[06:39] <babyface> utlemming,  job: precise-server-ec2-daily and quantal-server-ec2-daily  are failing
[07:38] <jamespage> babyface, would you be interested in learning how to diagnose failures with the ec2 tests?
[07:41] <babyface> jamespage,  I'm afraid I dont have that much efforts due to I have to handle all the failures on the iso and upgrade test on quantal and precise.  but anyway,  if you have any stuff(wikis, docs...) on ec2 test, you can share with me , I wanna learn more things when I'm not that busy
[07:41] <jamespage> babyface, hmm - well I took a look at the test failures you reported above
[07:41] <jamespage> its only around 2 failures of the 30 or so test cases run
[07:42] <jamespage> some are related to not being able to access launchpad for some reason
[07:42] <jamespage> and the others are ec2 related - i.e. something odd going on the in the ec2 infrastructure itself
[07:43] <babyface>   jamespage    you mean this one?  bzr: ERROR: Invalid http response for https://xmlrpc.launchpad.net/bazaar/: Unable to handle http code 502: Bad Gateway ,  ok, that's a test environment issue,  next time, when I met this kind of error, I will wait for the result of next run
[07:44] <jamespage> yes - we get that sort of transient issue from time to time
[07:44] <jamespage> generally if the majority of tests are passing and you see one or two failures its one of these two issues
[07:45] <jamespage> trends are more interesting - i.e. a single arch failing across the board, a single instance type failing across the board
[07:48] <Kartagis> do I have to reboot my computer to be able to install a vm? I've installed xen
[07:52] <babyface> jamespage,  ack. I will ignore the test envrionment issue later and  report you the real failure
[08:52] <soren> zul, adam_g, Daviey: If you're interested, I have branches of your Quantal packaging branches here: lp:~cisco-openstack/$project/folsom  I'll be pushing my changes there, and probably pulling from your Quantal branches regularly. If you want to pull from my branches too that'd be cool. I don't see much reason for divergence.
[09:13] <Hexch> Hi I have two ubuntu x64 10.04 server where I have attached an qnap - freenas 8.2 as nas.
[09:13] <Hexch> On the one server I see lots of nfs not responding
[09:13] <Hexch> The other one works great
[09:14] <Hexch> what could be wrong?
[10:26] <MatBoy> what shall I do these days for Ubuntu... istall open-vm-tools that include the packages for the guest or just add e vmware repo ?
[10:57] <AlphaWolf> I'm trying to extend the partition my install is on, and I messed it up last time. I've tried GParted but I get the error "Cannot have a partition out the disk!". Here's my current output: http://pastebin.com/A95tkvzw I only need the data on the last 2 partitions (5th is / and 6th is /swap). Also, I have full backups now :)
[10:59]  * xnox ponders to suggest wipe them all and restore from backup.
[11:02] <xnox> AlphaWolf: disks are linear. You now have "small partition"-> loads of empty space -> "partition" -> "linux" -> "swap"
[11:02] <xnox> AlphaWolf: what you want in gparted is to Move the partition from current location to the "loads of empty space"
[11:03] <xnox> it will only be possible to extend it up to the "partition"
[11:03] <xnox> unless you create LVM out of "Loads of empty space"
[11:03] <xnox> move your install there
[11:03] <xnox> and then add the space where "linux" is currently on to the LVM.
[11:03] <xnox> to use both for the LVM
[11:04] <xnox> please note your install will not boot from LVM, unless you chroot into it and modify it to support LVM.
[11:04] <AlphaWolf> xnox: If I were to just wipe it, is it best to do a re-install (re-formatting during install) and then restore, or wipe and restore? It sounds like that's harder to mess up!
[11:05] <xnox> AlphaWolf: no, because wiping everything will remove the other W95 partition you preserved.
[11:06] <xnox> AlphaWolf: a single partition cannot "jump" a partition in the middle of the disk.
[11:06] <xnox> And /dev/sda4       251138097   312590879    30726391+   f  W95 Ext'd (LBA)
[11:06] <xnox> is in the middle of nowhere.
[11:06] <xnox> What is /dev/sda4 ? is it actually real
[11:06] <xnox> because it overlaps with /dev/sda5
[11:07] <xnox> I'd delete /dev/sda4 and then move the linux partition with Gparted or simply dd to the beginning of the free space
[11:07] <xnox> and extended up-to swap.
[11:07] <xnox> then resize filesystem and you are done.
[11:08] <AlphaWolf> I'm not too sure, it was created when I installed Ubuntu Server.
[11:08] <AlphaWolf> I know it's an "extended" partition
[11:13] <AlphaWolf> GParted does not show my partition table though, which I guess is to do with it being incorrect
[11:14] <soren> xnox: sda4 is an extended partition. DOS partition tables only allow 4 partitions. To overcome this limitation, a primary partition can be desginated an "extended" partition which in turn can be split into a bunch of "logical" partitions.
[11:14] <soren> Logical partitions are the ones numbered 5 and up on Linux.
[11:15] <soren> fdisk always shows it this way.
[11:15] <soren> (i.e. the extended partition spans the logical ones)
[11:16] <xnox> soren: aha. /me uses GPT for the past 6 years.
[11:17] <xnox> soren: AlphaWolf: but you have deleted two primary partitions.
[11:17] <xnox> soren: can you convert extended into primary?
[11:17] <soren> xnox: You trend setter, you.
[11:17] <soren> xnox: I think you can make an extended partition primary instead, but it's a wee bit delicate.
[11:18] <soren> (like everything else that involves fiddling with you partition table)
[11:18] <xnox> soren: it's fine AlphaWolf has backups =)
[11:18] <soren> Just delete the logical partitions and the extended one and make a new primary one that has the same start and end as the logical one it's meant to replace had.
[11:19] <soren> I'm not sure why, though?
[11:22] <AlphaWolf> So basically extend partition 1 and then move the data to it?
[12:09] <sazawal> Can I make bottom panel in gnome 3 shell to unhide when a notification is received? Say a notification for message received in pidgin or xchat?
[12:09] <_ruben> err, how's that related to ubuntu-server?
[12:26] <zul> soren: cool ill have a look today
[12:28] <jibel> how can I update grub configuration when provisioning local cloud images with cloud-init ? I want to add an argument to the default kernel command line
[12:29] <Daviey> soren: How do you feel about pushing back to the core branches?
[12:37] <soren> Daviey: As long as everyone's happy with the fact that I'm not testing this at all on Quantal, sure.
[12:38] <soren> Daviey: I even applied for membership yesterday.
[12:41] <Daviey> soren: hah.. well.. the ~ubuntu-server-dev branches should be the source of precice and quantal.
[12:42] <soren> Daviey: They're not.
[12:42] <soren> Daviey: TAlked to adam_g last night.
[12:42] <soren> See scrollback.
[12:42] <Daviey> soren: hmm, i thought that was the direction we are going in?
[12:42] <Daviey> .. reading
[12:43] <Daviey> soren: gah, yes.. sorry i mixed ~ubuntu-server-dev with ~openstack-ubuntu-testing,
[12:44] <soren> Whatever direction you're going, the current state of affairs is that the -proposed under the testing team is where stuff happens.
[12:45] <soren> If you want to keep that team tightly controlled, that's fine. You can pull from my branches at your own pace.
[12:45] <soren> I have a shell script that does the same in the opposite direction.
[12:45] <soren> It's a 15 second task each day. Meh.
[12:47] <Daviey> soren: I'm quite keen on one place being the source of goodness, with specific changes not suitable for Ubuntu being separate.
[12:47] <Daviey> Ie, your tree only has the minimal magic you need.
[12:47] <Daviey> (Which IMO, should be less work for you long term anyway)
[12:47] <soren> ATM there's *no* diff. I don't really expect there to be.
[12:47] <Daviey> (and it's good for the soul)
[12:47] <Daviey> rocking
[12:47] <soren> I just need to be on top of the flow of updates, that's all.
[12:48] <soren> So I'll be keeping my own branches, but I'm fine keeping them in sync.
[12:49] <Daviey> ok, sounds good.
[13:50] <Daviey> jamespage: hey.. can you comment on what it is asserted against? https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-server-amd64_minimal-virtual/111/testReport/test/MinimalVirtualTest/testInstallSize/ kkthnx
[13:51] <Daviey> max_install_size == ?
[13:55] <jamespage> Daviey: https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/job/quantal-server-amd64_minimal-virtual/111/artifact/tests/test_cases/minimal-virtual/test/*view*/
[13:55] <jamespage> tells me that max_install_size on i386 is 668000 with some extra headroom for amd64 (due to multiarch)
[13:58] <jamespage> Daviey, units are 1k blocks
[13:58] <Daviey> jamespage: mind if i add a logigng output saying what the testing size is? :)
[13:58] <jamespage> Daviey, feel free
[13:58] <jamespage> are we actually going to fix that or are we dropping minimal-virtual
[13:59] <jamespage> ?
[14:00] <Daviey> jamespage: well, smoser seems to have stolen the task of fixing it.. and it seems it can be fixed
[14:01] <smoser> i have stolen no such tasks
[14:01] <smoser> if i'im going to steal something its going to be something i *want*
[14:02] <Daviey> rocking.
[14:05] <smoser> so it seems that this is definitely fallout of the kernel removal of -virtual
[14:05] <smoser> but it really seems to me that it is test case failure
[14:05] <smoser> as the kernel packages seem to me to be correct
[14:05] <smoser> linux-virtual gets you a small /lib/modules (22M)
[14:05] <smoser> the cloud-images get this right.
[14:05] <smoser> linux-generic gets you a big /lib/modules.
[14:05] <smoser> i'm not sure how we're selecting the kernel in that test, but that would seem to me to need updating.
[14:13] <AlphaWolf> Is it a good idea to store temp downloads in the /tmp folder? I assume there's some sort of garbage collection so I might lose files?
[14:16] <sagaci> AlphaWolf, as long as you don't need the files after reboot
[14:19] <soren> zul, adam_g: Just a heads up: I've pushed a bunch of changes to all of your -proposed branches. None of them should affect the built packages, but just fixes divergence in the packaging, enabling test suites, etc.
[14:20] <zul> soren: coolio
[14:21] <soren> I hate version control. It'll forever remember my poor grammar. Gah.
[14:22] <ogra_> ahah, same here
[14:22] <ogra_> though i hate *all changelogs* not vcs ones specifically :)
[15:13] <mojolive> hi, I'm having a serious issue. We were having some performance issues with our RAID-5 set, rebooted the server and now all disks are spares (minus 1 that apparently died on the reboot). Any ideas?
[15:32] <g0bl1n> hi, what's the channel for ubuntu in the cloud ? AWS
[15:32] <g0bl1n> I believe there's one...
[15:37] <g0bl1n> join ##aws
[15:37] <g0bl1n> there ;)
[15:42] <zul> hallyn: ping
[15:42] <hallyn> zul: .
[15:43] <zul> hallyn: so 0.10.0 is out but im going to wait until 0.10.1 is out to ask for a FFE
[15:44] <hallyn> zul: ok
[15:44] <hallyn> zul: note, the 0.9.13 is serving us well - it fixed hggdh's bug of multiple VMstarts in parallel failing
[15:44] <TheLordOfTime> anyone know where i can find pending SRU stuff for a given bug (and the progress on said SRU(s) )
[15:45] <zul> hallyn: sweet...yeah i think 0.10 will be easier to maintain in the long run..
[15:59] <mojolive> g0bl1n: #aws is generic amazon web services, but if it works for you ;)
[15:59] <g0bl1n> mojolive: it did :-)
[16:03] <mojolive> I have been having some issues with iscsi on the 3.2 kernel so I went to 3.5 (from mainline repository)
[16:04] <mojolive> 3.5.0 was fine, 3.5.2 was fine, now on 3.5.3 and it doesn't seem to create the configfs for target. any ideas? Most scripts that restore the config bork out on /sys/kernel/config/target (folder) missing
[17:15] <Psi-Jack> Hmmm. So I have this server that is throwing me a lot of "Resource temporarily unavailable" messages from a daemon process manager called daemon. Seems that this issue only occurs on the actual user running the process, so I'm starting to think it may be some kind of resource limits.
[17:18] <chmac> How do I undo the effects of `sudo passwd www-data` ?
[17:19] <Psi-Jack> vipw -s, and set the password to *
[17:20] <chmac> Psi-Jack: Ok, thanks, will check that out.
[17:21] <chmac> RTFM, man passwd reveals `sudo passwd -d www-data` :-)
[17:21] <Psi-Jack> Not quite the same thing.
[17:22] <chmac> Psi-Jack: Ahh, you're quite right, it sets a blank password in /etc/shadow, not *
[17:22] <chmac> Psi-Jack: What's the difference in practice?
[17:22] <Psi-Jack> Precisely.
[17:22] <Psi-Jack> Blank password = no password. * == disabled, no password will ever work, period.
[17:24] <chmac> Psi-Jack: Well I'm gladd I asked, not sure I'd ever have found vipw otherwise :-)
[17:26] <chmac> `ssh www-data@host` works if I set a password, but it's not accepting the key in /var/www/.ssh/authorized_keys
[17:27] <chmac> Permissions are 700 on /var/www/.ssh/ and 600 on /var/www/.ssh/authorized_keys
[17:31] <chmac> Got it, syslog, chmod g-w /var/www/... :-(
[17:31] <chmac> Psi-Jack: Thanks for your help
[17:45] <jkyle> is it possible to tell dpkg to look for dependencies in the current directory and install them if needed?
[17:45] <jpds> jkyle: No, you'll need to just add the packages to the dpkg -i command.
[18:05] <jgcampbell300> !juju
[18:05] <jgcampbell300> !juju channel
[18:05] <Psi-Jack> Heh
[18:05] <Psi-Jack> Oh god.
[18:05]  * Psi-Jack wonders why he's talking to himself.
[18:05] <jgcampbell300> hello
[18:06] <Psi-Jack> I effing HATE that term that's so meaningless, "in the cloud"
[18:06] <jgcampbell300> you know i hate it to ... but im trying to learn
[18:06] <jgcampbell300> and its kicking my ass
[18:32] <Daviey> zul: hey, any progress on essex->folsom on precise upgrade testing ?
[18:33] <zul> Daviey: just debugging my script
[18:33] <Daviey> coolio
[18:44] <jgcampbell300> does anyone here know where the configs for MAAS are handeled
[18:49] <cr3> has anyone tried deploying the current quantal server image with MaaS? if so, I suspect it should be using the new filesystem.squashfs for super fast deployment, right?
[18:52] <jgcampbell300> cr3: i am haveing a hard time just getting my system built lol ... do you know where maas leaves its configs for dhcp and dns ?
[18:53] <cr3> jgcampbell300: I believe MaaS provisions those with dnsmasq, so probably its config is where you want to look
[18:54] <jgcampbell300> cr3: ok thank you
[18:57] <jgcampbell300> wow i need to find a good uptodate guide to install all the latest packages to build a web hosting cloud
[19:00] <jgcampbell300> ok ... i checked .. /etc/dnsmasq.conf and it says addn-hosts = /var/lib/cobbler/cobbler_hosts ... so i check that file and its empty .... i wonder if i have finaly found the place i can add in dns name servers
[19:39] <smoser> hallyn, ping
[19:49]  * Daviey eww's at SpamapS.. help2man in packaging.
[19:52] <smoser> hallyn, http://paste.ubuntu.com/1174511/
[19:53] <smoser> maybe stgraber would know.
[19:53] <smoser> what am i doing there that is stupid
[19:53] <smoser> system (remote system) goes to lunch when i do that.
[19:53] <smoser> by goes to lunch, i mean network connection dies
[19:58] <smoser> comon... someone help me
[19:58] <soren> Daviey: The idea is fine... I just wish help2man's output was nicer.
[19:59] <smoser> there, that guy can help.
[19:59] <smoser> soren, what am i doing wrong.
[19:59] <soren> Man, where to start?
[19:59] <soren> :p
[20:00] <smoser> thank you. thank you very much.
[20:00] <soren> Does it disappear immediately?
[20:00] <soren> As soon as kvm starts?
[20:00] <soren> Or does it happen after a little bit (i.e. likely when the guest starts booting and fiddling with networking)
[20:00] <smoser> the last thing that gets printed is the brctl line
[20:00] <smoser> i dont eve see the kvm echo
[20:02] <smoser> but if i comment out 'kvm' everything is fine.
[20:02] <Daviey> soren: No, if i wanted the same content in --help as i did man.. i'd just --help.
[20:02] <Daviey> soren: using it as a base, sure.
[20:03] <soren> smoser: Interesting.
[20:03] <soren> smoser: Where is this server?
[20:03] <Troy^> does anyone have any experience with streaming 1080p files from a weird file server to a wireless comp/laptop on 802.11n?
[20:03] <Troy^> wired*
[20:04] <soren> smoser: In Canonical's data centre?
[20:04] <smoser> soren, on a canonistack instance
[20:04] <soren> smoser: Oh.
[20:04] <soren> That's why.
[20:04] <soren> The bridge probably grabs the tap device's mac.
[20:05] <soren> Try hardcoding the bridge's mac to match eth0's.
[20:07] <soren> smoser: I think adding "bridge_hw 00:11:22:33:44" or whatever to /e/n/interfaces should do the trick.
[20:10] <smoser> http://paste.ubuntu.com/1174543/
[20:10] <smoser> ?
[20:11] <stgraber> smoser: it's likely that your bridge will change mac address when doing that
[20:11] <stgraber> smoser: which may explain what you're seeing
[20:11] <smoser> right
[20:11] <smoser> i think i might be just doing osmething stupid
[20:12] <smoser> it might be better to just put an lxc instance on the lxcbr0 and then try to kvm off that.
[20:12] <smoser> rather than having the "real eth0" on a bridge.
[20:13] <smoser> hard coding the MAC didn't seem to have an affect
[20:17] <hallyn> smoser: eth0 is on br0?  and that's done right?
[20:17] <smoser> it was.
[20:18] <hallyn> heh just noticed the comment.
[20:18] <hallyn> that green is too 'ignore me'-ish
[20:19] <hallyn> smoser: wondering whether your instance is behind some 'smart' switch which says "you're trying dhcp with >1 mac, I'm shutting your port down"
[20:19] <soren> smoser: Don't you have VNC access?
[20:19] <hallyn> have you tried it on a local kvm instance?
[21:07] <smoser> hallyn, afte rlxc-create -t... and lxc-start, how do i get an IP address of the system ?
[21:08] <smoser> and what is the escape key after i did 'lxc-console'
[21:11] <hallyn> smoser: after lxc-console, escape with 'ctrl-a q'
[21:11] <hallyn> (ctrl-a ctrl-a q if in screen)
[21:12] <hallyn> smoser: if you're on quantal, you can get the ip address using:
[21:12] <hallyn> sudo lxc-attach -s NETWORK -n q1 -- /sbin/ifconfig
[21:12] <hallyn> but, i thought lxc-ip was supposed to be shipped with lxc now
[21:12] <hallyn> stgraber: ^ ?
[21:13] <stgraber> hallyn: nope, lxc-net was removed and replace by "ip netns"
[21:14] <stgraber> *replaced
[21:18] <hallyn> smoser: so i did:
[21:18] <hallyn> lxc-ps -n q1 (to get ip of container's init)
[21:18] <hallyn> mkdir -p /var/run/netns
[21:19] <hallyn> ln -s /proc/$containerinit/ns/net /var/run/netns/q1
[21:19] <hallyn> sudo ip netns exec q1 ifconfig -a
[21:19] <hallyn> $profit
[21:25] <Daviey> cr3: hey
[21:25] <Daviey> roaksoax: can you confirm bug 1042906 for cr3 please?
[21:25] <cr3> Daviey, roaksoax: hi folks, you guys might like to know about bug #1042906. the workaround for now is simply to create /tmp/live-installer in the early_command of the preseed
[21:26] <cr3> roaksoax: if you can confirm you've observed the same problem, that would be a good start but there's more
[21:27] <riz0n> Hey guys I have a question about Ubuntu Server and Postfix. In my main.cf file, I the home_mailbox = Maildir/ but when messages come in, they go to mbox unless I comment the mailbox_command = procmail -a "$EXTENSION" line out. Do I need the mailbox_command line in tihs file? If so, what should I change about this to make it go to Maildir/ instead of mbox? Thanks.
[21:27] <Daviey> cr3: If there is no response, i will validate.. and work on it tomorrow.
[21:27] <cr3> roaksoax: once you actually work around to downloading the filesystem.squashfs, the installation proceeds nicely until the end. but, when the system reboots, there's no kernel in /boot/vmlinuz-* so the system just boots into memtest86 because grub finds no other sensible boot option
[21:28] <cr3> Daviey: the installation goes blazingly fast though, so it's a well worth investment of time
[21:28] <roaksoax> Daviey: confirm
[21:29] <cr3> roaksoax: yay! what have you done so far in an attempt to get support for the filesystem.squashfs file?
[21:29] <roaksoax> cr3: today was just starting to look at it, 4 months ago i got it working with cjwatson's help
[21:31] <cr3> roaksoax: so there's hope :) any hints I might try to get this working? for example, did you encounter the problem with booting into memtest86 when you last worked on the problem?
[21:31] <roaksoax> cr3: nope, it was pretty simple but tricky, and unfortunately i lost all the changes i had made, so I'm trying to figure it out all over again
[21:32] <cr3> roaksoax: I'm working in the same direction then, so maybe we can keep each other posted on our progress as cjwatson is currently on holidays :)
[21:33] <roaksoax> cr3: alright cool :)
[21:34] <adam_g> roaksoax: where in cobbler or preseed can i set the kernel parameter that will be used on first boot after installation?
[21:34] <roaksoax> adam_g: let me check
[21:38] <roaksoax> adam_g: i believe it is this: d-i     debian-installer/add-kernel-opts        string $kernel_options_post
[21:38] <adam_g> roaksoax: sweet, thanks
[21:55] <cr3> roaksoax: I need to jet but I haven't found anything significant yet. by the way, when I said that there was no kernel installed under /boot/vmlinuz* I didn't mention that I noticed the linux-image-*-generic package installed: chroot /target dpkg -l 'linux-image-*' returned something starting with 'ii '
[23:29] <arrrghhh> hey all.  anyone use PS3MediaServer?  I'm having an issue with it, after i upgraded my server and subsequently PS3MediaServer, and now it won't start with the upstart job.  It seems eventually it starts OK if I  open it manually in a screen session or something
[23:36] <arrrghhh> bbl....