentries and inodes at a "fair" rate with respect to pagecache and
  931. swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
  932. to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
  933. causes the kernel to prefer to reclaim dentries and inodes.
  934. dirty_background_ratio
  935. ----------------------
  936. Contains, as a percentage of total system memory, the number of pages at which
  937. the pdflush background writeback daemon will start writing out dirty data.
  938. dirty_ratio
  939. -----------------
  940. Contains, as a percentage of total system memory, the number of pages at which
  941. a process which is generating disk writes will itself start writing out dirty
  942. data.
  943. dirty_writeback_centisecs
  944. -------------------------
  945. The pdflush writeback daemons will periodically wake up and write `old' data
  946. out to disk. This tunable expresses the interval between those wakeups, in
  947. 100'ths of a second.
  948. Setting this to zero disables periodic writeback altogether.
  949. dirty_expire_centisecs
  950. ----------------------
  951. This tunable is used to define when dirty data is old enough to be eligible
  952. for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
  953. Data which has been dirty in-memory for longer than this interval will be
  954. written out next time a pdflush daemon wakes up.
  955. legacy_va_layout
  956. ----------------
  957. If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
  958. will use the legacy (2.4) layout for all processes.
  959. lower_zone_protection
  960. ---------------------
  961. For some specialised workloads on highmem machines it is dangerous for
  962. the kernel to allow process memory to be allocated from the "lowmem"
  963. zone. This is because that memory could then be pinned via the mlock()
  964. system call, or by unavailability of swapspace.
  965. And on large highmem machines this lack of reclaimable lowmem memory
  966. can be fatal.
  967. So the Linux page allocator has a mechanism which prevents allocations
  968. which _could_ use highmem from using too much lowmem. This means that
  969. a certain amount of lowmem is defended from the possibility of being
  970. captured into pinned user memory.
  971. (The same argument applies to the old 16 megabyte ISA DMA region. This
  972. mechanism will also defend that region from allocations which could use
  973. highmem or lowmem).
  974. The `lower_zone_protection' tunable determines how aggressive the kernel is
  975. in defending these lower zones. The default value is zero - no
  976. protection at all.
  977. If you have a machine which uses highmem or ISA DMA and your
  978. applications are using mlock(), or if you are running with no swap then
  979. you probably should increase the lower_zone_protection setting.
  980. The units of this tunable are fairly vague. It is approximately equal
  981. to "megabytes," so setting lower_zone_protection=100 will protect around 100
  982. megabytes of the lowmem zone from user allocations. It will also make
  983. those 100 megabytes unavailable for use by applications and by
  984. pagecache, so there is a cost.
  985. The effects of this tunable may be observed by monitoring
  986. /proc/meminfo:LowFree. Write a single huge file and observe the point
  987. at which LowFree ceases to fall.
  988. A reasonable value for lower_zone_protection is 100.
  989. page-cluster
  990. ------------
  991. page-cluster controls the number of pages which are written to swap in
  992. a single attempt. The swap I/O size.
  993. It is a logarithmic value - setting it to zero means "1 page", setting
  994. it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
  995. The default value is three (eight pages at a time). There may be some
  996. small benefits in tuning this to a different value if your workload is
  997. swap-intensive.
  998. overcommit_memory
  999. -----------------
  1000. Controls overcommit of system memory, possibly allowing processes
  1001. to allocate (but not use) more memory than is actually available.
  1002. 0 - Heuristic overcommit handling. Obvious overcommits of
  1003. address space are refused. Used for a typical system. It
  1004. ensures a seriously wild allocation fails while allowing
  1005. overcommit to reduce swap usage. root is allowed to
  1006. allocate slightly more memory in this mode. This is the
  1007. default.
  1008. 1 - Always overcommit. Appropriate for some scientific
  1009. applications.
  1010. 2 - Don't overcommit. The total address space commit
  1011. for the system is not permitted to exceed swap plus a
  1012. configurable percentage (default is 50) of physical RAM.
  1013. Depending on the percentage you use, in most situations
  1014. this means a process will not be killed while attempting
  1015. to use already-allocated memory but will receive errors
  1016. on memory allocation as appropriate.
  1017. overcommit_ratio
  1018. ----------------
  1019. Percentage of physical memory size to include in overcommit calculations
  1020. (see above.)
  1021. Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
  1022. swapspace = total size of all swap areas
  1023. physmem = size of physical memory in system
  1024. nr_hugepages and hugetlb_shm_group
  1025. ----------------------------------
  1026. nr_hugepages configures number of hugetlb page reserved for the system.
  1027. hugetlb_shm_group contains group id that is allowed to create SysV shared
  1028. memory segment using hugetlb page.
  1029. laptop_mode
  1030. -----------
  1031. laptop_mode is a knob that controls "laptop mode". All the things that are
  1032. controlled by this knob are discussed in Documentation/laptop-mode.txt.
  1033. block_dump
  1034. ----------
  1035. block_dump enables block I/O debugging when set to a nonzero value. More
  1036. information on block I/O debugging is in Documentation/laptop-mode.txt.
  1037. swap_token_timeout
  1038. ------------------
  1039. This file contains valid hold time of swap out protection token. The Linux
  1040. VM has token based thrashing control mechanism and uses the token to prevent
  1041. unnecessary page faults in thrashing situation. The unit of the value is
  1042. second. The value would be useful to tune thrashing behavior.
  1043. drop_caches
  1044. -----------
  1045. Writing to this will cause the kernel to drop clean caches, dentries and
  1046. inodes from memory, causing that memory to become free.
  1047. To free pagecache:
  1048. echo 1 > /proc/sys/vm/drop_caches
  1049. To free dentries and inodes:
  1050. echo 2 > /proc/sys/vm/drop_caches
  1051. To free pagecache, dentries and inodes:
  1052. echo 3 > /proc/sys/vm/drop_caches
  1053. As this is a non-destructive operation and dirty objects are not freeable, the
  1054. user should run `sync' first.
  1055. 2.5 /proc/sys/dev - Device specific parameters
  1056. ----------------------------------------------
  1057. Currently there is only support for CDROM drives, and for those, there is only
  1058. one read-only file containing information about the CD-ROM drives attached to
  1059. the system:
  1060. >cat /proc/sys/dev/cdrom/info
  1061. CD-ROM information, Id: cdrom.c 2.55 1999/04/25
  1062. drive name: sr0 hdb
  1063. drive speed: 32 40
  1064. drive # of slots: 1 0
  1065. Can close tray: 1 1
  1066. Can open tray: 1 1
  1067. Can lock tray: 1 1
  1068. Can change speed: 1 1
  1069. Can select disk: 0 1
  1070. Can read multisession: 1 1
  1071. Can read MCN: 1 1
  1072. Reports media changed: 1 1
  1073. Can play audio: 1 1
  1074. You see two drives, sr0 and hdb, along with a list of their features.
  1075. 2.6 /proc/sys/sunrpc - Remote procedure calls
  1076. ---------------------------------------------
  1077. This directory contains four files, which enable or disable debugging for the
  1078. RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
  1079. be set to one to turn debugging on. (The default value is 0 for each)
  1080. 2.7 /proc/sys/net - Networking stuff
  1081. ------------------------------------
  1082. The interface to the networking parts of the kernel is located in
  1083. /proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
  1084. some of them, depending on your kernel's configuration.
  1085. Table 2-3: Subdirectories in /proc/sys/net
  1086. ..............................................................................
  1087. Directory Content Directory Content
  1088. core General parameter appletalk Appletalk protocol
  1089. unix Unix domain sockets netrom NET/ROM
  1090. 802 E802 protocol ax25 AX25
  1091. ethernet Ethernet protocol rose X.25 PLP layer
  1092. ipv4 IP version 4 x25 X.25 protocol
  1093. ipx IPX token-ring IBM token ring
  1094. bridge Bridging decnet DEC net
  1095. ipv6 IP version 6
  1096. ..............................................................................
  1097. We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
  1098. only minor players in the Linux world, we'll skip them in this chapter. You'll
  1099. find some short info on Appletalk and IPX further on in this chapter. Review
  1100. the online documentation and the kernel source to get a detailed view of the
  1101. parameters for those protocols. In this section we'll discuss the
  1102. subdirectories printed in bold letters in the table above. As default values
  1103. are suitable for most needs, there is no need to change these values.
  1104. /proc/sys/net/core - Network core options
  1105. -----------------------------------------
  1106. rmem_default
  1107. ------------
  1108. The default setting of the socket receive buffer in bytes.
  1109. rmem_max
  1110. --------
  1111. The maximum receive socket buffer size in bytes.
  1112. wmem_default
  1113. ------------
  1114. The default setting (in bytes) of the socket send buffer.
  1115. wmem_max
  1116. --------
  1117. The maximum send socket buffer size in bytes.
  1118. message_burst and message_cost
  1119. ------------------------------
  1120. These parameters are used to limit the warning messages written to the kernel
  1121. log from the networking code. They enforce a rate limit to make a
  1122. denial-of-service attack impossible. A higher message_cost factor, results in
  1123. fewer messages that will be written. Message_burst controls when messages will
  1124. be dropped. The default settings limit warning messages to one every five
  1125. seconds.
  1126. netdev_max_backlog
  1127. ------------------
  1128. Maximum number of packets, queued on the INPUT side, when the interface
  1129. receives packets faster than kernel can process them.
  1130. optmem_max
  1131. ----------
  1132. Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
  1133. of struct cmsghdr structures with appended data.
  1134. /proc/sys/net/unix - Parameters for Unix domain sockets
  1135. -------------------------------------------------------
  1136. There are only two files in this subdirectory. They control the delays for
  1137. deleting and destroying socket descriptors.
  1138. 2.8 /proc/sys/net/ipv4 - IPV4 settings
  1139. --------------------------------------
  1140. IP version 4 is still the most used protocol in Unix networking. It will be
  1141. replaced by IP version 6 in the next couple of years, but for the moment it's
  1142. the de facto standard for the internet and is used in most networking
  1143. environments around the world. Because of the importance of this protocol,
  1144. we'll have a deeper look into the subtree controlling the behavior of the IPv4
  1145. subsystem of the Linux kernel.
  1146. Let's start with the entries in /proc/sys/net/ipv4.
  1147. ICMP settings
  1148. -------------
  1149. icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
  1150. ----------------------------------------------------
  1151. Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
  1152. just those to broadcast and multicast addresses.
  1153. Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
  1154. destination address your network may be used as an exploder for denial of
  1155. service packet flooding attacks to other hosts.
  1156. icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
  1157. ---------------------------------------------------------------------------------------
  1158. Sets limits for sending ICMP packets to specific targets. A value of zero
  1159. disables all limiting. Any positive value sets the maximum package rate in
  1160. hundredth of a second (on Intel systems).
  1161. IP settings
  1162. -----------
  1163. ip_autoconfig
  1164. -------------
  1165. This file contains the number one if the host received its IP configuration by
  1166. RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
  1167. ip_default_ttl
  1168. --------------
  1169. TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
  1170. hops a packet may travel.
  1171. ip_dynaddr
  1172. ----------
  1173. Enable dynamic socket address rewriting on interface address change. This is
  1174. useful for dialup interface with changing IP addresses.
  1175. ip_forward
  1176. ----------
  1177. Enable or disable forwarding of IP packages between interfaces. Changing this
  1178. value resets all other parameters to their default values. They differ if the
  1179. kernel is configured as host or router.
  1180. ip_local_port_range
  1181. -------------------
  1182. Range of ports used by TCP and UDP to choose the local port. Contains two
  1183. numbers, the first number is the lowest port, the second number the highest
  1184. local port. Default is 1024-4999. Should be changed to 32768-61000 for
  1185. high-usage systems.
  1186. ip_no_pmtu_disc
  1187. ---------------
  1188. Global switch to turn path MTU discovery off. It can also be set on a per
  1189. socket basis by the applications or on a per route basis.
  1190. ip_masq_debug
  1191. -------------
  1192. Enable/disable debugging of IP masquerading.
  1193. IP fragmentation settings
  1194. -------------------------
  1195. ipfrag_high_trash and ipfrag_low_trash
  1196. --------------------------------------
  1197. Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
  1198. of memory is allocated for this purpose, the fragment handler will toss
  1199. packets until ipfrag_low_thresh is reached.
  1200. ipfrag_time
  1201. -----------
  1202. Time in seconds to keep an IP fragment in memory.
  1203. TCP settings
  1204. ------------
  1205. tcp_ecn
  1206. -------
  1207. This file controls the use of the ECN bit in the IPv4 headers. This is a new
  1208. feature about Explicit Congestion Notification, but some routers and firewalls
  1209. block traffic that has this bit set, so it could be necessary to echo 0 to
  1210. /proc/sys/net/ipv4/tcp_ecn if you want to talk to these sites. For more info
  1211. you could read RFC2481.
  1212. tcp_retrans_collapse
  1213. --------------------
  1214. Bug-to-bug compatibility with some broken printers. On retransmit, try to send
  1215. larger packets to work around bugs in certain TCP stacks. Can be turned off by
  1216. setting it to zero.
  1217. tcp_keepalive_probes
  1218. --------------------
  1219. Number of keep alive probes TCP sends out, until it decides that the
  1220. connection is broken.
  1221. tcp_keepalive_time
  1222. ------------------
  1223. How often TCP sends out keep alive messages, when keep alive is enabled. The
  1224. default is 2 hours.
  1225. tcp_syn_retries
  1226. ---------------
  1227. Number of times initial SYNs for a TCP connection attempt will be
  1228. retransmitted. Should not be higher than 255. This is only the timeout for
  1229. outgoing connections, for incoming connections the number of retransmits is
  1230. defined by tcp_retries1.
  1231. tcp_sack
  1232. --------
  1233. Enable select acknowledgments after RFC2018.
  1234. tcp_timestamps
  1235. --------------
  1236. Enable timestamps as defined in RFC1323.
  1237. tcp_stdurg
  1238. ----------
  1239. Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
  1240. default is to use the BSD compatible interpretation of the urgent pointer
  1241. pointing to the first byte after the urgent data. The RFC793 interpretation is
  1242. to have it point to the last byte of urgent data. Enabling this option may
  1243. lead to interoperability problems. Disabled by default.
  1244. tcp_syncookies
  1245. --------------
  1246. Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
  1247. syncookies when the syn backlog queue of a socket overflows. This is to ward
  1248. off the common 'syn flood attack'. Disabled by default.
  1249. Note that the concept of a socket backlog is abandoned. This means the peer
  1250. may not receive reliable error messages from an over loaded server with
  1251. syncookies enabled.
  1252. tcp_window_scaling
  1253. ------------------
  1254. Enable window scaling as defined in RFC1323.
  1255. tcp_fin_timeout
  1256. ---------------
  1257. The length of time in seconds it takes to receive a final FIN before the
  1258. socket is always closed. This is strictly a violation of the TCP
  1259. specification, but required to prevent denial-of-service attacks.
  1260. tcp_max_ka_probes
  1261. -----------------
  1262. Indicates how many keep alive probes are sent per slow timer run. Should not
  1263. be set too high to prevent bursts.
  1264. tcp_max_syn_backlog
  1265. -------------------
  1266. Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
  1267. in listen(2) only specifies the length of the backlog queue of already
  1268. established sockets. When more connection requests arrive Linux starts to drop
  1269. packets. When syncookies are enabled the packets are still answered and the
  1270. maximum queue is effectively ignored.
  1271. tcp_retries1
  1272. ------------
  1273. Defines how often an answer to a TCP connection request is retransmitted
  1274. before giving up.
  1275. tcp_retries2
  1276. ------------
  1277. Defines how often a TCP packet is retransmitted before giving up.
  1278. Interface specific settings
  1279. ---------------------------
  1280. In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
  1281. interface the system knows about and one directory calls all. Changes in the
  1282. all subdirectory affect all interfaces, whereas changes in the other
  1283. subdirectories affect only one interface. All directories have the same
  1284. entries:
  1285. accept_redirects
  1286. ----------------
  1287. This switch decides if the kernel accepts ICMP redirect messages or not. The
  1288. default is 'yes' if the kernel is configured for a regular host and 'no' for a
  1289. router configuration.
  1290. accept_source_route
  1291. -------------------
  1292. Should source routed packages be accepted or declined. The default is
  1293. dependent on the kernel configuration. It's 'yes' for routers and 'no' for
  1294. hosts.
  1295. bootp_relay
  1296. ~~~~~~~~~~~
  1297. Accept packets with source address 0.b.c.d with destinations not to this host
  1298. as local ones. It is supposed that a BOOTP relay daemon will catch and forward
  1299. such packets.
  1300. The default is 0, since this feature is not implemented yet (kernel version
  1301. 2.2.12).
  1302. forwarding
  1303. ----------
  1304. Enable or disable IP forwarding on this interface.
  1305. log_martians
  1306. ------------
  1307. Log packets with source addresses with no known route to kernel log.
  1308. mc_forwarding
  1309. -------------
  1310. Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
  1311. multicast routing daemon is required.
  1312. proxy_arp
  1313. ---------
  1314. Does (1) or does not (0) perform proxy ARP.
  1315. rp_filter
  1316. ---------
  1317. Integer value determines if a source validation should be made. 1 means yes, 0
  1318. means no. Disabled by default, but local/broadcast address spoofing is always
  1319. on.
  1320. If you set this to 1 on a router that is the only connection for a network to
  1321. the net, it will prevent spoofing attacks against your internal networks
  1322. (external addresses can still be spoofed), without the need for additional
  1323. firewall rules.
  1324. secure_redirects
  1325. ----------------
  1326. Accept ICMP redirect messages only for gateways, listed in default gateway
  1327. list. Enabled by default.
  1328. shared_media
  1329. ------------
  1330. If it is not set the kernel does not assume that different subnets on this
  1331. device can communicate directly. Default setting is 'yes'.
  1332. send_redirects
  1333. --------------
  1334. Determines whether to send ICMP redirects to other hosts.
  1335. Routing settings
  1336. ----------------
  1337. The directory /proc/sys/net/ipv4/route contains several file to control
  1338. routing issues.
  1339. error_burst and error_cost
  1340. --------------------------
  1341. These parameters are used to limit how many ICMP destination unreachable to
  1342. send from the host in question. ICMP destination unreachable messages are
  1343. sent when we cannot reach the next hop while trying to transmit a packet.
  1344. It will also print some error messages to kernel logs if someone is ignoring
  1345. our ICMP redirects. The higher the error_cost factor is, the fewer
  1346. destination unreachable and error messages will be let through. Error_burst
  1347. controls when destination unreachable messages and error messages will be
  1348. dropped. The default settings limit warning messages to five every second.
  1349. flush
  1350. -----
  1351. Writing to this file results in a flush of the routing cache.
  1352. gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
  1353. ---------------------------------------------------------------------
  1354. Values to control the frequency and behavior of the garbage collection
  1355. algorithm for the routing cache. gc_min_interval is deprecated and replaced
  1356. by gc_min_interval_ms.
  1357. max_size
  1358. --------
  1359. Maximum size of the routing cache. Old entries will be purged once the cache
  1360. reached has this size.
  1361. max_delay, min_delay
  1362. --------------------
  1363. Delays for flushing the routing cache.
  1364. redirect_load, redirect_number
  1365. ------------------------------
  1366. Factors which determine if more ICPM redirects should be sent to a specific
  1367. host. No redirects will be sent once the load limit or the maximum number of
  1368. redirects has been reached.
  1369. redirect_silence
  1370. ----------------
  1371. Timeout for redirects. After this period redirects will be sent again, even if
  1372. this has been stopped, because the load or number limit has been reached.
  1373. Network Neighbor handling
  1374. -------------------------
  1375. Settings about how to handle connections with direct neighbors (nodes attached
  1376. to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
  1377. As we saw it in the conf directory, there is a default subdirectory which
  1378. holds the default values, and one directory for each interface. The contents
  1379. of the directories are identical, with the single exception that the default
  1380. settings contain additional options to set garbage collection parameters.
  1381. In the interface directories you'll find the following entries:
  1382. base_reachable_time, base_reachable_time_ms
  1383. -------------------------------------------
  1384. A base value used for computing the random reachable time value as specified
  1385. in RFC2461.
  1386. Expression of base_reachable_time, which is deprecated, is in seconds.
  1387. Expression of base_reachable_time_ms is in milliseconds.
  1388. retrans_time, retrans_time_ms
  1389. -----------------------------
  1390. The time between retransmitted Neighbor Solicitation messages.
  1391. Used for address resolution and to determine if a neighbor is
  1392. unreachable.
  1393. Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
  1394. IPv4) or in jiffies (for IPv6).
  1395. Expression of retrans_time_ms is in milliseconds.
  1396. unres_qlen
  1397. ----------
  1398. Maximum queue length for a pending arp request - the number of packets which
  1399. are accepted from other layers while the ARP address is still resolved.
  1400. anycast_delay
  1401. -------------
  1402. Maximum for random delay of answers to neighbor solicitation messages in
  1403. jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
  1404. yet).
  1405. ucast_solicit
  1406. -------------
  1407. Maximum number of retries for unicast solicitation.
  1408. mcast_solicit
  1409. -------------
  1410. Maximum number of retries for multicast solicitation.
  1411. delay_first_probe_time
  1412. ----------------------
  1413. Delay for the first time probe if the neighbor is reachable. (see
  1414. gc_stale_time)
  1415. locktime
  1416. --------
  1417. An ARP/neighbor entry is only replaced with a new one if the old is at least
  1418. locktime old. This prevents ARP cache thrashing.
  1419. proxy_delay
  1420. -----------
  1421. Maximum time (real time is random [0..proxytime]) before answering to an ARP
  1422. request for which we have an proxy ARP entry. In some cases, this is used to
  1423. prevent network flooding.
  1424. proxy_qlen
  1425. ----------
  1426. Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
  1427. app_solicit
  1428. ----------
  1429. Determines the number of requests to send to the user level ARP daemon. Use 0
  1430. to turn off.
  1431. gc_stale_time
  1432. -------------
  1433. Determines how often to check for stale ARP entries. After an ARP entry is
  1434. stale it will be resolved again (which is useful when an IP address migrates
  1435. to another machine). When ucast_solicit is greater than 0 it first tries to
  1436. send an ARP packet directly to the known host When that fails and
  1437. mcast_solicit is greater than 0, an ARP request is broadcasted.
  1438. 2.9 Appletalk
  1439. -------------
  1440. The /proc/sys/net/appletalk directory holds the Appletalk configuration data
  1441. when Appletalk is loaded. The configurable parameters are:
  1442. aarp-expiry-time
  1443. ----------------
  1444. The amount of time we keep an ARP entry before expiring it. Used to age out
  1445. old hosts.
  1446. aarp-resolve-time
  1447. -----------------
  1448. The amount of time we will spend trying to resolve an Appletalk address.
  1449. aarp-retransmit-limit
  1450. ---------------------
  1451. The number of times we will retransmit a query before giving up.
  1452. aarp-tick-time
  1453. --------------
  1454. Controls the rate at which expires are checked.
  1455. The directory /proc/net/appletalk holds the list of active Appletalk sockets
  1456. on a machine.
  1457. The fields indicate the DDP type, the local address (in network:node format)
  1458. the remote address, the size of the transmit pending queue, the size of the
  1459. received queue (bytes waiting for applications to read) the state and the uid
  1460. owning the socket.
  1461. /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
  1462. shows the name of the interface, its Appletalk address, the network range on
  1463. that address (or network number for phase 1 networks), and the status of the
  1464. interface.
  1465. /proc/net/atalk_route lists each known network route. It lists the target
  1466. (network) that the route leads to, the router (may be directly connected), the
  1467. route flags, and the device the route is using.
  1468. 2.10 IPX
  1469. --------
  1470. The IPX protocol has no tunable values in proc/sys/net.
  1471. The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
  1472. socket giving the local and remote addresses in Novell format (that is
  1473. network:node:port). In accordance with the strange Novell tradition,
  1474. everything but the port is in hex. Not_Connected is displayed for sockets that
  1475. are not tied to a specific remote address. The Tx and Rx queue sizes indicate
  1476. the number of bytes pending for transmission and reception. The state
  1477. indicates the state the socket is in and the uid is the owning uid of the
  1478. socket.
  1479. The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
  1480. it gives the network number, the node number, and indicates if the network is
  1481. the primary network. It also indicates which device it is bound to (or
  1482. Internal for internal networks) and the Frame Type if appropriate. Linux
  1483. supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
  1484. IPX.
  1485. The /proc/net/ipx_route table holds a list of IPX routes. For each route it
  1486. gives the destination network, the router node (or Directly) and the network
  1487. address of the router (or Connected) for internal networks.
  1488. 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
  1489. ----------------------------------------------------------
  1490. The "mqueue" filesystem provides the necessary kernel features to enable the
  1491. creation of a user space library that implements the POSIX message queues
  1492. API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
  1493. Interfaces specification.)
  1494. The "mqueue" filesystem contains values for determining/setting the amount of
  1495. resources used by the file system.
  1496. /proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
  1497. maximum number of message queues allowed on the system.
  1498. /proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
  1499. maximum number of messages in a queue value. In fact it is the limiting value
  1500. for another (user) limit which is set in mq_open invocation. This attribute of
  1501. a queue must be less or equal then msg_max.
  1502. /proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
  1503. maximum message size value (it is every message queue's attribute set during
  1504. its creation).
  1505. 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
  1506. ------------------------------------------------------
  1507. This file can be used to adjust the score used to select which processes
  1508. should be killed in an out-of-memory situation. Giving it a high score will
  1509. increase the likelihood of this process being killed by the oom-killer. Valid
  1510. values are in the range -16 to +15, plus the special value -17, which disables
  1511. oom-killing altogether for this process.
  1512. 2.13 /proc/<pid>/oom_score - Display current oom-killer score
  1513. -------------------------------------------------------------
  1514. ------------------------------------------------------------------------------
  1515. This file can be used to check the current score used by the oom-killer is for
  1516. any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
  1517. process should be killed in an out-of-memory situation.
  1518. ------------------------------------------------------------------------------
  1519. Summary
  1520. ------------------------------------------------------------------------------
  1521. Certain aspects of kernel behavior can be modified at runtime, without the
  1522. need to recompile the kernel, or even to reboot the system. The files in the
  1523. /proc/sys tree can not only be read, but also modified. You can use the echo
  1524. command to write value into these files, thereby changing the default settings
  1525. of the kernel.
  1526. ------------------------------------------------------------------------------
  1527. 2.14 /proc/<pid>/io - Display the IO accounting fields
  1528. -------------------------------------------------------
  1529. This file contains IO statistics for each running process
  1530. Example
  1531. -------
  1532. test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
  1533. [1] 3828
  1534. test:/tmp # cat /proc/3828/io
  1535. rchar: 323934931
  1536. wchar: 323929600
  1537. syscr: 632687
  1538. syscw: 632675
  1539. read_bytes: 0
  1540. write_bytes: 323932160
  1541. cancelled_write_bytes: 0
  1542. Description
  1543. -----------
  1544. rchar
  1545. -----
  1546. I/O counter: chars read
  1547. The number of bytes which this task has caused to be read from storage. This
  1548. is simply the sum of bytes which this process passed to read() and pread().
  1549. It includes things like tty IO and it is unaffected by whether or not actual
  1550. physical disk IO was required (the read might have been satisfied from
  1551. pagecache)
  1552. wchar
  1553. -----
  1554. I/O counter: chars written
  1555. The number of bytes which this task has caused, or shall cause to be written
  1556. to disk. Similar caveats apply here as with rchar.
  1557. syscr
  1558. -----
  1559. I/O counter: read syscalls
  1560. Attempt to count the number of read I/O operations, i.e. syscalls like read()
  1561. and pread().
  1562. syscw
  1563. -----
  1564. I/O counter: write syscalls
  1565. Attempt to count the number of write I/O operations, i.e. syscalls like
  1566. write() and pwrite().
  1567. read_bytes
  1568. ----------
  1569. I/O counter: bytes read
  1570. Attempt to count the number of bytes which this process really did cause to
  1571. be fetched from the storage layer. Done at the submit_bio() level, so it is
  1572. accurate for block-backed filesystems. <please add status regarding NFS and
  1573. CIFS at a later time>
  1574. write_bytes
  1575. -----------
  1576. I/O counter: bytes written
  1577. Attempt to count the number of bytes which this process caused to be sent to
  1578. the storage layer. This is done at page-dirtying time.
  1579. cancelled_write_bytes
  1580. ---------------------
  1581. The big inaccuracy here is truncate. If a process writes 1MB to a file and
  1582. then deletes the file, it will in fact perform no writeout. But it will have
  1583. been accounted as having caused 1MB of write.
  1584. In other words: The number of bytes which this process caused to not happen,
  1585. by truncating pagecache. A task can cause "negative" IO too. If this task
  1586. truncates some dirty pagecache, some IO which another task has been accounted
  1587. for (in it's write_bytes) will not be happening. We _could_ just subtract that
  1588. from the truncating task's write_bytes, but there is information loss in doing
  1589. that.
  1590. Note
  1591. ----
  1592. At its current implementation state, this is a bit racy on 32-bit machines: if
  1593. process A reads process B's /proc/pid/io while process B is updating one of
  1594. those 64-bit counters, process A could see an intermediate result.
  1595. More information about this can be found within the taskstats documentation in
  1596. Documentation/accounting.
  1597. ------------------------------------------------------------------------------