[00:00] <minashokry> hello guys, I have subdomain.domain.com and svn running on http://www.domain.com/svn, the subdomain is being redirected to another service with mod_proxy like "ProxyPass / http://localhost:xxxx" now, when I try to access the svn I see the request is sent to this other service and I get 404 error. any help?
[00:04] <zroysch> can I not mount and use a mdadm raid1 drive partition when the other drive failed?
[00:08] <SpamapS> you can
[00:09] <SpamapS> zroysch: you should have received a warning on bootup that the array was degraded
[00:09] <zroysch> SpamapS: server has yet to be rebooted. i was actually alerted of it with the md email telling me the one drive has been removed from the array and marked faulty
[00:14] <SpamapS> zroysch: that shouldn't prevent you from using it.
[00:15] <zroysch> well here's what it's looking like md0 : active raid1 sdb[1] sda[2](F)
[00:15] <zroysch>       1953514496 blocks [2/1] [_U]
[00:15] <zroysch> sdb still apparently ok
[00:16] <SpamapS> Yeah, thats a degraded array
[00:16] <SpamapS> it should still be working
[00:16] <zroysch> the scary/weird thing is sda and sdb dont even show up in fdisk -l anymore
[00:17] <zroysch> sudo mdadm -E /dev/sdb
[00:17] <zroysch> mdadm: No md superblock detected on /dev/sdb
[00:18] <zroysch> ah got something with sudo mdadm --detail /dev/md0
[00:19] <zroysch> http://pastebin.com/26T0xh4M
[00:19] <SpamapS> zroysch: shouldn't be anything scary at all.. you have backups right?
[00:20] <zroysch> nope
[00:21] <SpamapS> many have made this mistake, so do not feel bad baout it. RAID is not a backup solution. :)
[00:21] <zroysch> i already know this
[00:21] <zroysch> lost data a few months ago on a raid5 failure
[00:21] <zroysch> i have more hard drive failures than anyone i know
[00:21] <zroysch> i really don't know how or why
[00:21] <zroysch> pissing me off
[00:22] <SpamapS> Hard drives were made to fail
[00:22] <Daviey> SpamapS: you are buying the wrong brand then :)
[00:23] <SpamapS> Daviey: my brother in mac hardware arms.. any ideas on how to get Natty onto a mac book air? ;)
[00:23] <SpamapS> Its failing miserably for me at the moment. :(
[00:23] <Daviey> funny you say that
[00:23] <SpamapS> About to d/l vmware fusion
[00:24] <Daviey> https://help.ubuntu.com/community/MacBookAir :)
[00:24] <SpamapS> no
[00:24] <SpamapS> thats useless
[00:24] <Daviey> bah
[00:24] <Daviey> SpamapS: you have a new laptop then?
[00:24] <SpamapS> Yeah
[00:25] <Daviey> pass.. i had to spin my own iso with my kernel patch to get Maverick on this one.. :)
[00:25] <SpamapS> thats what I'm afraid of
[00:25] <SpamapS> why can't anything be easy?
[00:25] <SpamapS> >:
[00:25] <Daviey> life is a ...
[00:25] <SpamapS> when you look at it
[00:26]  * SpamapS heads off to run errands "IRL"
[00:26]  * Daviey heads to bed.
[00:26] <Daviey> nn peeps
[00:29] <T3rm> I've a small issue, I changed my default gateway, the route is correct, however, traffic is still going to the old default gateway. >.<
[00:30] <Daviey> T3rm: renew your dhcp lease or change your /etc/network/interfaces gateway
[00:31] <T3rm> well, it's statically defined.
[00:31] <Daviey> i assume you are using a different IP address, as your arp table might also need blating.
[00:32]  * Daviey really heads out of the door now. nn
[00:32] <T3rm> How would I restart all of that without a reboot?
[00:39] <qman__> new connections should use the new gateway
[00:39] <qman__> while persisting ones should use the old one
[00:39] <qman__> if new connections are not using the new gateway, the setting hasn't stuck or been done right
[00:39] <qman__> there's no way to change persisting connections over, they must be closed and reopened
[00:39] <qman__> so each program which has such a connection will need to reconnect or restart
[00:49] <T3rm> aha
[00:50] <T3rm> Next question... I'm a noob with sendmail, what quick commands can I use to delete mail from a particular sender email?
[01:41] <DanaG> hmm, I'm pondering removing that Intel nic, after all.
[01:42] <DanaG> Any of you have lots of experience with Intel and Broadcom NICs, to know if there's any real worthwhile difference between Broadcom 5723 and this Intel?
[01:42] <DanaG> lemme' dig up link.
[01:44] <DanaG> http://www.newegg.com/Product/Product.aspx?Item=N82E16833106033&Tpk=intel%20nic%20expi9301ctblk
[01:46] <DanaG> If anything, powertop shows just slightly higher wakeups with the Intel.
[01:57] <DanaG> okay, for now I'm leaving the Intel in place.
[02:17] <hellonew> is there anyone working on xen virtualization?
[02:17] <hellonew> really need some guide in ubuntu 10.4 para-virtalization
[02:19] <DanaG> heh, about the only thing I can think of putting in that PCIe slot, is a serial card.
[02:20] <hellonew> I have tried many appraoches, it still doesn't wok
[02:20] <hellonew> tried to upgrade from 8.04 to 10.04, install an compiled xen domU kernel and modules then replace that in an HVM ubuntu, all doesn't work
[02:27] <hellonew> no one answer....
[02:27] <hellonew> Oh,,, nopes
[02:34] <DanaG> har, this usb3.0 card has a SATA power connector:
[02:34] <DanaG> http://www.newegg.com/Product/ImageGallery.aspx?CurImage=15-644-001-02.jpg&Image=15-644-001-02.jpg&S7ImageFlag=0&WaterMark=0&Item=N82E16815644001&Depa=0&Description=HP%20USB%203.0%20SuperSpeed%20PCIe%20x1%20Card%20Model%20BM867AA
[02:34] <DanaG> Random and somewhat off-topic, but still nifty.
[02:45] <hellonew> newegg....
[02:47] <hellonew> eggache
[04:07] <SpamapS> lifeless: so, the Oneiric version of the lxc tools and templates is far superior to the natty version.. its taking a lot less time to create/start containers on oneiric...
[04:52] <lifeless> SpamapS: yeah, serge + upstream == win
[04:54] <SpamapS> lifeless: unfortunately, cloud-init is broken in natty, and pretty much everything is broken in oneiric.. so spawning containers that can be booted w/ cloud-init data is proving.. tricky
[04:55] <SpamapS> I'm toying with converting the cloud-init stuff to rc.local
[07:01] <Tommy_nmw> hello everybody
[07:05] <Tommy_nmw> hi
[07:06] <trapmax_> is it safe to edit olcDatabase\=\{1\}hdb.ldif while slapd is running? i need to add some not indexed olcDbIndex -parameters
[09:06] <iclebyte-work> I run about 20 servers now and I'd like to enable the automatic security updates however I want to be notified of each package before it's installed. My idea is to run a local apt-repo and then having a staging repo which comes from the live one. However I want to manually 'sign off' on packages for promotion into the local repo from the staging one. I can then point all servers to the main local cache. Are any tools avail
[09:06] <iclebyte-work> able to do this? I don't want to write it myself if it's already been done!
[11:11] <blinkiz> Hello. Where can I get help about failover with ISC DHCP?
[11:13] <blinkiz> It seems like the failover is not true failover. If one of two server goes down, after a couple of hours, the server that is working will start to complain about "peer holds all free leases". It should know that the other server is down and it can use all available IP addresses
[12:02] <cwillu_at_work> I don't suppose anyone has noticed if a recent update slightly broke samba?
[12:05] <zul> wha?
[12:05]  * cwillu_at_work would like to be more specific, but hasn't narrowed it down yet beyond "word 2003 won't save new files" and some unsuccessful-looking messages in /var/log/samba/log.winbindd-idmap
[12:07] <cwillu_at_work> lots of http://pastebin.com/QAWjthU7
[12:08] <cwillu_at_work> (smb.conf hasn't changed in months, this only started a couple days ago)
[12:17] <pmatulis> cwillu_at_work: what version of samba are you running?
[12:17] <twb> So what *did* change a couple of days ago?
[12:17] <cwillu_at_work> pmatulis, whatever is in lucid
[12:18] <cwillu_at_work> twb, I installed updates :p
[12:18] <pmatulis> cwillu_at_work: look at the changelog
[12:18] <cwillu_at_work> nothing is jumping out at me
[12:18] <cwillu_at_work> hence my asking
[12:43] <glycoknob> hi
[12:44] <glycoknob> I'm fighting with a strange bug in 10.4 lts server.
[12:44] <glycoknob> apt-get / lograte / grpconv all have permissions set to 000 after doing a apt-get dist-upgrade
[12:45] <glycoknob> I suspect webmin-virtualmin packages but I'm not sure. There has to be a bug in some post-install script
[12:46] <greppy> glycoknob: I think that may be one reason that webmin isn't really supported on ubuntu.
[12:46] <Pici> Or on Debian.
[12:47] <glycoknob> so the bug is known and related to webmin? I'm really not sure, this apparead just now after upting apt, logrotate and webmin. So I'm not sure wether it is webmin
[12:47] <greppy> not saying it's known, just that webmin isn't really supported due to some of the things it does to permissions and the contents of config files.
[12:49] <glycoknob> ok. I'd like to sort that bug out. Where can I look? It must appear in the post-install scripts of the packages? What else could create permissions 0000 after updating?
[12:50] <glycoknob> ok I can reproce this: apt-get --reinstall apt
[12:50] <glycoknob> after doing this:
[12:50] <glycoknob> ls -al /usr/bin/apt-get
[12:50] <glycoknob> ---------- 1 root root 122640 2011-07-13 13:11 /usr/bin/apt-get
[13:03] <Ursinha> bom dia amiguinhos
[13:50]  * cwillu_at_work blames a windows update
[13:59] <pyasi> has anyone successully upgrade a 6.06 server to 8.04 after the EOL date?  I'm getting errors about prerequisites missing running do-release-upgrade.
[13:59] <patdk-wk>  yep
[13:59] <patdk-wk> it's cause the 6.06 branch doesn't exist anymore, do-release-upgrade doesn't work
[14:00] <pyasi> I am using the old-release.subuntu.com as the URL in my sources.list
[14:00] <pyasi> er. old-releases.ubuntu.com
[14:00] <patdk-wk> I have never done that
[14:01] <patdk-wk> but to upgrade, I just edit sources to what I want, 8.04 in this case
[14:01] <patdk-wk> and to an upgrade, dist-upgrade
[14:01] <patdk-wk> normally works without any issues
[14:01] <patdk-wk> on my friends though he had an issue, but easy enough to fix
[14:01] <patdk-wk> defently not the recommended upgrade method
[14:01] <CrazyGir> pyasi: how has that server gone so long without an update?
[14:02] <pyasi> it's only a month after the EOL for 6.06
[14:02] <CrazyGir> so
[14:02] <patdk-wk> I normally start planning upgrades after 3years, so 2years before EOL
[14:02] <CrazyGir> exactly
[14:03] <CrazyGir> eg, not waiting until the last minute
[14:03] <patdk-wk> 1.6years till planned upgrade to 12.04 :)
[14:04] <pyasi> I would have rather upgraded it earlier too, but there was a business reason that was out of my hand to keep it the way it was, in any case I thought do-release-upgrade should work
[14:04] <patdk-wk> nope, never has for me, on the machines I forgot about
[14:05] <patdk-wk> those rarely touched machines that just do their job you never look at :)
[14:05] <patdk-wk> mainly cause they where installed with non-lts for some reason, so needed more constant updating
[14:06] <patdk-wk> but so far, just changing the source.list file and doing an update worked pretty good
[14:06] <pyasi> ok, i'll try the sources file, I can go right from dapper to hardy, not have to do gutsy, feisty in order?
[14:06] <oCean> pyasi: I think it is archive.ubuntu.com
[14:07] <patdk-wk> gutsy and feisty haven't existed for a long time
[14:07] <oCean> pyasi: https://help.ubuntu.com/community/EOLUpgrades/Dapper
[14:07] <patdk-wk> oh nice :)
[14:07] <pyasi> oCean, those instructions don't work anymore, 6.06 is not on archive.ubuntu.com now
[14:11] <oCean> pyasi: the packages are there
[14:12] <pyasi> but dapper isn't there afaik
[14:12] <oCean> pyasi: oh wait, right
[14:12] <oCean> but at old-releases they are?
[14:12] <oCean> so it seems
[14:15] <CrazyGir> pyasi: could you not just build a new system and migrate services?
[14:15] <CrazyGir> wouldn't that make more sense?
[14:16] <CrazyGir> I definitely don't know what other constraints you have going on
[14:16] <CrazyGir> but that seems to be a more sensible option
[14:38] <shan> hello! I have a ubuntu 8.04 lts server with Raid1 with 2 x 250GB Sata HDD
[14:38] <shan> I have ltsp server also running on it.
[14:39] <shan> All of a sudden today, the users were not able to log into the server through the terminals.
[14:40] <shan> I rebooted the server and while rebooting the server said ext3: no journal on filesystem on md2
[14:41] <shan> any idea as to how I could resolve this situation?
[14:42] <davros> why is server being a pain, somehow KDE was installed cant boot from usb anymore i wanna redo the install ugh
[14:44] <shan> any one who could help me out with pointers to the server booting problem? It would be great!
[14:46] <SpamapS> shan: are you booting into a recovery console then?
[14:47] <smoser> Ursinha, is http://people.canonical.com/~ursula/server/triage-report/ the right location for that report now?
[14:48] <shan> spamaps: yes, it says it that there is some error in the filesystem and to run fsck manually.
[14:48] <Ursinha> hi smoser! I managed to add it in http://status.qa.ubuntu.com/reports/ubuntu-server/triage-report.html
[14:48] <Ursinha> more obvious to everyone :)
[14:48] <shan> I did that too..
[14:49] <shan> but still the problem persists
[14:49] <Ursinha> smoser: it's update every 10 minutes, so it's pretty fresh
[14:49] <SpamapS> shan: are you certain your RAID is running ok? maybe look in /proc/mdstat
[14:49] <Ursinha> updated, even
[14:49] <Ursinha> Daviey: hola
[14:49] <shan> I did that and it says all the drives are working.
[14:50] <smoser> Ursinha, cool. i have http://ubuntu-server-new-bugs.notlong.com pointing to it. had to update that.
[14:50] <shan> for all the partitions it says active and [ _U]
[14:51] <shan> I guess it would be a F if the partition had failed?
[14:51] <jpds> shan: No, it's _ if it failed.
[14:51] <Ursinha> ah, sorry smoser, I'm working on setting things up yet
[14:52] <shan> jpds: so what does the U stand for?
[14:52] <shan> and its says active for all the partitions?
[14:52] <jpds> shan: And it's drives it's showing, not partitions.
[14:52] <shan> ok
[14:53] <shan> md2:active raid1 sdb6[1]
[14:53] <guhcampos> Is there a big difference between the KVM versions from Kernel 2.6.32 to 2.6.38? I plan to deploy a virtual host and I can't decide between CentOS 6 and Ubuntu 11.04
[14:53] <smoser> Ursinha, ah. i see. we're set now. http://ubuntu-server-new-bugs.notlong.com -> http://status.qa.ubuntu.com/reports/ubuntu-server/triage-report.html
[14:53] <shan> 29294400 blocks [2/1] [ _U]
[14:53] <smoser> there isnt as much need for my link as there was before as yours has tokens in its path that my awesome bar will find.
[14:54] <shan> this is a sample of what I get when I do a cat /proc/mdstat
[14:55] <Daviey> Ursinha: hola
[14:56] <shan> md2:active raid1 sdb6[1]
[14:56] <shan> 29294400 blocks [2/1] [ _U]
[14:57] <shan> jpds: ?
[14:57] <shan> spamaps: ?
[14:58] <shan> unused devices: <none>
[14:59] <PrickelPit> shan, whats so difficult? just read...your raid has a defect. your md2 raid1 is missing the second partition amongst /dev/sdb6.
[14:59] <jpds> shan: You have a dead drive, go and replace it?
[15:01] <shan> ok, thanks for the help guys!
[15:01] <SpamapS> to be clear though, that shouldn't cause FS corruption
[15:02] <SpamapS> so you should also consider verifying your data integrity against backups if at all possible
[15:02] <shan> ok
[15:03] <jo-erlend> I have an mdadm raid5 array with lvm partitions on it. I would like to mount these on a blank system. How do I do that?
[15:04] <jo-erlend> I thought it should be sufficient to use mdadm --assemble --scan and then just mount the partitions as usual, but only md0 shows up in /dev and md1 contains my data.
[15:14] <smoser> zul, does the 'apache -v' output at https://bugs.launchpad.net/ubuntu/+source/apache2/+bug/815865 make sense to you?
[15:14] <zul> looking
[15:14] <smoser> it reports 2.2.12 but https://launchpad.net/ubuntu/+source/apache2 shows nothing like that...
[15:14] <smoser> oh.. i guess he's running karmic?
[15:14] <smoser> is there somewhere where i could ssee what versions were in karmic?
[15:15] <jo-erlend> smoser, rmadison
[15:15] <zul> smoser: yeah that looks sane but clint already commented on that bug
[15:15] <smoser> verified
[15:15] <smoser> https://launchpad.net/ubuntu/+source/apache2
[15:16] <jo-erlend> I wish I could figure out how to mount my partitions. :(
[15:16] <jo-erlend> all help appreciated. This is becoming really frustrating.
[15:18] <PrickelPit> jo-erlend, lvm2 installed? vgdisplay -v saying anything?
[15:18] <smoser> zul, SpamapS was digging same time i was.
[15:19] <zul> ah
[15:19] <zul> ok
[15:20] <jo-erlend> PrickelPit, it was not. It is now and what lvdisplay shows seems correct. But I still can't find the partitions in  /dev
[15:20] <PrickelPit> jo-erlend, reboot.
[15:21] <jo-erlend> oh, ok. I'll give it a whirl. :)
[16:31] <DormantOden> Hey, I've just setup a raid 5 with 4 disks, but whenever I copy ISO files and exe files over to it they become corrupted. Anyone have any ideas why that might be happening?
[16:51] <Daviey> hallyn: You might want to set bzr whoami ?
[16:58] <hallyn> Daviey: feh, i had done a bzr launchpad-login...
[17:00] <hallyn> Daviey: there
[17:00] <hallyn> zul: https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/817606  the tree linked to there, can you push that?  (Or I can upload a debdiff if you prefer)
[17:00] <zul> debdiff please
[17:03]  * hallyn wonders about uvirbot sometimes
[17:03] <hallyn> zul: http://people.canonical.com/~serge/lxc-fix-lxc-ps.debdiff
[17:06] <zul> hallyn: done
[17:08] <hallyn> zul: thx.  now back to the shutdown issue.  Though I'm tempted to wait on that in the hopes that we push the full fix into kernel at the sprint...
[17:08] <zul> hallyn: okie dokie
[17:10] <hallyn> but yeah, the problem with oneiric is that /var/run is a symlink to /run, and the lxc monitor opens /proc/<init-pid>/root/var/run/utmp, but ends up opening /run/utmp on the host :)
[17:10] <zul> ewww...couldnt you check to see if the path exists or soemthing
[17:13] <hallyn> zul: yeah, that's what i'm gonna do - check for /run/utmp first
[17:13] <hallyn> the whole process is so hacky from the start that it makes you not want to keep it working :)
[17:14] <zul> heh
[17:32] <hallyn> kirkland: man, the hangs on t1.micro - it must just depend on who you're sharing with.  usually i haven't noticed trouble, but today it's horrible
[17:32] <kirkland> hallyn: yeah, agreed, me too
[17:32] <kirkland> hallyn: today seems really bad
[17:33] <kirkland> hallyn: i upgraded to a m1.small to share with iamfuzz earlier
[17:33] <kirkland> hallyn: i took byobu out of the equation, and it was the same thing
[17:33] <kirkland> hallyn: ssh -t <hostname> screen
[17:43] <hallyn> SpamapS: so how do i find a list of open high-priority sru-nominated bugs against qemu?  (so i can roll them all into one -proposed push)
[18:02] <hallyn> zul: http://people.canonical.com/~serge/lxc-fix-shutdown.debdiff  (last one for today :)  this fixes the shutdown hack for me
[18:06] <zul> done
[18:14] <hallyn> zul: thanks!
[18:15] <hallyn> now to send these upstream
[18:19] <zul> hallyn: no worries
[18:42] <SpamapS> hallyn: you should ask Ursinha
[18:42] <Ursinha> me
[18:43] <Ursinha> just a moment, link on the way
[19:00] <davros> should i use dynamic dhcp or static dchp for a set up of, modem>server>wirelessrouter>terminals ? giving the wireless router a static ip?
[19:01] <cloakable> I'd use static personally
[19:01] <davros> eth0 is modem the eth1 to the wireless giving the wireless a static ip
[19:04] <SpamapS> adam_g: hey are you here at oscon today?
[19:10] <dkn> can i increase the max number of /var/log/auth.log files so i can see even further back in time, or increase the file size limit?
[19:10] <SpamapS> dkn: yes look in /etc/logrotate*
[19:10] <dkn> @SpamapS thanks
[19:11] <Pici> dkn: It's in the rsyslog file
[19:31] <Ursinha> hallyn: sorry, the nominated report wasn't working due to a lp hiccup, and it's taking a while to be generated, but it's on its way
[19:31] <Ursinha> hallyn: it should be here: http://people.canonical.com/~ursula/server/sru-report/sru-accepted.html
[19:31] <Ursinha> in a few minutes, I believe
[19:35] <hallyn> Ursinha: cool, thanks.  Is that going to be regularly generated?
[19:36] <Ursinha> hallyn: yes, every ten minutes for the index, and every half an hour for the nominated
[19:36] <hallyn> Ursinha: awesome!  thanks, that could make a huge difference
[19:36] <hallyn> (in getting srus more organized for qemu and libvirt)
[19:37] <Ursinha> hallyn: cool, that's nice to know :) anything else you find useful, let me know
[19:50] <hallyn> Ursinha: though, mind you, i was going based on the url itself.  the page is not yet loading for me
[19:51] <Ursinha> hallyn: it's not because it's wrong, but because it's taking long for the script to finish
[19:51] <Ursinha> I'm trying some optimizations here
[19:51] <hallyn> ok
[19:51] <hallyn> no hurry then :)
[19:55] <Ursinha> hallyn: is qemu-kvm the package you're interested in?
[19:55] <hallyn> yeah
[19:55] <hallyn> well, and libvirt too
[19:56] <Ursinha> ok, just checking
[19:56] <Ursinha> because I check the packages ubuntu-server team is subscribed to
[19:56] <Ursinha> and the only qemu package on the list is qemu-kvm
[20:01]  * hallyn off to take the car in.  be back in awhile
[20:06] <robbiew> Ursinha: ping
[20:10] <Daviey> Ursinha: the next thing we need to do is smoke test SRU's :)
[20:31] <Aison> where can I register my own oid for ldap?
[21:11] <joren> Does anyone know how to track down which /dev/sdx device "ata1.00:" is referring to in dmesg?