UbuntuIRC / 2008 /10 /16 /#ubuntu-server.txt
niansa
Initial commit
4aa5fce
=== freaky[t] is now known as fReAky_t
=== fReAky_t is now known as freaky[t]
=== freaky[t] is now known as fReAky_t
[02:30] <fduplex> I just followed the instructions to install ubuntu JeOS under vmware server, and when I attempt to install vmware tools, vmware says it is incompatible and won't mount the iso image to the drive. Is there a known workaround for this?
=== PrivateVoid_ is now known as PrivateVoid
[05:35] <smultron> i'm trying to install bacula, but it seems to be missing the bacula-dir.conf file in /etc/bacula
[05:36] <smultron> this was installing using apt-get install bacula-server
[05:57] <greggman> In K-Network MAnager Icon I found out that my box uses eth2..How could I change it to eth0?
[05:58] <ScottK> greggman: Why do you care? What problem are you trying to solve?
[05:58] <greggman> ScottK, since I'm using the box for wifi gator..so I need it to change to eth0
[06:01] <greggman> ScottK, how would I change it? any idea?
[06:02] <ScottK> What I would do is run grep -r eth2 * in /etc and see where that's used. Carefully consider making changes. It'll be quite easy to end up with no networking at all. Also figure out what's using eth0 now.
[06:07] * ScottK is off to bed.
=== xelpud is now known as fduplex
=== fReAky_t is now known as freaky_t
[07:35] <chmac> I'm trying to connect via reverse ssh to an ubuntu machine and getting "ssh_exchange_identification: Connection closed by remote host"
[07:35] <chmac> Any suggestions?
[07:35] <chmac> I've got the person to log onto my machine with `ssh -R 1024:localhost:22 blah`
[07:35] <chmac> I'm now trying to log onto their machine with `ssh -p 1024 localhost` and getting that error
[07:36] <chmac> But, if I connect my laptop in the same way, it seems to work fine
[09:53] <chmac> Managed to get him sorted in the end :)
[10:11] <sylvaing> f:q!
[10:12] <sylvaing> sorry, bad window
[10:12] <sylvaing> :-D
[10:12] <ewook> whats so bad about this window? =/
[10:35] <sylvaing> hi all, it seems that directory of cyrus on hardy had bad owner (/var/lib/cyrus/* /var/spool/cyrus/* /var/spool/sieve/*)
[10:36] <sylvaing> is a packaging bug? or not?
=== freaky_t is now known as fReAky_t
=== fReAky_t is now known as freaky_t
[10:47] <_ruben> sylvaing: could very well be
[10:47] <_ruben> not using cyrus myself .. sieve's tempting, though im used to qmail style maildirs
[10:49] <sylvaing> but when i install just cyrus-imapd, few directory are "root:root" owner
[10:50] <sylvaing> and when I start cyrus mail.log full about permission denied
[10:51] <sylvaing> mail.log is full
[10:51] <sylvaing> (sorry for my english ;-) )
[10:58] <kraut> moin
[11:30] <pschulz01> Anyone here dealt with LDAP(slapd) and creating an addressbook for Evolution?
[11:31] <pschulz01> I'm looking for more information on the attributes used by the address book.
[12:49] <Safri> hello, when I connect to samba server from Windows, it keep saving the logged username and password.
[13:23] <sommer> pschulz01: you should be able to accomplish that using inetOrgPerson as your objectClass
[13:23] <sommer> pschulz01: the mail, cn, title, etc attribute should show up in evolution
=== mdz_ is now known as mdz
[16:17] <rrittenhouse> I need help getting Hardy Server LTS to install on a machine. I am running into this bug: https://bugs.launchpad.net/ubuntu/hardy/+source/linux/+bug/148466
[16:17] <uvirtbot`> Launchpad bug 148466 in linux "Won't boot on a DELL Poweredge 2500" [Undecided,Fix released]
[16:17] <rrittenhouse> Intrepid *WORKS* but I need hardy LTS
[16:17] <rrittenhouse> any ideas other than installing a new controller?
[16:27] <xbxbxb> Hey, I transfered a folder from my windows box to my linux box... I don't understand why I get different values in Windows explorer and with "find * -type f | du --apparent-size -b -c" for the same folder and it does not contain any hidden files. But all files have been transfered and are intact.
[17:08] <rrittenhouse> Is there anyway to drop to busybox with the Ubuntu Server cd?
[17:08] <rrittenhouse> similar to break=top with the other cds
=== Linuturk_ is now known as Linuturk
[17:48] <lantjie> hey folks. do you guys know how to setup a ubuntu-server
[17:48] <lantjie> i am running ubuntu 7.04
[17:49] <lantjie> cache
[17:49] <lantjie> sorry
[17:51] <lantjie> please someone help
[17:54] <greenfly> lantjie: sure you don't mean 8.04? that's the latest Long Term Support release
[17:54] <smultron> what user does apache2 run as? nobody?
[17:55] <sommer> smultron: www-data
[17:55] <smultron> sommer: thanks
[17:55] <sommer> np
[17:56] <smultron> just curious, is there a file that says that somewhere? wondering how to figure that out on other systems, too
[17:57] <sommer> probably /etc/init.d/apache2
[17:57] <sommer> but doing ps -ef | grep apache will show you as well
[17:57] <smultron> ah, thanks
[18:10] <mathiaz> nijaba: is there a reason why the default suite selected in vmbuilder is hardy?
[18:10] <mathiaz> nijaba: I'd expect it to be intrepid for the version in intrepid
[18:12] <nijaba> mathiaz: hmm, good question. not that I know of. Maybe soren preferred to do an LTS by default?
[18:12] <nijaba> mathiaz: if you feel it should be changed, I can do that in a snap
[18:13] <mathiaz> nijaba: well - I think it makes sense to install intrepid by default for intrepid
[18:14] <nijaba> mathiaz: allright, on my way
[18:14] <mathiaz> nijaba: it's a change in the distro plugin
[18:14] <nijaba> mathiaz: I know, in the set_defaults
[18:15] <nijaba> mathiaz: actually in register_settings
[18:16] <mathiaz> nijaba: group.add_option('--suite', default='hardy',
[18:16] <nijaba> mathiaz: already done
[18:17] <nijaba> mathiaz: and pushed
[18:17] * mathiaz pulls
[18:20] <yann2> is the default vmbuilder in intrepid the one in bash or python?
[18:21] <mathiaz> yann2: python
[18:21] <yann2> nice :)
[18:33] <IanFHood> how do I install and/or enable SSL in hardy (or python maybe?) I'm getting 'unknown url type: https' errors in zope/plone
[18:36] <mathiaz> nijaba: hmmm. not good. The image created by vmbuilder with an intrepid suite doesn't boot
[18:36] <mathiaz> nijaba: it seems that the kernel is not found at boot.
[18:37] <mathiaz> nijaba: a hardy image works though.
[18:39] <nijaba> mathiaz: hmmm... doing the test....
[18:43] <nijaba> mathiaz: I'll work on this after returning from dinner.
[18:55] <mathiaz> nijaba: it seems that the dest argument isn't supported in the configuration file.
[19:04] <uvirtbot`> New bug: #232493 in openvpn (main) "Knocks out openvpn tunnels before both ends have keys regenerated" [Undecided,Won't fix] https://launchpad.net/bugs/232493
[19:26] <nijaba> mathiaz: "Error 15: File not found" booting intrepid VM. Have you opened a bug?
=== erichammond1 is now known as erichammond
[19:32] <nijaba> mathiaz: looks like grub is corrupted for some reason. Easy to fix from grub but need to understand why a spurious UUID is inserted
[19:46] <uvirtbot`> New bug: #284572 in samba (main) "smbclient crashed with SIGSEGV in strlen()" [Undecided,New] https://launchpad.net/bugs/284572
[19:58] <jambooda> Hey does anyone know how I can get my system to show all 4GB of ram?
[19:58] <jambooda> i'm not running the 64-bit OS
[19:58] <jambooda> its only showing 3GB but I know I can enable PAE or something like that in the kernel
[19:58] <jambooda> but not sure how
[19:58] <zul> install the server kernel should do it
[20:04] <mjeanson> Anyone using "dfree command" in Samba config under hardy to change the reporting of free space?
[20:06] <mathiaz> nijaba: it may be related to the grub changes outlined in this email: http://www.nabble.com/GRUB-now-supports-specifying-the-root-by-UUID.-Please-help-test.-td19987017.html
[20:07] <acemo> with proftpd, is it possible to instead of using the user as owner let files they upload be owned by www-data?
[20:10] <nijaba> mathiaz: yep, looks like there is now an additional uuid line that we do not update. easy fix. Thanks for the pointer
[20:16] <nijaba> mathiaz: fix done, now testing. Have you opened a bug?
[20:16] <mathiaz> nijaba: nope
[20:16] <nijaba> mathiaz: np, I'll do it
[20:16] <mathiaz> nijaba: why?
[20:16] <nijaba> mathiaz: just for tracking purposes, thought it would be a good idea, am I wrong?
[20:17] <mathiaz> nijaba: well - IMO it's just unecessary overhead
[20:17] <mathiaz> nijaba: we're not tracking *every* bug we fix in LP
[20:17] <nijaba> mathiaz: ok, I'll skip that then
[20:17] <mathiaz> nijaba: as long as the change/fix is documented in the changelog entry that's ok IMO
[20:18] <nijaba> mathiaz: fine, you are way ahead of me on this type of process
[20:19] <mathiaz> nijaba: so instead of filing a bug report for a bug you've just fixed, you could work on figuring out why dest isn't support in the configuration file :D
[20:20] <nijaba> mathiaz: I m on it too
[20:40] <J_P> hi all
[20:40] <J_P> people, why this "01 00 * * * root /etc/scripts/clock/shour.sh" not works in crontab? If I execute in shell /etc/scripts/clock/shour.sh works. why ?
[20:42] <greenfly> J_P: did you wait until one minute after midnight?
[20:42] <J_P> greenfly: sure!
[20:42] <J_P> I wait 10 minutes
[20:43] <J_P> and nothing :-(
[20:43] <greenfly> and you intend this script to run only once per day at 00:01 ?
[20:44] <greenfly> also, did you get any interesting error messages sent to your root email account, or alternatively, anything intersting in /var/log/ ?
[20:44] <J_P> greenfly: ahh sorry, what is woring.. I need that that run each minute..
[20:45] <greenfly> you want that script to run every minute?
[20:46] <J_P> greenfly: yes
[20:46] <greenfly> then you need the line to read:
[20:46] <greenfly> * * * * * root /etc/scripts/clock/shour.sh
[20:46] <greenfly> the way your cron entry was set up, the script would only run once a day at 00:01
[20:47] <greenfly> if it still doesn't work, make sure that /etc/scripts/clock/shour.sh is executable
[20:47] <J_P> ok, I will change that
[20:53] <nijaba> mathiaz: my grub fix did not work. It seems that update-grub pushes a modifcation later on. As I did not write this part, it may take a while for me to understand the full process...
[20:54] <mathiaz> nijaba: you may wanna ask kirkland for some helkp
[20:54] <mathiaz> nijaba: he worked on grub-install lately
[20:55] <nijaba> mathiaz: I think I should leave kirkland alone for a bit, as he is knee deep in ecryptfs
[20:55] <kirkland> nijaba: i'm getting the counter right
[20:55] <kirkland> nijaba: but i'm happy to help with grub after
[20:55] <nijaba> kirkland: that's a very good news
[20:55] <kirkland> nijaba: post your questions here and i'll answer when i can
[20:55] <mathiaz> nijaba: well - of all the bug I've found this is probably the most important
[20:56] <mathiaz> nijaba: as of know, vmbuilder doesn't produce a functionning intrepid vm
[20:56] <nijaba> kirkland: sure. I'll try to move forward in the mean time.
[20:58] <nijaba> mathiaz: agreed. simple work around would be for me to update the menu.lst at the very end, but I am unsure this would be proper.
[20:58] <mathiaz> nijaba: does the partition plugin/code generate UUID?
[20:59] <mathiaz> nijaba: I mean generate a UUID for the block devices that are created?
[20:59] <nijaba> mathiaz: nope. we set groot as (hd0,0) but it is modified by ??? later on in the process, and this is what I am trying to figure out
[20:59] <Bryan> I'm trying to share a printer, via samba. My two windows xp machines can not find the driver.
[20:59] <Bryan> any help?
=== ajmitch_ is now known as ajmitch
[21:19] <mathiaz> nijaba: hm - dapper-server-i386 created by vmbuilder doesn't boot
[21:19] <mathiaz> nijaba: it fails with grub error2
[21:20] <acemo> how do i get the group of my user to have read and write rights on files i upload? with proftpd
[21:20] <mathiaz> nijaba: same for gutsy-server-amd64
[21:23] <nijaba> yes, intrepid is fixed....
[21:23] <nijaba> mathiaz: am I glad you are doing some real testing
[21:24] <mathiaz> nijaba: ok - so now dapper and gutsy fail to load with a grub error 2
[21:24] <sommer> acemo: you could try using umask: http://www.tech-faq.com/umask.shtml
[21:24] <mathiaz> nijaba: that means: Selected disk doesn't exist"
[21:25] <mathiaz> nijaba: http://www.uruk.org/orig-grub/errors.html
[21:25] <acemo> sommer: great thanks!
[21:25] <sommer> np
[21:26] <mathiaz> nijaba: what was the issue with intrepid?
[21:27] <nijaba> mathiaz: we were setting groot (hd0,0) but update-grub was trying to be smart and was trying to guess the UUID from a loop device. a mess
[21:28] <nijaba> mathiaz: fix pushed to trunk if you want to look
[21:28] * mathiaz pulls
[21:28] * nijaba take a brief pause before looking at error 2
[21:31] <mathiaz> nijaba: could you be more explicit in your changelog entries?
[21:31] <mathiaz> nijaba: intrepid guest now boot -> explain why that has been done (mention UUID)
[21:35] <nijaba> mathiaz: oui prof
[21:37] <nijaba> mathiaz: would "intrepid needs groot=uuid, else update-grub tries to outsmart us later on
[21:37] <nijaba> and it wont boot." be proper?
[21:37] <mathiaz> nijaba: ok
[21:38] <madyogi> Hi guys. Can anyone help me out with an Ubuntu-Server 8.04 installation? It's about aptitude/apt-get. I've managed to install a minimal system and try to update the backages. I am getting Segfault when procps installs and dpkg exits with exit status 139 and the package remains unconfigured...
[21:38] <mathiaz> nijaba: but this may be a bug in grub though.
[21:38] <mathiaz> nijaba: it seems that update-grub doesn't handle loopback device correctly.
[21:39] <nijaba> mathiaz: well, would grub be tested for loopback device?
[21:39] <mathiaz> nijaba: apparently not
[21:39] <mathiaz> nijaba: I don't actually know
[21:39] <nijaba> mathiaz: I certainly would not expect it to
[21:40] <mathiaz> nijaba: well - here is a valid use case
[21:40] <mathiaz> nijaba: so this may need to be fixed in grub
[21:40] <mathiaz> nijaba: I'm sure kirkland would love to investigate that issue
[21:41] <nijaba> mathiaz: once he is done fixing other urgent issues, I am sure he would
[21:42] <nijaba> on with your error #2 now
[21:44] <mathiaz> nijaba: great - intrepid boots now :D
[21:44] <nijaba> mathiaz: glad you can confirm it :)
[21:46] * nijaba grumbles: his cache did not include dapper, so I am downloading every package....
[21:55] <madyogi> I am getting Segfault, when I try to configure some allready installed packages with dpkg on a fresh install of the ubuntu-server 8.04. It concerns random packages like procps and screen. These packages just stay unconfigured... Does anyone have some Ideas about this?
[21:56] <madyogi> now also apache won't install because of segfaults on the configure stage. Another packages are installed without problems.
[21:57] <madyogi> I have googled about that problem, but nobody seem to have a solution and all forums questions are unsolved :-(
[22:04] <mathiaz> nijaba: I don't understant what is the FS image setup
[22:04] <mathiaz> nijaba: when is that used?
[22:04] <nijaba> mathiaz: for Xen
[22:04] <nijaba> mathiaz: xen wants one file per partition, not per disk.... really weird stuff
[22:10] <nijaba> madyogi: what I do not understand is that procps being part of the ubuntu-minimal seed, you should not have to install it afterward. How did you install your "minimal" system?
[22:13] <madyogi> nijaba: yes, I've installed my system twice allready. Becaus first time I thought, that it's just some installation bug
[22:13] <madyogi> nijaba: and procps was installed and configured from the begonning
[22:14] <madyogi> nijaba: when I tried to uptade-upgrade the packages, procps configuration threw a segfault
[22:14] <mathiaz> madyogi: if you get random segfault you may wanna check your memory
[22:14] <madyogi> nijaba: then I tried to degrade the package back and I've had the same result
[22:15] <madyogi> mathiaz: ok. I'll try it right away...
[22:15] <madyogi> mathiaz: but the packages I am getting segfault are always the same.
[22:16] <nijaba> mathiaz: solved one pb for dapper, but there seem to be another one
[22:20] <nijaba> mathiaz: did you test edgy?
[22:23] <madyogi> yes
[22:24] <madyogi> mathiaz: oh. no, sorry. I've tested etch
[22:24] <madyogi> but that's not ubuntu...
[22:24] <madyogi> :-)
[22:26] <mathiaz> nijaba: nope - edgy is not longer supported
[22:27] <mathiaz> nijaba: I haven't tested feisty either
[22:27] <mathiaz> nijaba: as it won't be supported when intrepid is released :)
[22:27] <madyogi> ah. I've forgot the main part I suppose. This ubuntu is running on a Xen as domU
[22:27] <madyogi> and dom0 is debian etch
[22:28] <madyogi> and now while I've started to test memory ssh threw me out. Have to talk to our sysadmin tomorrow I think :-) I can't get it back any more...
[22:31] <mathiaz> nijaba: http://paste.ubuntu.com/58534/ <- list of tests I've got setup
[22:31] <mathiaz> nijaba: the naming convention is suite-flavour-arch
[22:37] <madyogi> alright, it's over for me for today. Anyway thanx for help. I'll let you know if I've found out something
[22:52] <nxmehta> what's a simple way to make a backup of a server that allows for bare metal restore, and also allows use of the system during the backup? for my windows desktop i used to acronis...
[22:55] <nijaba> nxmehta: backups are never simple, but I would combine a disk image (done once, cold) and bacula...
[23:04] <smultron> nxmehta: backuppc is simpler than most others.
[23:05] <smultron> nxmehta: i'd also recommend rdiff-backup
[23:09] <mathiaz> nijaba: I've pushed a branch to fix the templates option: lp:~mathiaz/vmbuilder/fix-templates-option
[23:14] <nijaba> mathiaz: ok, great.