[00:00] <Xpistos|work> hggdh: I think I did this inadvertently when I added my wifes NFS shares when I got her moved over to linux. I didn't have anything in my downloads for a few weeks that is why I didn't' notice and I must have looked at the wrong line when I check my exports before.
[00:00] <Xpistos|work> hggdh: Thanks for you help
[00:01] <Xpistos|work> and oddly enought I can fo things in my NFS share now
[00:01] <Xpistos|work> hmm?
[00:02] <hggdh> :-)
[00:07] <RoAkSoAx> is there a list of all the server packages that use upstart instead of the LSB init scripts?
[00:25] <MatBoy> is there a date around when 10.04 can be released ?
[00:31] <ScottK> MatBoy: April 29
[01:10] <zul> happy happy joy joy
[01:45] <medex> How can I update my version from the terminal no gui?
[01:47] <Pici> medex: do-release-upgrade, check --help for options
[03:17] <hazmat> is there a way to make ubuntu not try to use gcj for all the java packages?
[03:19] <hazmat> i've got a 'real' java installed and identified as such for java alternatives
[03:21] <hungnv> hello all, I create samba share for other ubuntu clients, then I mount at clients, some works well, some I get problem : mount : cannot allocate memory
[03:22] <hungnv> http://pastebin.com/bFanKYYZ at client dmesg shows that problem
[03:23] <hungnv> please give me a solution
[03:26] <ScottK> hazmat: There is an #ubuntu-java.  I don't know if they'd be able to help.
[03:31] <hazmat> ScottK, thanks
[04:02] <lifestream> Could anyone help me with .... setting up NAT(?) so I can use Linux (host) to connect to a program (server) that is on a virtualbox guest? (I am on the #vbox channel too)
[04:02] <lifestream> I don't understand what this means:
[04:02] <lifestream> We have a Guest Machine with a running ssh server which accepts connections on the TCP port 22. Our goal is to make any packet arriving at a given TCP port (i.e. 2222) of the Host machine, to be forwarded to the TCP port 22 of the Guest Machine.
[04:03] <lifestream> Wait... nevermind, I think I'm following the wrong tutorial.
[04:18]  * lifestream is away: afffkkkkk
[04:19] <maxagaz> I can I check the activity of the network in live ?
[04:19] <maxagaz> how
[04:20] <KurtKraut> maxagaz, what do you mean precisely by activity? Give us an example.
[04:22] <maxagaz> KurtKraut, I'd like to be able to check the activity of the network between each node
[04:22] <KurtKraut> maxagaz, 'activity' still an abstract idea.
[04:23] <KurtKraut> maxagaz, bandwith? what IP and port being used? Read the packets content?
[04:23] <maxagaz> KurtKraut, I mean the size of datas going between nodes
[04:24] <KurtKraut> maxagaz, in real time or you want it logged?
[04:24] <maxagaz> KurtKraut, in real time
[04:25] <maxagaz> KurtKraut, my network is often slow, but I don't understand why, it would be easier to see if I can monitor the traffic the traffice between computers
[04:31] <KurtKraut> maxagaz, try the following packages? bwm-ng and nethogs
[04:31] <KurtKraut> maxagaz, try the following packages? bwm-ng and nethogs
[04:31] <KurtKraut> oops
[04:31] <KurtKraut> maxagaz, try the following packages: bwm-ng and nethogs. They may give you the data you need.
[04:32] <KurtKraut> maxagaz, another good resource is http://www.ubuntugeek.com/bandwidth-monitoring-tools-for-linux.html
[05:24] <krezel> I'm running Karmic and I need to run an old webapp that relies on PHP4 and I believe to be suspect from a security standpoint. What's my best option? chroot jail? Compile php4 and run it alongside php5 but with the app isolated by mpm-itk?
[05:26] <ScottK> Best is don't do it.
[05:27] <krezel> I seem to hear that a lot :)
[05:28] <ScottK> It's good advice.
[05:28] <ScottK> If you really need to run php4, set up a Dapper system (in a vm or something).  Dapper has another ~year of security support and has php4.
[05:29] <krezel> Hmm, I'm already on a vm (linode) but I could probably set up dapper in a chroot jail, right?
[05:30] <ScottK> No idea, but presumably.
[05:31] <kees> krezel: chroots are not secure containers.  they should only be used for testing.  if you have no choice, I would port it to run with php5, and then confine it with apparmor.  there are good examples in lucid, but should work in karmic too.
[05:32] <krezel> kees: Sadly I don't think this steaming pile of dung will take the upgrade to php5 well. Its basically a big custom phpwiki and I wouldn't even know where to start
[05:32] <ScottK> Which gets back to my initial advice
[05:33] <kees> :(
[05:35] <krezel> The fun never ends
[06:02] <jasonmchristos> what tracker software does torrent.ubuntu.com use? i want to setup my own tracker
[06:19] <darkk^> kees, by the way, what's wrong with chroots as soon as user does not have root in the chroot?
[06:19] <darkk^> yes, root can almost always escape from the chroot, but how can ordinary user do that?
[08:15] <_minerva> hi
[08:27] <kees> darkk^: depends on the chroot, but many have /proc mounted, which can expose kernel interfaces.  chroots are better than nothing, but there are much better solutions.  :)
[09:03] <jjohansen> ttx: for bug #546743 is the proper solution but we are kind of out of time there so I think for server we can get away with changing /etc/modprobe.d/radeon-kms.conf to have options radeon modeset=0
[09:05] <ttx> jjohansen: that would only affect server ? Which package is this shipped in ?
[09:05] <jjohansen> xserver-xorg-video-radeon
[09:05] <jjohansen> ttx: hrmm, well I assume there is a way to only make it affect server
[09:06] <jjohansen> I could be wrong
[09:06] <ttx> jjohansen: maybe it shouldn't just affect server anyway
[09:07] <ttx> jjohansen: I don't think server ships with xserver-xorg-video-radeon ? Or do I get something wrong ?
[09:07] <jjohansen> well I think some desktop people would be awful unhappy if kms was off by default
[09:07] <ttx> .. probably :)
[09:08] <jjohansen> ttx: well that is the desktop package
[09:08] <jjohansen> I guess
[09:09] <ttx> jjohansen: can we move this to #ubuntu-release ?
[09:09] <jjohansen> its installed here, I didn't even think to look in the server install
[09:09] <jjohansen> sure
[10:14] <jjohansen> ttx: is there anyone else with ES1000 hardware who can test soon?
[10:15] <ttx> jjohansen: there are affected servers in the London DC, but I don't have contacts there
[10:15] <ttx> jjohansen: we can also ask in the bug, reporters were pretty responsive
[10:16] <jjohansen> ttx: sure, but I am talking with in hours
[10:24] <ttx> jjohansen: otherwise fader should be up in ~3 hours
[10:24] <jjohansen> okay I'll let apw know
[10:25]  * apw jerks awake
[10:35] <Guest70176> hi there
[10:35] <Guest70176>  maybe someone got a clue! i installed in UEC mode from the cd installer. one host for clc walrus etc, and one host as a nc. everything is working fine. QUESTION: how do i log into the nc?? in the installation process i wasn't ask for a passwort :o
[10:39]  * ttx admires the kernel team capability to suspend/resume
[10:39] <ttx> Guest70176: you should use the same as the CLC
[10:40] <ttx> Guest70176: it's preseeded from whatever you chose on the CLC
[10:40] <ttx> the NC takes a lot of installation defaults from the CLC
[10:46] <Guest70176> ttx: aaah so it's the user / pwd i choosed from the installation on the CLC?
[10:47] <ttx> yes, and feel free to change them if you need different ones
[10:49] <Guest70176> ttx: ohman, thanks! that should go in the documentation :)
[10:49] <ttx> Guest70176: it's a wikipage, fix it :)
[10:50] <Guest70176> ttx: this one, aye? https://help.ubuntu.com/community/UEC/
[10:50] <ttx> Guest70176: I'm not sure it's up to date anyway
[10:50]  * ttx looks
[10:51] <ttx> a quick note on Step3 that the node will inherit user/password from the CLC could help yes
[10:53] <e-DIO-t> ad ogni modo: Yo!
[11:05] <VSpike> Is there any way I can get squid-2.7.STABLE9 onto 9.10 server?
[11:40] <franjpr> I am running ubuntu server 9.10. The other day it failed to start. It halts in the init steps. Any idea what to do?. Thanks
[11:48] <VSpike> franjpr: did you try the recovery mode?
[11:48] <VSpike> franjpr: do the last few lines of output contain anything interesting?
[12:19] <franjpr> vspike: yeah, I tried the recovery mode. Same result
[12:20] <alvin> franjpr: And the last messages you see?
[12:20] <franjpr> last lines are init: ureadahead-other...
[12:20] <franjpr> or whatever process
[12:20] <alvin> ah, try pressing 'M' at that point.
[12:20] <franjpr> let me check
[12:20] <alvin> If my suspicion is right, it should give you a recovery shell
[12:22] <franjpr> no, it justs prints M or m in the screen
[12:22] <franjpr> the boot process stops with init steps
[12:22] <alvin> And if you press 'S'?
[12:23] <franjpr> the same
[12:23] <franjpr> I can type in the screen
[12:23] <franjpr> but there is no command line
[12:24] <franjpr> last line, init: ureadahead-other main process (652) terminated with status 4
[12:24] <franjpr> I read on a thread that this message does not interfere in the boot process
[12:24] <franjpr> and I do not know what to do
[12:24] <alvin> I have the same on all servers, but if I press 'M', the recovery shell appears and I can mount remaining filesystems manually. Just like you, the last lines are ureadahead... and I can't see what error actually occurs (filesystems not mounting)
[12:25] <franjpr> not in my case
[12:25] <alvin> franjpr: What's your disk layout? Are you using mdadm or lvm?
[12:25] <franjpr> there is no recovery shell
[12:25] <franjpr> lvm
[12:25] <alvin> Do you have snapshots?
[12:26] <franjpr> no
[12:26] <alvin> Do you have more than 4 lvm volumes?
[12:26] <franjpr> no
[12:26] <alvin> In that case, it's something I did not encounter yet.
[12:26] <franjpr> just two hdd
[12:26] <franjpr> one /dev/sda
[12:27] <franjpr> and another /dev/sdb
[12:27] <franjpr> in /dev/sda
[12:27] <alvin> no mirror?
[12:27] <franjpr> no
[12:27] <franjpr> it is a very simple server
[12:27] <alvin> and single user mode (recovery in grub) gives you the same output?
[12:27] <franjpr> in /dev/sda I have /dev/server-root
[12:28] <franjpr> and /dev/server-root/boot
[12:28] <franjpr> or something like this
[12:28] <franjpr> yes
[12:28] <franjpr> I go to grub menu
[12:28] <franjpr> and select recovery
[12:28] <franjpr> and I get the same result
[12:29] <alvin> Ah, a separate boot! Since karmic, I have stopped using that because of bug 462961
[12:29] <alvin> This is a fresh install or an upgrade?
[12:29] <franjpr> this is an upgrade from 9.04
[12:29] <franjpr> but it has been working
[12:30] <franjpr> normally
[12:30] <franjpr> untill yesterdayi
[12:30] <alvin> With 9.10 in between? Or did you switch sources?
[12:30] <franjpr> I upgraded from 9.04 to 9.10, just following the upgrade process from ubuntu
[12:30] <alvin> Ah, ok. I thought you were on Lucid.
[12:31] <franjpr> ah no
[12:31] <alvin> Nothing else in /etc/fstab? NFS mounts, CIFS mounts?
[12:31] <franjpr> fstab?
[12:31] <franjpr> I can run an ubuntu server cd
[12:31] <alvin> Yes, /etc/fstab, where your mount points are defined
[12:31] <franjpr> and edit whatever file
[12:32] <franjpr> I have not changed anything these days
[12:32] <alvin> I would do that. Check the contents of /etc/fstab (or post them here, using pastebin)
[12:32] <alvin> No recent kernel upgrade either?
[12:32] <franjpr> I can check the file
[12:32] <franjpr> I can't tell right now
[12:32] <alvin> You didn't change anything in the BIOS?
[12:33] <franjpr> however I  have tried to select other kernels from grub menu, with the same result
[12:33] <whom> i have a local code <div>'ça'  </div> , remote server is ubuntu, via ftp, i upload my code, i check it back from server, output is:  'Ã§a'   , what is wrong?
[12:33] <franjpr> no, when before this issue. Afterwards, I have tried.
[12:34] <franjpr> let me check fstab
[12:38] <franjpr> alvin: I am running ubuntu server cd
[12:38] <franjpr> repair broken system
[12:38] <alvin> Yes, answer a bit more questions and choose to have a shell in /
[12:38] <franjpr> I have /dev/sda /dev/sdb /dev/server/root /dev/server/swap_1
[12:39] <alvin> /dev/server/root
[12:41] <franjpr> let's the fstab
[12:43] <franjpr> the seems to be correct
[12:43] <franjpr> nothing has been changed
[12:52] <alvin> franjpr: I'm not suggesting anything is wrong with your /etc/fstab. Neither is anything wrong with mine, but most servers do not boot either.
[12:53] <whom> i did set binary, and transfer one line to that ubuntu server, i read it back , it gives me broken chracters. then i also send in ASCII same. whats the point now?
[12:54] <whom> i have a local code <div>'ça'  </div> , remote server output is:  'Ã§a'  ?
[12:54] <franjpr> alvin: I am a bit lost. I do not know how to proceed to recover the server.
[12:56] <alvin> franjpr: Me neither. I encounter boot problems on a daily basis, but yours are new. I wish there was a log somewhere. I think you can set mountall to debug mode, but do not know how. Anyone?
[12:58] <alvin> Aer your lvm volumes defines by UUID in /etc/fstab, or are they /dev/mapper/server-root, etc,...
[12:59] <franjpr> by uuid
[13:00] <alvin> Try changing them back to /dev/mapper
[13:00] <franjpr> ok+
[13:01] <franjpr> however it seems they are mounted correctly
[13:01] <whom> alvin, franjpr : in my centos LANG=en_US.UTF-8, i have a local file with only this value: 'ça' , when i use upload ftp (ascii/binary)  that file, which is hosted ubuntu server,  and again if i download it back, the same file value gets changed to this: 'Ã§a' , do you read me!!
[13:01] <franjpr> at least they are identified and fsck checks them
[13:01] <franjpr> afterwards, in the init steps...
[13:04] <alvin> whom: We're not a helpdesk. Read up on ASCII vs BINARY file transfers by FTP.
[13:05] <alvin> franjpr: Still halts? (try 'M' now if it does)
[13:06] <whom> alvin, i am not  a client also, came in the community, i did ASCII and Binary transfer both same result, but when i cent my centos box to another remotelocation same centos the same characters output same. What is the point, ubuntu ftp server is not listening or miss configured?
[13:12] <whom> alvin, anyway you are not listening, like all of those broken neck. you dont know this simple point dont you? shame on you, playing with big toys.
[13:49] <franjpr> alvin: I changed in fstab all uuid to /dev/... with the same result
[13:49] <franjpr> the partitions are correctly identified
[13:49] <alvin> This is bad. What could it be then?
[13:49] <franjpr> no idea
[13:49] <alvin> Did you press 'm' this time?
[13:50] <franjpr> partitions are checked
[13:50] <franjpr> then stops
[13:50] <alvin> You could file a bug against mountall, but I wouldn't know what information to provide in your case, except for the description; https://bugs.launchpad.net/ubuntu/+source/mountall
[13:50] <franjpr> after the last line init:whatever...
[13:51] <franjpr> the cursor blinks
[13:51] <franjpr> you can type m or M
[13:51] <franjpr> hit enter
[13:51] <franjpr> go to the next line
[13:51] <franjpr> but nothing happens
[13:51] <alvin> No, it should give you a rescue shell. In your case, something crashes, but I don't know what.
[13:52] <franjpr> yes
[13:52] <franjpr> i do not know
[13:52] <franjpr> and i do not know what to report
[13:55] <alvin> Just where your boot halts, what the last messages are, your kernel version, mountall version and /etc/fstab I think.
[13:56] <ttx> smoser: around ?
[13:56] <smoser> here
[13:56] <smoser> whats up ?
[13:57] <ttx> smoser: about openldap's latest comments from mathiaz, I assume you uploaded before he commented ?
[13:57] <ttx> smoser: if so, We should create a bug to track the remaining items
[13:58] <ttx> alvin: re: mountall logs: you can edit /etc/init/mountall.conf so that the mountall line reads:
[13:58] <smoser> we did upload before he commented, kyes.
[13:58] <ttx> exec mountall --debug --daemon $force_fsck $fsck_fix > /dev/mountall.log
[13:59] <ttx> smoser: ok, I'll create a bug for the remaining stuff
[14:01] <smoser> ttx, yeah. i can try to gake a quick stab at it and race the Final Freeze
[14:01] <ttx> smoser: I doubt that, given we are already frozen
[14:02] <smoser> oh.
[14:02] <smoser> did not realize that.
[14:02] <smoser> so yeah, thats an unlikely race to win
[14:03]  * smoser searches for that flux capacitor ... i know its around here somewhere
[14:03] <ttx> alvin: about your boot issues on 10.04
[14:04] <ttx> alvin: iiuc you have two issues: one is that it fails to mount things, and the other is that it fails to tell you about it. Is that correct ?
[14:05] <ttx> alvin: about the first, are you running on LVM ? about the second, what are your kernel options ? About both, do you already have opened bugs about those ?
[14:06] <alvin> franjpr: look at what ttx posted for your mountall debug options. (I'll enable that too)
[14:07] <ttx> alvin, franjpr: fwiw I usually make a copy of /dev/mountall.log to /ev/copy.log while in M mode, I saw some file clobbering when I made the install continue after that
[14:08] <ttx> franjpr: that said, I don't think it would work for 9.10
[14:08] <alvin> ttx: Actually, I have more than those 2 issues with Lucid. Still examining a bit before filing the bug. Plymouth fails to tell me when using ubuntu-server, but the keys do work (S, M,...) The other one is: as soon as I create a snapshot, one machine gets into grub rescue, the other will not be able to mount all filesystems.
[14:09] <ttx> alvin: I'll try to reproduce the "fails but keys do work" one
[14:09] <alvin> and the last is: when creating a snapshot, sometimes the snapshot gets mounted instead of its parent. I filed that one yesterday and it's only on upgrades. > bug 563117
[14:09] <alvin> ttx: Easiest way is to snapshot an lvm volume
[14:10] <alvin> and reboot
[14:10] <ttx> I know that Foundations are still working on a slew of mountall issues
[14:10] <alvin> WHat is Foundations?
[14:11] <ttx> the group that works on ubuntu foundations for both desktop and server, that includes the boot process
[14:11] <alvin> Ah, cool
[14:12] <alvin> I see the trouble with lvm snapshots on both server and desktop, but they react differently. Probably because the server has mdadm raid1 too. That one is the worst because it only gets into as far as grub rescue.
[14:25] <Jeeves_> Any apt specialists here?
[14:29] <Pici> Depends what the question is.
[14:32] <Genk1> hello, is there a way to crypt all the traffic network of a LAN ? I mean for every protocol used in the LAN !
[14:33] <Jeeves_> Pici: It's about the gpg checking.
[14:33] <Jeeves_> But I found what's going wrong
[14:34] <Jeeves_> All I need to do now is fix my mirror :)
[14:34] <Pici> Jeeves_: Why not just ask the question and we'll see if we can answer.  Or nevermind.
[14:34] <alvin> Genk1: You can tunnel about anything over ssh, but I have no experience with that. All traffic is a bit much, no?
[14:35] <Genk1> alvin, yes it's a high traffic !
[14:35] <Jeeves_> Pici: Nevermind :)
[14:37] <alvin> I think encryption is easier for point to point connections. For example, file transfers over sftp. CIFS also has encryption, but that is really th protocol itself. If you want all X traffic in the lan encrypted, I wouldn't know the answer to that.. You can tunnel X through ssh easily, but a broadcast is something else.
[14:39] <Genk1> alvin, hmmm.. I see !!
[14:39] <Genk1> thank you
[14:56] <_ruben> Genk1: that's pretty much what ipsec is used for
[14:59] <zul> Daviey: ping wtf with the ubuntu-fortunes-server?
[15:00] <Daviey> zul: what about it?
[15:00] <Daviey> zul: it's all fixed now.. but it seems the MIR wasn't actioned ages ago.
[15:00] <zul> Daviey: ok
[15:01] <Daviey> zul: fwiw, i thought it was in main already - prepaired an upload yesterday, and didn't notice.
[15:01] <zul> Daviey: gotcha
[15:02] <iclebyte> anyone familar with drbd? after creating and performing the inital sync, should you only create the file system on /dev/drbd0 on the primary note? i.e. will the creation of the filesystem replicate to the slave node via the block device?
[15:09] <zul> ttx: is it your day for daily triage?
[15:10] <ttx> zul: no, was yesterday
[15:10] <ttx> today's smoser's
[15:10] <smoser> yeah. my triage day.
[15:11] <alvin> You assign each other triage jobs?
[15:12] <smoser> alvin, its just that each of us has agreed to spend time doing triage one day of the week.
[15:13] <smoser> you, and anyone, is more than welcome to triage server bugs any time you'd like (and I suggest you help out on Thursdays :)
[15:13] <smoser> https://bugs.launchpad.net/ubuntu/+bugs?field.searchtext=&orderby=datecreated&field.status%3Alist=NEW&field.importance%3Alist=UNDECIDED&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_supervisor=ubuntu-server&field.bug_commenter=&field.subscriber=&field.component-empty-marker=1&field.status_upstream-empty-marker=1&field.omit_dupes.used=&field.omit_dupes=on&field.has_patch.used=&field.has_cve.used=&field.affects_
[15:13] <smoser> me.used=&field.tag=&field.tags_combinator=ANY&field.has_no_package.used=&search=Search
[15:13] <smoser> wow. silly long link.
[15:13] <zul> smoser: slacker
[15:14] <alvin> I'm currently at work. We use ubuntu-server a lot, so a bit of triaging during work hours couldn't hurt
[15:14] <smoser> http://ubuntu-server-new-bugs.notlong.com
[15:15] <alvin> If you're triaging now, try rebooting lucid after you made a snapshot of an lvm parition. I have yet to file the bug though.
[15:15] <ttx> link: https://wiki.ubuntu.com/ServerTeam/KnowledgeBase#Bug%20Triager%20resources
[15:16] <mathiaz> zul: bug 561750
[15:17] <mathiaz> zul: it seems that squid reload kills squid
[15:17] <zul> mathiaz: ok ill have a look
[15:18] <alvin> Ah, triaging is not testing, but setting the importance and status of bugs. I see.
[15:18] <ttx> mathiaz !
[15:18] <mathiaz> ttx: o/
[15:24] <Italian_Plumber> so... wget doesn't recursively follow javascript links?
[15:31] <_ruben> i doubt wget does any javascript at all
[15:32] <Italian_Plumber> well fudgenuggets
[15:41] <Italian_Plumber> what about links that exist on the pages as <area> instead of <a>... or is that javascript
[15:53] <ttx> alvin: reproduced, filing bug
[15:54] <alvin> ttx: Wait, I just filed two
[15:54] <ttx> ah
[15:54] <ttx> bug # ?
[15:54] <alvin> bug 563895
[15:54] <alvin> looking for the other one
[15:55] <alvin> bug 563902
[15:55] <alvin> I hope I chose the right packages
[15:55] <ttx> ok, will file for the lack of M / S prompt
[15:55] <alvin> Good, I didn't file that one. Please, post the number here so  can subscribe
[16:00] <persia> So, I installed a new lucid server, and have a new, cool, byobu theme.  On an upgraded server, I still have the old theme.  What do I need to do to get the new theme on the upgraded server?
[16:01] <kirkland> persia: F9-> Select background (black), Select foreground (white)
[16:01] <persia> kirkland: Thanks!
[16:01] <kirkland> persia: or ... printf "BACKGROUND=k\nFOREGROUND=w\nMONOCHROME=0" > ~/.byobu/color
[16:02] <kirkland> persia: if you want the "monochrome" theme to make the grey-and-white look of Lucid
[16:02] <jasonmchristos> what tracker software does torrent.ubuntu.com use? i want to setup my own tracker
[16:04] <ttx> alvin: bug 563916
[16:04] <persia> kirkland: That's even better: now if I get a custom selection I can just scp it about.  Thanks again.
[16:05] <alvin> ttx: thanks
[16:07] <kirkland> persia: cheers, yeah
[16:12] <kirkland> persia: i actually use different color combinations for each host (so that i recognize what machine i'm on by a visual cue)
[16:12] <Pici> Me too
[16:12] <persia> Hmm.  So maybe I oughtn't have "fixed" that :)
[16:13]  * persia reverts to prevent a future mistake in advance
[16:17] <kirkland> persia: ?
[16:17] <kirkland> persia: it's totally up to you
[16:17] <persia> kirkland: Yeah, but it's a *really* good idea to have a visual cue to differentiate hosts :)
[16:18] <persia> kirkland: Also, thanks for the seed changes: http://cdimage.ubuntu.com/ubuntu-server/ports/daily/current/ is all <700MB now.
[16:18] <kirkland> persia: http://people.canonical.com/~kirkland/colors.png
[16:19] <persia> Oh my.  I don't have quite that server count, but I can see why you almost need it.
[16:20] <binBASH> I rebooted the machine where cloud controller etc. is on, now it shows 0 free vms :/
[16:20] <binBASH> anyone knows what's wrong?
[16:20] <binBASH> in logfile I see it gets timeout on select()
[16:39] <mathiaz> ttx: bug 423252
[16:39] <mathiaz> ttx: ^^ we should have a look at this one as well
[16:40] <ttx> ew
[16:41] <mathiaz> hggdh: are you using the uec test rig?
[16:42] <ttx> mathiaz: added to watch list on ReleaseStatus page
[16:42] <hggdh> mathiaz: was, but I can give it up
[16:43] <mathiaz> hggdh: kirkland and I are going to look at the multi-machine topo
[16:43] <hggdh> mathiaz: a Q: the /etc/eucalyptus/eucalyptus-ipaddr.conf should point to the correct addresses of the CLC, CC, Walrus, etc, correct?
[16:43] <mathiaz> hggdh: is this the topo that is currently installed?
[16:43] <mathiaz> hggdh: yes
[16:43] <hggdh> mathiaz: lucid-adm64-topo2
[16:43] <hggdh> mathiaz: they are not
[16:43] <hggdh> mathiaz: to the point I am not sure how it even works
[16:47] <mathiaz> hggdh: they all look good
[16:48] <mathiaz> hggdh: eucalyptus-ipaddr.conf is only used by the avahi publishing jobs
[16:48] <kirkland> hggdh: mathiaz: actually, /etc/eucalyptus/eucalyptus-ipaddr.conf isn't very accurate ... i started working on a fix for that (using avahi, and grepping the logs)
[16:48] <mathiaz> hggdh: to pick up the correct IP address to publish on the nework
[16:48] <hggdh> ah, OK
[16:48] <kirkland> hggdh: mathiaz: but i figured it was too late to commit that kind of a change
[16:49] <hggdh> mathiaz, kirkland: the rig is all yours. Want me to logoff?
[16:50] <mathiaz> hggdh: ok thanks
[16:50] <ttx> hggdh: you have three work items on https://blueprints.launchpad.net/ubuntu/+spec/server-lucid-uec-testing about testing of the Beta2 milestone, it would be good to cover them soon, with whatever topology works at the time... If already done please mark those work items as "DONE"
[16:50] <ttx> because next week we'll switch to testing RC candidates
[16:51] <ttx> so the beta2 post-milestone tests need to be covered by then
[16:51] <mathiaz> ttx: doesn't the WI cover *all* topologies?
[16:51] <ttx> or should they be considered covered already ?
[16:51] <hggdh> ttx: as soon as mathiaz and kirkland are done
[16:51] <hggdh> ttx: only the all-in-one was tested
[16:51] <ttx> mathiaz: I prefer partial results this week to no results at all for B2
[16:52] <mathiaz> hggdh: have you already tested beta2 on topo1?
[16:52] <hggdh> mathiaz: I had, yes
[16:52] <ttx> hggdh: you still have a few days, so let's give all the topologies a chance :)
[16:53] <ttx> hggdh: just saying that next week you'll be busy with RC candidates validation, so we'll have to go with whatever you can come up with by... Monday
[16:53] <hggdh> ttx: ack
[16:53]  * hggdh runs around
[16:55] <e-DIO-t> seeya
[17:30] <kirkland> smoser: ping
[17:30] <kirkland> smoser: see nurmi's noted;  neil +1'd your euca2ools change
[17:30] <kirkland> smoser: shall i sponsor and get that uploaded?
[17:45] <matt3206_> is this purely server support
[17:46]  * matt3206_ wonders wtf is everyone
[17:48] <ScottK> matt3206_: This is Ubuntu Server support and development.
[17:49] <matt3206_> thanks
[17:49] <matt3206_> i need multimedia support
[17:49] <smoser> kirkland, please do sponsor that
[17:49] <matt3206_> do you know which channel i could find that in?
[17:50] <smoser> kirkland, bug 551847
[17:51] <matt3206_> anyone work with sun micro system hardware?
[17:52] <ScottK> matt3206_: General Ubuntu support is #ubuntu
[17:52] <jzy> anyone using nginx on ubuntu-server?
[17:52] <matt3206_> cool thanks
[17:52] <matt3206_> scott you ever heard of an NCC
[18:32] <mathiaz> hggdh: I've fixed the bug on the uec test rig
[18:32] <mathiaz> hggdh: and pushed it to my branch
[18:32] <mathiaz> hggdh: I've also merged your console output patch
[18:32] <mathiaz> hggdh: everything can be found at lp:~mathiaz/+junk/uec-testing-scripts
[18:32] <mathiaz> hggdh: could you mark your branch merged?
[18:34] <hggdh> mathiaz: ẁill do
[18:34] <hggdh> mathiaz: and I will update my local copy of your branch
[18:34] <hggdh> mathiaz: what was it?
[18:37] <hggdh> mathiaz: forget it, read the comments ;-)
[18:38] <hggdh> mathiaz: can I use the rig now?
[18:38] <mathiaz> hggdh: yes
[18:38] <hggdh> mathiaz: thank you
[19:07] <hggdh> ttx: I do not have access to the beta2 images from the rig, can I use the daily?
[19:10] <aaaccc> !search: german: No such pack or chat room
[19:12] <cemerick> is there any easier way to get the unlimited java policy files installed, short of going to oracle's site?
[19:12] <RoyK> cemerick: aren't those in the repository?
[19:13] <hggdh> nowadays I do not know, but Sun used to require you to get there and accept the conditions
[19:13] <cemerick> RoyK: not that I can see
[19:13] <cemerick> hggdh: they're totally unrestricted at this point *shrug*
[19:13] <RoyK> apt-get install ubuntu-restricted-extras ?
[19:16] <cemerick> RoyK: sheesh, not sure how else I would have found that.  Been googling for a while.  Thank you very much :-)
[19:18] <RoyK> that installs a bunch of stuff, though
[19:21] <cemerick> I can live with that vs. carrying around another couple of jars or automating a wget from somewhere. :-)
[19:26] <binBASH> Hi RoyK
[19:26] <RoyK> hi, shelly :)
[19:28] <smoser> mathiaz, ping
[19:29] <binBASH> Hi smoser
[19:29] <smoser> hey binBASH
[19:37] <cemerick> RoyK: that doesn't seem to actually include anything related to JCE -- just a metapackage that happens to include the sun jvm I think...
[19:37] <RoyK> comwhat is JCE?
[19:38] <RoAkSoAx> kirkland, ping!!
[19:38] <cemerick> java cryptography -- the API for using AES, etc
[19:38] <cemerick> if you want to use larger keys, etc., you need to have the "unlimited strength" crypto policy files installed.  These used to be export-controlled, but are now free and clear.
[19:38] <RoyK> oh - I'm not a java person, really, except for the blackish fluid, so I don't know - sorry
[19:39] <RoAkSoAx> kirkland, I just upgraded and libvirt seems to have broken something since I cannot launch any VM from virt-manager
[19:39] <binBASH> RoAkSoAx: Sounds like you're doomed :/
[19:39] <RoAkSoAx> kirkland, it shows me something like: Error starting domain: could not remove profile for 'libvirt-f0530b2a-fe14-70cb-e036-7f50c06e2b67'
[19:39] <RoyK> is there something like a nice gui for kvm?
[19:40] <RoAkSoAx> binBASH, LoL i'm actually am... have a presentation later today and have to reinstall everything :'(
[19:40] <RoAkSoAx> RoyK, virt-manager?
[19:40] <RoyK> does that allow stuff like live migration?
[19:40] <RoyK> sorry - I'm on a slow link - asking too many questions :þ
[19:41] <binBASH> RoAkSoAx: I was doomed today. my Cloud Controller didn't show any free vms, neither totals. 6 hours later I noticed it was caused by a firewall rule :/
[19:42] <RoAkSoAx> binBASH, haha but in my case is totally local ;/
[19:42] <binBASH> RoyK: Maybe proxmox or abicloud, dunno if virt-manager does live migration.
[19:43] <binBASH> RoAkSoAx: Luckily I'm just testing, not productive yet :p
[19:43] <RoyK> heh - 25% of european air traffic is down... :)
[19:44] <RoyK> seems the Icelanders have made their payback for the finincial crisis
[19:44] <binBASH> hehe
[19:44] <binBASH> I read on a newspage that there will be a second erruption at another crate maybe.
[19:45] <binBASH> In history it was always like this...
[19:45] <RoyK> the bad thing is that Eyjafjallajökull has never erupted in known history (1200 years) without trigging Katla, which is the largest volcano in Iceland
[19:46] <binBASH> They wrote, maybe it could be like in a hollywood movie.
[19:46] <RoyK> the two eruptions so far are quite small
[19:46] <RoyK> if Katla wakes up, it'll be bad
[19:48] <RoAkSoAx> kirkland, it is not even creating new VM's with same kind of error
[19:53] <RoAkSoAx> kirkland, nevermind I just foudn the bug report :/ will have to wait till the fix is published
[21:11] <thebishop> hello
[21:12] <thebishop> got a question about nfs configuration.  I'd like for owner on the client side to appear as a different user than the owner on the host side.  is this possible?
[21:15] <alvin> thebishop: You want user mapping. I never did that, but I think it's possible. There's probably a config file for that kind of thing. (idmap?)
[21:17] <mathiaz> smoser: hi!
[21:17] <smoser> hey.
[21:17] <smoser> you've a mail from me, regarding server-dev team
[21:18] <alvin> apparently not idmapd.conf I just read the man page
[21:18] <mathiaz> smoser: yop - I'll reply to it
[21:18] <mathiaz> smoser: just discussed that with kirkland
[21:22] <alvin> thebishop: I found something for you. Look here: http://ubuntuforums.org/showthread.php?t=552579
[21:23] <alvin> map_static. I have need of such a thing too. Just never got around to it until you asked your question.
[21:30] <ewook> mathiaz: and you'll get one from me soon, since I don't know what do do ;)
[21:31] <ttx> hggdh: the idea of the B2 test is to use the milestone to get a reference point...
[21:37] <hggdh> ttx: but we did go thru them when beta2 was being released (at this time I was still doing it in a way not to be hit by the issue)
[21:38] <hggdh> and I did use the beta2 ISO
[21:38] <ttx> hggdh: then you should mark them as done :)
[21:38] <hggdh> ttx: anyway -- the rig does not have access to archive.ubuntu.com
[21:38] <hggdh> ttx: so this should be considered -- how do we get it open to archive?
[21:38] <ttx> there is some redundancy in those items
[21:39] <hggdh> yes I thought so, but a list is a list...
[21:39] <ttx> there is one about candidate testing and one about milestone testing. Obviously the last candidate is also the milestone :)
[21:39] <hggdh> indeed
[21:40] <hggdh> but I am going thru them, anyway, using today's image
[21:40] <ttx> ok, feel free to mark those as done, with a note about having run only topo1
[21:41] <hggdh> ttx, I only did not run the multi on the last candidate
[21:41] <ttx> hggdh: I'll let you handle it, going to bed :)
[21:42] <ewook> aw man. my landscape trial just died out, without me even testing it once =(
[21:44] <hggdh> ttx: marked done (but still testing the daily ;-)
[21:44] <hggdh> ewook: it is interesting, been testing it
[21:46] <ewook> hggdh: oooh. what did you think of it?
[21:46] <kirkland> smoser: RoAkSoAx: sorry, been busy ... just catching up to backlog
[21:46] <kirkland> smoser: sponsoring 551847 now
[21:46] <smoser> ok
[21:47] <hggdh> ewook: seems really nice, only thing is to have mail filters set if you subscribe to all alerts
[21:48] <RoAkSoAx> kirkland, np :), the fix was already in the archives but by the time It wasn't built yet, so I just build it locally and install it :)
[21:48] <hggdh> ewook: really nice to command selective updates remotely
[21:49] <ewook> hggdh: hurpf.. I'll try to beg for a new trial when I know I have time for it :)
[21:56] <kirkland> smoser: the patch at https://bugs.edge.launchpad.net/ubuntu/+source/euca2ools/+bug/551847 does not apply cleanly to euca2ools in lucid
[21:56] <kirkland> smoser: please doctor accordingly
[21:56] <kirkland> smoser: neither hunk
[21:56] <smoser> what ?
[21:57] <smoser> i'll look at it
[21:58] <smoser> kirkland, it applies here to trunk of lp:~ubuntu-core-dev/eucalyptus/euca2ools
[21:59] <smoser> is that not where this should be going ?
[21:59] <kirkland> smoser: hrm
[21:59] <kirkland> smoser: what rev is head?
[22:00] <thebishop> alvin, the thing i'm consistently seeing about user masking is that it doesn't work
[22:00] <smoser> kirkland, 277
[22:00] <thebishop> if you have the UIDs don't match between host and client, inconvenient things can happen
[22:02] <alvin> That's why I make sure users have the same UID. It's a bit of a pain though. I have yet to try that masking so I can't say whether it works or not.
[22:04] <sandberg> I'm looking into bug #423252, but I'm afraid I'm stuck. Anyone feel like giving a hand, or just some hints?
[22:04] <smoser> kirkland, ^^
[22:05] <kirkland> smoser: bugger, i had a push fail
[22:05] <kirkland> smoser: no ...
[22:05] <kirkland> hmm
[22:05]  * kirkland starts over
[22:06] <smoser> i applied that both to your  ~coredev branch and to lp:ubuntu/lucid/euca2ools cleanly
[22:10] <kirkland> smoser: jeebus.  my fault.  trying to do too many things at once
[22:10] <kirkland> smoser: applies perfectly
[22:10] <kirkland> smoser: when i apply the right patch
[22:10] <kirkland> smoser: and use p0
[22:10] <kirkland> yeesh
[22:20] <kirkland> RoAkSoAx: okay, good, so you're back on track with libvirt?
[22:21] <RoAkSoAx> kirkland, you mean if its working correctly? Yes everything is working smoothly.
[22:23] <kirkland> RoAkSoAx: cool
[22:24] <RoAkSoAx> kirkland, btw I'm almost done with school so I'll be also be poking you about testdrive
[22:26] <kirkland> jdstrand: https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/564075
[22:27] <kirkland> jdstrand: can you take a look at the dmesg output attached to that bug?
[22:27] <kirkland> jdstrand: http://launchpadlibrarian.net/44314688/bug.txt
[22:28] <kirkland> jdstrand: there's an apparmor issue in the dmesg, it looks like
[22:28] <kirkland> jdstrand: but also a kernel bug
[22:28] <jdstrand> kirkland: have the person update to .21. jjohansen did a lot of work to fix some of these issues
[22:28] <kirkland> jjohansen: can you take a look at that stck trace to?
[22:29] <kirkland> jdstrand: ah
[22:30] <kirkland> jdstrand: jjohansen: bug updated
[22:30] <jdstrand> kirkland: also, make sure that the person has the latest libvirt ubuntu24 from today
[22:31] <jdstrand> kirkland: once the person has ubuntu24, virt-aa-helper denied reads are non-fatal (it is virt-aa-helper trying to see if the file has a backingstore-- something that a file in /dev/mapper won't)
[22:33] <kirkland> jdstrand: thanks
[22:33] <jdstrand> kirkland: what bug is that? is the xml available?
[22:33] <jjohansen> kirkland: interesting trace
[22:34] <kirkland> jdstrand: jjohansen: https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/564075
[22:34] <jdstrand> ah 564075
[22:35] <jjohansen> kirkland: yeah I caught that
[22:36] <jdstrand> kirkland: actually, you can see in the dmesg that the denied messages were non-fatal: right after those, you get a profile_load of the VM-- that means virt-aa-helper did its job
[22:36] <jjohansen> I'm subscribed
[22:50] <arthurjohnson> If you have any servers running clamav, you might want to make sure its 0.95 or later.
[22:51] <arthurjohnson> I'm getting segfaults, causing a race condition, on my older mailservers.
[22:59] <hggdh> kirkland: got some failures on a stress test on euca
[22:59] <kirkland> hggdh: oh, interesting
[22:59] <kirkland> hggdh: bug filed?
[23:00] <hggdh> kirkland: not yet, just ended a stress on single, (in)famous topo2
[23:01] <hggdh> kirkland: trying to figure out how many different failures there were
[23:03] <kirkland> hggdh: pastebin the results somewhere that mathiaz and i can see them
[23:20] <hggdh> kirkland, mathiaz: http://people.canonical.com/~cerdea/single_test.log.2010-04-15_171304
[23:25] <kirkland> hggdh: is this against beta2, or today's archive?
[23:27] <kirkland> hggdh: also, what version of the kernel, qemu-kvm, and libvirt are on those NCs ?
[23:37] <hggdh> kirkland: all based on today's daily, 20100415
[23:37] <kirkland> hggdh: and the guest image?  same?
[23:37] <kirkland> smoser: around?
[23:39] <smoser> whats up kirkland
[23:39] <hggdh> kirkland: uec is also from 20100415
[23:39] <kirkland> smoser: did the newest kernel make it into today's uec image build?
[23:40] <kirkland> hggdh: smoser: in http://people.canonical.com/~cerdea/single_test.log.2010-04-15_171304, looks like 2.6.32-20-server
[23:40] <smoser> http://uec-images.ubuntu.com/lucid/20100415/ has manifests of whats in it
[23:40] <smoser> http://uec-images.ubuntu.com/lucid/20100415/lucid-server-uec-amd64.manifest
[23:40] <smoser> linux-image-2.6.32-20-virtual 2.6.32-20.30
[23:40] <kirkland> we should be testing 2.6.32-21-server at this point
[23:40] <kirkland> smoser: thanks
[23:40] <smoser> it pulls whatever is in the archive
[23:40] <kirkland> smoser: timing issue then
[23:40] <kirkland> smoser: can you push the button that says "make a new uec image now" ?
[23:41] <smoser> i can. the whole thing takes 2 hours or so
[23:41] <smoser> it will organically pop out in about 5 or 6
[23:41] <hggdh> then we can wait for tomorrow's
[23:41] <smoser> so pushing button now saves 2
[23:41] <smoser> err 4.
[23:43] <smoser> yesterdays' thing was published at 03:39:14 UTC
[23:43] <smoser> so, right at 5 hours from now