| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10
Comments:
<0> alekzandr: http://centos.karan.org/
<1> gambolputty: wait it out
<2> all ssh operations fail
<2> ok
<1> how much ram, and how many 'kernel*' packages are installed?
<3> gambolputty: please put the output of "cat /etc/cpuinfo" into a paste bin
<0> alekzandr: actually nevermind. That package hasn't been aproved in extras yet :-D
<3> gambolputty: do you have ide or scsi drives?
<1> @donavan ... patience -- wait for one answer before asking another question
<2> I don't have a cpuinfo file
<3> err /proc/cpuinfo
<1> gambolputty: donavan meant /proc/cpuinfo
<4> orc_orc: ahh, i see, this is my first centos server, moving from debian and trying to adjust to the differences
<1> alekzandr: welcome
<1> alekzandr: we have full docs to help orient you at: http://www.centos.org/docs/
<2> for kernel stuff http://pastebin.ca/100912
<2> I have one IDE drive
<2> 640 megs of memory
<1> gambolputty: I asked because yum eats much more memory when old kernel packages are present - if you decide not to wait it out, it would make sense to rpm -e the oler ones you have no intention of using again
<1> s//older/
<4> is there a replacement web interface to manaage mysql that centos does package?
<3> sorry orc_orc missed your question
<1> alekzandr: I am aware of none
<1> donavan: np of course
<3> Evolution: has packaged up mysqladmin somewhere
<2> ok
<3> gambolputty: about that /proc/cpuinfo ?
<2> I thought old kernel stuff like this would be removed automaticaly
<2> automatically, typo there
<1> gambolputty: it is intentionally perserved, as a precautionary measure
<2> based on the kernel pastebin I made, what stuff can be removed?
<2> my cpu info http://pastebin.ca/100918
<1> gambolputty: if you have rebooted into the latest kernel, and are happy with it, you may remove older ones -- this will speed things
<3> gambolputty: and what is the output of "hdparm -tT /dev/hda"
<5> donavan: it's in centos.karan.org
<2> hdparm info http://pastebin.ca/100923
<6> Is there a way to run yum over an ssh socks proxy?
<2> I am removing some old kernel stuff
<3> gambolputty: you may paste the output of "uname -a" in here
<5> McPheer: define http_proxy
<3> without the prompt and command line
<5> McPheer: you'll want to consult the yum faq/wiki/documentation for possible pitfalls
<6> Evolution: Will that work with a standard ssh socks tunnel?
<6> Evolution: k
<3> 3.27 MB/S is horrid performance btw
<2> [root@wilt ~]# uname -a
<2> Linux wilt 2.6.9-34.0.2.EL #1 Fri Jul 7 19:24:57 CDT 2006 i686 athlon i386 GNU/Linux
<3> when did yum start acting up?
<2> today
<3> what is the output of "uptime"
<2> I will leave the 34.0.1 kernel just in case
<1> gambolputty: good plan ;)
<3> leave some older ones too
<2> 11:43:58 up 8 min, 1 user, load average: 0.06, 0.22, 0.16
<2> I put a monitor on the computer
<3> when did you start using the .0.2 kernel?
<2> uptime is short as a result
<2> can't remember
<2> in the past month I think
<2> whenever it was released
<3> and have you run yum since the last kernel update?
<5> McPheer: yeah. it works. I've done it once before, but it's NOT pretty..
<2> not sure
<2> was trying to run it today
<5> and of course I didn't doco any of what I had done...
<2> the box primarily runs asterisk
<3> okay gambolputty reboot with the -34 kernel and see if the hdparm -tT /dev/hda performance improves
<6> Evolution: Care to elaborate or do I just need to read?
<3> then try yum update and see how it performs
<2> my kernel stuff nowhttp://pastebin.ca/100934
<2> now that is
<1> McPheer: http://linux.duke.edu/projects/yum/
<6> orc_orc: thanks
<1> np
<1> McPheer: quick answer is: yum largely relies on python modules coded by others for proxy support -- it is spotty in quality at the edges
<2> I will try to leave the box running this time without interrupting it
<2> I have a yum list command going
<1> gambolputty: ;) if you wish, start it with the -d 5 option and you will be entertained by more screen content
<2> that will help
<2> Reading repository metadata in from local files
<2> Setting up Package Sacks
<2> that's where it is occupied now
<1> gambolputty: when it hangs for more than 5 minutes, pastebin is a nice place for the last 5 lines ;)
<2> ok
<7> is it possible to deactivate all usb/serial ports and activate it agian (as in restart)
<2> we will know in 5 minutes then
<7> besides "reboot"
<2> now it is consuming swap space
<1> gambolputty: p a t i e n c e
<8> hurray for competition, hurray for apt !
<5> hehe
<5> dag: hey, stick around for a moment, I've got some issues for you.
<8> evolution: email sticks better :)
<1> tee hee
<5> bah. fair enough.
<8> evolution: go ahead, I'm not leaving
<9> bah apt and yum are for infidels.. there is only tar xzvf
<8> orc :)
<1> smooge: you forgot the && ./configure && make && sudo make install ;)
<5> dag: I've built an updated perl package (5.8.8) for some updated web features, and it'll be in testing soon. Will 5.8.8 allow you to continue to update clam* packages? or will that screw your other package builds?
<9> orc_orc, no no no.. these are precompiled tar balls.
<1> smooge: slacker
<9> I forgot the /
<5> orc_orc: no... it's s/er/ware/ on that last statement of yours...
<1> heh
<9> orc_orc, it was an argument yesterday with a solaris person over why solaris patch and old-fashioned tar balls are better than RPM/Conary/etc
<1> smooge: they are clearly right : -- remember to bury them when you get tired of them and kill them to make them stop talking
<9> eric-xx, depends on the level of deactivation you want. You can remove the usb modules and turn off serial access and reinit the system.
<8> evolution: how are perl and clamav related ?
<2> 5 minutes, no change in debug lines
<1> gambolputty: and what did I suggest .... ?
<9> orc_orc, sadly the request to get taller raised floors did not go through
<2> you said to wait, but nothing has happened
<1> now about the new tape safe?
<1> gambolputty: and ....
<5> dag: crap.. maybe it wasn't clam... there was a thread a while back about you not updating a package for your rhel/centos tree because it wanted a newer perl..
<8> evolution: I still have clamav packages from EL2.1 until EL4
<2> all the swap space is used up
<9> tape safes are no longer allowed. We have to store things in large vault libraries.
<2> I would say something drastic is wrong
<1> gambolputty: please consult your scrollback for hte next step to follow .....
<8> evolution: ah, but I cannot replace perl in my buildsystem, unless I expect everyone to replace their perl (and perl packages)
<2> Reading repository metadata in from local files
<2> Setting up Package Sacks
<2> this has not changed
<2> what should I do?
<1> smooge: Wen Ho Lee's method was inadequate?
<9> he is not to spoken of
<10> Evolution: It was amavisd-new I think.
<1> gambolputty: USE THE PASTEBIN FOR THE LAST 5 lines
<8> evolution: so yes, in any case, if the perl of the distribution is too old for something, I'm stuck
<5> dag: true. I didn't know how you had your build-env set up. I didn't know if that would make it easier for you or not.
<1> smooge: heh
<1> make
<8> yes, amavisd-new is a problem, and perl-Net-Server as well (in combination with amavisd-new)
<5> iWolf: yeah. that sounds right.
<2> http://pastebin.ca/100955
<8> evolution: especially when there are dependencies with base packages I have to be very careful and I cannot (ever) replace base packages if there might be ABI/API issues
<9> gambolputty, Centos 3 or 4?
<2> 4.3
<1> oh ... my ... gawd
<2> what?
<5> dag: fair nuff. I just thought it might help alleviate some issues, but the solution may be worse than the problem.
<1> gambolputty: nothing ;)
<1> just having a bit of fun
<2> did you see anything wrong from my last pastebin?
<5> other than the breakage causing your problem?
Return to
#centos
or
Go to some related
logs:
z60_hdparm.rules
#centos
#linux
test mailserver hello
install pdftoppm +ubuntu
cipherphunk
#gcc
#php
arch vsd gentoo
#osdev
|
|