logs archiveIRC Archive / Freenode / #debian / 2010 / April / 09 / 1
phogg
linelevel: it would be better to test first that this could have been the problem by seeing what mode directories have when you mkdir.
f00bar80
Titan8990, is there no way to upgade the DBUS package ?
linelevel
phogg: `mkdir foo`, `ls -l` shows that foo/ has mode drwxr-xr-x
phogg
linelevel: aka 755 which means that this is not the problem
linelevel
phogg: well the problem according to checkinstall is that it has mode 2755 instead of 0755
abrotman
f00bar80: did you try?
f00bar80
abrotman, how to ?
phogg
that's very strange
abrotman
f00bar80: maybe you should try setting your system to squeeze and rmeove all references to unstable and lenny ..
         

f00bar80
abrotman, and then run apt-get install gnome dbus ?
abrotman
f00bar80: for the love of god .. use aptitude
linelevel
phogg, if that's not the problem, do you know what else could it be?
phogg
linelevel: nothing springs to mind
qwd
Anyone here use Revelatin password manager under Sid? I can't start it and I'd like to know if it's a problem in Sid or if I'm the only one experiencing this.
*Revelation
abrotman
i use it in squeeze, worked fine the last time i tried to start it
f00bar80
abrotman, so i'll remove both sources for lenny and sid from sources.list and only i'll keep the one for squeeze , then run aptitude -i gnome dbus , is it correct ?
abrotman
qwd: try running it from an xterm
f00bar80: try it .. might explode
no worse than what you have now
qwd
abrotman: oh, now it works. that's odd, no restart or anything.
abrotman
glad i could help ...
qwd
abrotman: yes, thanks :)
f00bar80
abrotman, what's the apt source url for squeeze ?
macrobat
i cant unpause ncmpcpp without restarting mpd. i dont understand why it gets paused in the first place
abrotman
f00bar80: where are you geographically ?
f00bar80
abrotman, middle-east
abrotman
more specifically, i'm trying to pick a mirror near you
just the country is enough
f00bar80
abrotman, egypt
abrotman
though i vaguely recall there aren't any
isn't that .. africa? :)
"deb http://debian.mirror.ac.za/debian/ squeeze main"
and you also need "deb http://security.debian.org/ squeeze/updates main"
amphi
heh, .za is near .eg ?
Vasistha
I think Egypt identifies more with the middle east than mainland africa, though it's on that continent
         

amphi
.de is nearer
abrotman
amphi: wasn't sure if they'd get in trouble for using .il
Vasistha
heh
well even if not, there may not be many cable connections ...
abrotman
actually .. .tr would have been much better, oh well
dfinn
what's the correct way to disable ipv6 in squeeze? Google hasn't given me a difinitive answer
Bop
(Action) Boa Noite srs e sras do SLB ! Mais informo que o voo com destino a lisboa se encontra em transito ... Porta de embarque nÂș -> <-
bpgoldsb
I've got some new hardware which isn't supported in Lenny's kernel. Compiling the modules from source is easy enough. But I would like to do it with module-assistant if possible. That way if I upgrade kernels, the module is rebuild. Can anyone point me to docs for doing this?
Tramp
!tell dfinn -about noipv6
MikePer
Hello. Someone here did compile compat-wireless for squeeze? I think it's not compatible with 2.6.33, can someone confirm that?
abrotman
!tell bpgoldsb aboutkmuto
valdyn
MikePer: 2.6.33 wont get into squeeze
MikePer
valdyn: oh right, sorry, 2.6.32
babilen
bpgoldsb: You could either give more details, or just install a newer kernel from backports (/msg dpkg bpo)
f00bar80
abrotman, still the same issue, that's what i got in .xession-errors http://pastebin.com/tqvuNpx8, after kept only the mentioned sources above then aptitude install gnome dbus and reboot
abrotman
f00bar80: aptitude full-upgrade ?
f00bar80
abrotman, 0 packages upgraded, newly installed, to remove and not upgraded
abrotman
f00bar80: did you do an 'aptitude update' first ?
i did ask you ages ago if dbus was actually running
also, try it as a new user
f00bar80
abrotman, yes dbus is running
Titan8990
f00bar80: what is your current dbus version?
sdk_
Heh, pretty cool.. my php script crashed Epiphany. :)
mohadib
howdy
are solid state drives supported?
babilen
sure
mohadib
sweet
jelly-home
in as much as they feel and act like any other disk drive, yes
mohadib
nice
babilen
!tell mohadib -about ssd
mohadib
ty
babilen
!tell mohadib -about ssd limit
f00bar80
abrotman, 1.2.1-5+lenny1
babilen
mohadib: The last is important if you want to enjoy your ssb disk for a long time
mohadib
think ssd would be good to run a db on?
pretty active db
jelly-home
(Action) gives babilen an X25-M Gen2 with trim support, but no userspace or kernelspace actually doing TRIM.
mohadib: postgres? oracle?
Vasistha
mohadib: active = lots of accesses or lots of transactions? It's the multiple writes that wear down an ssd
mohadib
mysql :\
well, i see the the wait state in top is always about 50%
jelly-home
mohadib: I don't call that a db! :->
mohadib
im not real sure how often it hits the hardrive
jelly-home: :D
but the wait is caused by mysql
we retrive about 10 million records at a time
maybe once every 10 minutes
jelly-home
mohadib: use something like "iostat -x 1" to see if the disk is a bottleneck
HardDisk_WP
hey
mohadib
jelly-home: ok, let me try that now, thanks
HardDisk_WP
anyone here knows how to get http://www.devco.net/archives/2006/03/14/exim_vacation_messages.php working?
jelly-home
mohadib: average wait, service time and %util columns are interesting
mohadib: sysstat package, fwiw
babilen
HardDisk_WP: No, but check the "vacation" package :)
,info vacation
judd
vacation (mail, extra): email autoresponder. Version: 3.3.0-0.3; Size: 13.1k; Installed: 68k
MikePer
Noone compiled compat-wireless for squeeze? I would be happy to get the ath9k_htc driver...
eldowan
I'm currently playing around with tc (and tcng) for traffic shaping on sarge (I know way out of date). I know I must be blind, but I can't seem to find how to import / export the ruleset. Does anyone have a quick pointer?
mohadib
jelly-home: http://pastebin.com/SYp7aipB">http://pastebin.com/SYp7aipB
HardDisk_WP
babilen, uh, I wont see any mails that arrive?
mohadib
jelly-home: look like disk is the bottle neck?
jelly-home
mohadib: sure seems to be a lot of small writes to it
mohadib
think a ssd would help then?
jelly-home
throwing hardware at an issue can't make things worse, if you've exhausted all the configuration tuning options
mohadib
yeah, ill screw with the mysql settings and see if i can do anything
thanks
babilen
HardDisk_WP: Surely not - "man vacation"
HardDisk_WP: Be careful with mailing lists
HardDisk_WP
ah damn, what I need is a way to see the messages I get when I get back and have people know why i dont react
[Jasper]
hej guys, I want to make a .deb from a backport
how can I do this?
\amethyst
[Jasper]: what do you mean "from a backport" ?
[Jasper]: do you mean you want to make a backport?
[Jasper]
I have the deb from backports in my sources
now I want the newer drbd8.3 from the backports..but I don't want all the build stuff on my system
so I want to make it on another system ... a .deb which I can transfer to the production server
\amethyst
[Jasper]: I'm not sure I understand... what's wrong with the .deb from backports?
[Jasper]: why can't you copy that one?
vsmatck
deb packages don't need to be compiled. They are binaries.
\amethyst
[Jasper]: oh, it's a kernel module..
Cain
Hi, How to add a user to a group ?
\amethyst
[Jasper]: so you got drbd8-source from backports, you mean
[Jasper]: when you built the modules, it should have produced .debs, probably in /usr/src/
[Jasper]
ah ok
\amethyst
[Jasper]: they will require the same kernel version you built them for, though
[Jasper]
it's the same :P
\amethyst
[Jasper]: unless they were built with dkms, anyway... but that's what m-a a-i does... it makes .debs, then installs them
dkms has a 'mkdeb' action, I think, but I haven't played with it so I don't know the details
[Jasper]
hmm \amethyst it made a drbd8.tar.b2
bz2
I don't see a .deb
can anyone tell me how to do this?
Titan8990
[Jasper]: what exactly have you done?
[Jasper]
Titan8990 i got drbd8-source from backport
it's a mdoule
module
now I want to make .deb so I can transfer it to another box
Titan8990
[Jasper]: is the package supossed to be built with m-a?
[Jasper]
how do I know?
http://packages.debian.org/lenny-backports/drbd8-source
\amethyst
[Jasper]: install the linux-headers-* package for the kernel you're using on the other machine
Titan8990
[Jasper]: I would say yes, since one of its depends is m-a
\amethyst
[Jasper]: then m-a -l '2.6.26-9-xyz' a-b drbd8-source where 2.6.29-9-xyz is that kernel version
err, maybe without the '-source'
that should give you a drbd-modules-2.6.26-9-xyz package in /usr/src
Titan8990
\amethyst: I haven't tried it but looking at the manpage, shouldn't it be -k instead of -l?
\amethyst
Titan8990: you use -l to specify a version, -k to specify a kernel headers directory
Titan8990: either works, but usually there's no need to do both
[Jasper]
so \amethyst I will do m-a -l '2.6.26-2-openvz-686' a-b drbd8-source?
\amethyst
[Jasper]: yeah, but I might be mistaken and it might need to be drbd8 instead of drbd8-source
[Jasper]
only 1 way to find out
\amethyst
this is assuming that version of drbd supports that version of the kernel
I believe a-b will install the linux-headers package you need; if not, you might need to install linux-headers-2.6.26-2-openvz-686 first
[Jasper]
yes, just did that
\amethyst
anyway, I'm off, but I'm sure someone else here can help if you run into problems
[Jasper]
thanks anyways
jeward
Hi, I just did a netinstall of a 64 bit lenny and it isn't updating the kernel?
Oh, nevermind. :-)
jhutchins_lt
judd: versions kernel
judd
Sorry, no package named 'kernel' was found.
jeward
(Action) can't read.
babilen
,kernels
judd
Available kernel versions are: experimental: 2.6.33-2-686 (2.6.33-1~experimental.4); sid: 2.6.32-4-686 (2.6.32-11); squeeze: 2.6.32-3-686 (2.6.32-9); lenny-backports: 2.6.32-bpo.3-686 (2.6.32-9~bpo50+1); lenny: 2.6.26-2-686 (2.6.26-21lenny4); etchnhalf: 2.6.24-etchnhalf.1-686 (2.6.24-6~etchnhalf.9etch3); etch: 2.6.18-6-686 (2.6.18.dfsg.1-26etch2)
pickcoder
I've tried adding back a couple of devices that failed in mdadm, but it's just putting them back as spares and is not rebuilding. I'm unable to stop the array at this point and I can not remove the LVM setup onto of it.
is there a way to force mdadm to put the drives back in as specific device numbers?
LaidBack_01
I had to so something similar about six months ago. Had to work with the UIDs on the drives in question. Made changes to the newone to look liek the old one as I recall.
pickcoder
they're the same drives
LaidBack_01
sure, but the uid isn't
there's a unique ID on the drive that's set for each one - usually by filesystem or partition, I cannot recall.
pickcoder
originally, the 4th drive of a RAID5 failed and mdadm went to lunch completely
no mdstat or anything
LaidBack_01
it's not normal to have to do that, but it's a way to get around it.
so... when you cat /proc/mdstat it's empty?
pickcoder
I was getting a null file before, yes
after I rebooted the machine it showed the last drive failed
LaidBack_01
wow, that's pretty bad. okay, so it's back then. that's good.
pickcoder
so I removed it.. what's wierd is the same drive showed back up as a newer drive letter
rhineheart_m
hello! how to check if in CLI my other 2 SATA drives?
pickcoder
sde originally
now it's sdf
LaidBack_01
so, did you fail the drive becore you removed it?
pickcoder
mdadm failed it already
when I added it back it started to rebuild
but it was rebuilding as a spare
!pastebin
dpkg
Do not paste more than 2 lines to this channel. Instead, use: http://paste.debian.net/ http://pastebin.ca/ http://pastie.org/ http://pastebin.com/ or http://picpaste.com/ for pics. Use wikisend.com for large files (think tar.gz) up to 100MB. Remember to tell us the URL of your paste! Cannot cut and paste? Ask me about <pastebinit>.
LaidBack_01
well, sh*t, that's better than nothing. I guess you are letting that happen and working towards a solution - hence why you are here.
pickcoder
uh.. it's degraded at this point
rhineheart_m
I have 3 sata drives.. my concern is.. how to check if the system has abe to recognize those? what's the command please? thanks.
pickcoder
I tried removing the fauly device number, which still showed up
--remove fault
--remove failed
and 2 of the 4 disappeared
http://paste.debian.net/68078/
it's totally fubar now
(and this all happened on a live database setup)
LaidBack_01
so, it sounds normal... just like what the instructions say.
pickcoder
unfortunately, it won't rebuild
LaidBack_01
anyway, let me read the pastie
pickcoder
and I can stop the array
s/can/can't
LaidBack_01
so... that's pretty scary.
pickcoder
I have no idea what to do at this point
I need to rebuild /dev/md0 with a new set of drives, which will probably end up being the same drive letters
« prev 1 2 3 4 5 6 7 8 9 10 11 12 13 14 next »