API tools faq
paste
Advertisement
Malsasa

#bash Masalah Truncated History

Malsasa
Aug 26th, 2014
50
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
mIRC 12.02 KB | None | 0 0
raw download clone embed print report
  1. #bash pada Tuesday, August 26, 2014 09:38 PM
  2. Mengenai cara mencegah bash history yang selalu terpotong
  3. =========================================================
  4.  
  5. [21:07:35] <Malsasa> I found my bash history always truncated into 500 lines again, after I open mode than 3 bash instance (3 Terminal Emulators concurrently). My .bashrc code for keeping the history is here http://paste.ubuntu.com/8145842/. I have specified histappend, make HISTSIZE big, using history -a -r, etc. but the history still truncated. How to make the bash history eternal, for all bash instance in any number Terminal window opened in one time? I need it, it is so important for me (tracking error, etc). Thanks.
  6. [21:07:36] <osse> the advantage of my method is that if the number of files is not equal to 55 files you will gzip a number of files unequal to 45, and if there are fewer than 10 files you get thousands of usage texts
  7. [21:07:57] <Riviera> Malsasa: I believe the FAQ addresses that.
  8. [21:08:04] <-- Naib (~j@fu/hw/naib) has quit (Quit: Nothing to see here)
  9. [21:08:07] <osse> !faq history > Malsasa
  10. [21:08:08] <greybot> Malsasa: http://mywiki.wooledge.org/BashFAQ/051 -- I want history-search just like in tcsh. How can I bind it to the up and down keys?
  11. [21:08:09] <Malsasa> Riviera: I will see it.
  12. [21:08:13] <osse> ooops
  13. [21:08:15] <osse> !faq history > lose
  14. [21:08:15] <greybot> lose: http://mywiki.wooledge.org/BashFAQ/051 -- I want history-search just like in tcsh. How can I bind it to the up and down keys?
  15. [21:08:16] <geirha> Malsasa: HISTSIZE wraps at 2^32 iirc
  16. [21:08:21] --> tandroid (~Adium@50-201-154-94-static.hfc.comcastbusiness.net) has joined #bash
  17. [21:08:23] <osse> !faq lose > Malsasa
  18. [21:08:24] <greybot> osse: No matches found at http://mywiki.wooledge.org/BashFAQ
  19. [21:08:24] <Aldaron> geirha: yeah, it's retarded because a retarded program needs to undestand it :)
  20. [21:08:53] <Aldaron> geirha: thanks for mentioning that, because I do get trouble when the month changes
  21. [21:08:53] <Riviera> Would of course be totally unretarded to have filenames that follow the limitations of a crappy shell.
  22. [21:09:04] <osse> !faq losing > Malsasa
  23. [21:09:05] <greybot> Malsasa: http://mywiki.wooledge.org/BashFAQ/088 -- How can I avoid losing any history lines?
  24. [21:09:07] <geirha> Malsasa: If you have bash 4.3, you can set them to -1 to get infinite history
  25. [21:09:15] --> Naib (~j@5751e858.skybroadband.com) has joined #bash
  26. [21:09:15] <-- Naib (~j@5751e858.skybroadband.com) has quit (Changing host)
  27. [21:09:15] --> Naib (~j@fu/hw/naib) has joined #bash
  28. [21:09:18] <-- Naib (~j@fu/hw/naib) has quit (Client Quit)
  29. [21:09:24] <geirha> on older versions, the highest number you can set is $((2**31 - 1))
  30. [21:09:42] <Malsasa> osse: I am going.
  31. [21:09:53] --> Naib (~j@5751e858.skybroadband.com) has joined #bash
  32. [21:09:53] <-- Naib (~j@5751e858.skybroadband.com) has quit (Changing host)
  33. [21:09:53] --> Naib (~j@fu/hw/naib) has joined #bash
  34. [21:10:18] <osse> $((2**32 - 1)) you mean?
  35. [21:10:41] <geirha> osse: I don't remember if it's signed or unsigned. Best go 31 to be on the safe side
  36. [21:10:58] <geirha> Malsasa: Also, I recommend changing HISTFILE as well. That avoids running bash --norc truncating it to 500 lines again
  37. [21:11:15] <geirha> HISTFILE=~/.bash_history.$HOSTNAME  for instance
  38. [21:11:18] <osse> geirha: isn't the max 64 bits or something? Not for HISTFILE, but for $(())
  39. [21:11:32] <osse> # echo $((2**50))
  40. [21:11:33] <shbot> osse: 1125899906842624
  41. [21:11:56] <geirha> osse: yes, but HISTSIZE uses 32 bit int internally
  42. [21:12:07] <osse> ah
  43. [21:12:17] --> jasondotstar (~jasondots@2001:420:2280:1310:942e:7355:7ad4:7387) has joined #bash
  44. [21:12:50] <geirha> (( HISTFILESIZE = HISTSIZE = 2**31 - 1 ))
  45. [21:13:04] <-- darkbasic (~quassel@niko.linuxsystems.it) has quit (Quit: No Ping reply in 180 seconds.)
  46. [21:13:14] <Malsasa> geirha: thank you. But my HISTFILE was HISTFILE=~/.bash_history. Is it okay?
  47. [21:13:26] --> darkbasic (~quassel@niko.linuxsystems.it) has joined #bash
  48. [21:13:45] <geirha> Malsasa: that's the default, so it won't guard against bash --norc
  49. [21:14:00] <Savemech> osse: geirha:ty guys now itsworks really fine, now i can sleep well http://i.imgur.com/nCShuOl.png
  50. [21:14:12] <Savemech> Riviera:
  51. [21:14:38] <geirha> heh
  52. [21:14:44] <lhunath> Malsasa: by guard he means, if you ever start a default shell, it will wipe your history clean on exit.
  53. [21:14:57] --> elkorn (~elkorn@188.121.21.132) has joined #bash
  54. [21:14:59] --> somerandomnick12 (~a@gatek.thls.bbc.co.uk) has joined #bash
  55. [21:15:06] --> f-a (~f-a@nl1x.mullvad.net) has joined #bash
  56. [21:15:12] <-- Whiskey (whiskey@31.44.232.185) has quit (Ping timeout: 272 seconds)
  57. [21:15:13] <Malsasa> geirha: ooooow. Okay. I will change it.
  58. [21:15:30] <Malsasa> lhunath: thank you.
  59. [21:15:43] <f-a> so, due to me battling with bash I have not a file called --exclude=speccy . How to remove this?
  60. [21:15:53] --> jottr (~jottr@unaffiliated/jottr) has joined #bash
  61. [21:15:54] <geirha> well, wipe it down to the 500 line default
  62. [21:15:56] <-- koderok (~koderok@14.141.153.182) has quit (Quit: koderok)
  63. [21:16:14] <kerframil> f-a: rm -- <filename>
  64. [21:16:53] <f-a> thanks, that worked
  65. [21:17:03] <-- somerandomnick12 (~a@gatek.thls.bbc.co.uk) has quit (Client Quit)
  66. [21:17:29] <Malsasa> geirha: lhunath: osse: thank you. I have changed my bash history file name on .bashrc. I wanna back here soon after seeing the result.
  67. [21:17:47] <_val_> f-a: in case you're interested rm ./--exclude=speccy   works too
  68. [21:18:06] <_val_> what the heck is a speccy?
  69. [21:18:11] <geirha> Malsasa: Don't forget to change the HISTSIZE and HISTFILESIZE values too
  70. [21:18:17] <f-a> a folder with, 'lo and behold, spectrum games
  71. [21:18:20] <f-a> a lot of them
  72. [21:18:27] <Malsasa> geirha: yes. I have specified big big size for them.
  73. [21:18:51] <geirha> Malsasa: make sure it does not exceed 2^31-1
  74. [21:18:57] <geirha> (which your original did)
  75. [21:19:05] <Malsasa> geirha: ow. Okay, okay. Now I understand.
  76. [21:19:38] --> aquinas_ (~aquinas_@unaffiliated/aquinas) has joined #bash
  77. [21:19:52] --> somerandomnick12 (~a@gatek.thls.bbc.co.uk) has joined #bash
  78. [21:20:54] --> lpax (~lpax@unaffiliated/lpax) has joined #bash
  79. [21:20:58] <Malsasa> geirha: OOT: is this channel have a log? I have googled just a few seconds ago.
  80. [21:21:00] <geirha> Malsasa: Oh and don't export variables like HISTCONTROL and PROMPT_COMMAND
  81. [21:21:16] --> fork (~fork@unaffiliated/milefork) has joined #bash
  82. [21:21:20] <geirha> no official logs
  83. [21:21:33] <-- kahen (~infenwe@0118200527.0.fullrate.dk) has quit (Ping timeout: 250 seconds)
  84. [21:22:00] <-- Aldaron (1000@nblzone-228-25.nblnetworks.fi) has left #bash
  85. [21:22:18] --> sqlnoob (~techgaun@113.199.170.206) has joined #bash
  86. [21:22:55] --> semeion (~semeion@unaffiliated/semeion) has joined #bash
  87. [21:23:05] --> oxhak (~OxHaK@ns330132.ip-37-59-37.eu) has joined #bash
  88. [21:23:47] --> fork|2 (~fork@unaffiliated/milefork) has joined #bash
  89. [21:24:06] <-- semeion (~semeion@unaffiliated/semeion) has quit (Client Quit)
  90. [21:25:24] --> dennisw (~dennis@128.42.164.155) has joined #bash
  91. [21:26:01] <-- fork (~fork@unaffiliated/milefork) has quit (Ping timeout: 245 seconds)
  92. [21:26:13] --> semeion (~semeion@unaffiliated/semeion) has joined #bash
  93. [21:27:51] --> ntz (~jdoe@static-84-42-228-122.net.upcbroadband.cz) has joined #bash
  94. [21:27:53] <ntz> hello
  95. [21:28:04] <-- f-a (~f-a@nl1x.mullvad.net) has left #bash ("bai")
  96. [21:28:21] <ntz> how am I supposed to check in bash if the perms are exactly 600 and owner $EUID ?
  97. [21:28:42] <ntz> i can realize many ugly ways with sed and so
  98. [21:28:50] <geirha> chmod 600 file && echo "permissions are right"
  99. [21:29:16] <ntz> I need to check if the file has already that perms
  100. [21:29:21] <-- TomyWork (~tomy@85.199.84.100) has quit (Ping timeout: 250 seconds)
  101. [21:29:23] <ntz> I don't want to touch that file ..
  102. [21:29:32] <Jetchisel> find?
  103. [21:29:42] <ntz> yeah, that's one of "nasty"ways
  104. [21:29:46] <capt_mercaptan> geirha: lol
  105. [21:29:54] <_val_> UsingFind has some xamples I think.
  106. [21:29:57] <ntz> but probably find is good
  107. [21:29:58] <capt_mercaptan> ntz: stat
  108. [21:29:59] <geirha> or stat(1) if your system has one
  109. [21:30:02] <_val_> s/xamples/examples.
  110. [21:30:08] --> Alex-SF (~Alex-SF@71.198.230.165) has joined #bash
  111. [21:30:29] --> ScriptFanix (vincent@Jedrik.DMZ.riquer.fr) has joined #bash
  112. [21:30:33] <-- PotatoGim^Home (~potatogim@211.115.242.166) has left #bash ("전 이만 갑니다.")
  113. [21:30:38] <geirha> The find would be something like  [[ $(find file -perm 600 -print) ]]
  114. [21:30:44] <-- dreamon (~dreamon@unaffiliated/dreamon) has quit (Ping timeout: 240 seconds)
  115. [21:30:46] --> Nirgali42 (~dmart@64.132.167.18) has joined #bash
  116. [21:31:12] <geirha> silly find doesn't have a useful exit status, so you're left with testing output
  117. [21:31:12] <capt_mercaptan> find would be easier than stat
  118. [21:31:38] --> avostrik (~avostrik@178-16-155-142.obit.ru) has joined #bash
  119. [21:31:53] <capt_mercaptan> with stat, you'd get the permissions as a string (preferably in octal) and then have to string-compare with "0600"
  120. [21:32:04] <geirha> [[ $(find file -user "$EUID" -perm 600 -print) ]]
  121. [21:32:20] --> ggherdov (sid11402@gateway/web/irccloud.com/x-zeadfhvhwvzcpqpa) has joined #bash
  122. [21:32:28] <-- danemacmillan (~danemacmi@modemcable206.143-37-24.static.videotron.ca) has quit (Quit: WeeChat 1.0)
  123. [21:32:31] <ntz> Jetchisel: find /tmp/foo -perm 0600 -uid $EUID || something_else()
  124. [21:32:38] --> danemacmillan (~danemacmi@modemcable206.143-37-24.static.videotron.ca) has joined #bash
  125. [21:32:40] <ntz> yes, thanks .. that's all-in-one and good
  126. [21:32:57] --> magyar (~magyar@76.21.119.81) has joined #bash
  127. [21:33:00] <geirha> ntz: no. find doesn't have a useful exit status
  128. [21:33:02] <_val_> ntz: did you read the above thing? Find has no exit status.
  129. [21:33:05] <-- magyar (~magyar@76.21.119.81) has quit (Changing host)
  130. [21:33:05] --> magyar (~magyar@unaffiliated/magyar) has joined #bash
  131. [21:33:06] <-- Dary (~Dary@111.161.114.122) has quit (Ping timeout: 245 seconds)
  132. [21:33:10] <lhunath> -exec + can control find's exit code.
  133. [21:33:13] <ntz> it works
  134. [21:33:24] <geirha> and -user is posix, -uid is not
  135. [21:33:26] <ntz> i tested it with changing perms and owner ..but thanks
  136. [21:33:39] <osse> _val_: *no* exit status? So if you run find and then echo $? the shell SIGSEGVs? :O:O:O
  137. [21:33:47] --> PotatoGim^Laptop (~potatogim@211.115.242.166) has joined #bash
  138. [21:34:00] --> gargola (quassel@nat/hp/x-rrpumapugqbxtcbp) has joined #bash
  139. [21:34:34] <-- PotatoGim^Laptop (~potatogim@211.115.242.166) has left #bash
  140. [21:34:42] <Malsasa> geirha: okay, thank you. I have changed it.
  141. [21:35:01] <lhunath> and although I can't find a reference for it in the spec, all finds I have return 1 when the file doesn't exist.
  142. [21:35:04] <ntz> geirha: I know that -user is posix .. doing it on solaris
  143. [21:35:09] <capt_mercaptan> Can I use bash to randomize the instruction pointer's contents?
  144. [21:35:20] <geirha> ntz: Well, it won't work with GNU find at least
  145. [21:35:21] <ggherdov> hi. Is there anything to "decolorize" outputs ? I have some scripts that where written with a TTY in mind, and output fancy colors all around. Now I am running them headless-ly (automatic testing and the like) and I fetch their stdout. Those color codes are really disturbing.
  146. [21:35:21] <ggherdov> Any program I can pipe that stuff into, to get rid of colors?
  147. [21:35:24] <-- mdih (~dvillamer@112.210.72.68) has quit (Ping timeout: 260 seconds)
  148. [21:35:26] <ntz> it does
  149. [21:35:26] <lhunath> the best reference is that find Idontexist doesn't "All path operands were traversed successfully."
  150. [21:35:38] <ntz> i tested it right now on linux and solaris .. works reasonably
  151. [21:35:54] <capt_mercaptan> ggherdov: a sed script can be used to remove codes
  152. [21:36:07] <geirha> # touch file; chmod 666 file; find ./file -uid 0 -perm 600; echo $?
  153. [21:36:09] <shbot> geirha: 0
  154. [21:36:22] <geirha> ^ returns success even though the file has mode 666
  155. [21:36:34] <lhunath> ggherdov: TERM=dumb ?
  156. [21:36:48] <capt_mercaptan> ggherdov: you'd have to be more specific about what codes you're actually getting, like, are they ANSI escapes or terminal control code
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!