=== c0nv1ct_ is now known as c0nv1ct [02:14] anyone here got any exp with pfsense? === dendro-afk is now known as dendrobates [02:36] New bug: #754210 in bacula (main) "package bacula-director-mysql 5.0.1-1ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/754210 [04:00] Im having a strange issue with 10.04.2 - it slows down after a few days of uptime or heavy data copying to it over the network. Gets so slow that its tough to login and run commands. Once top or similar actually ran though, no CPU Load or runaway process visible at all. iostat, free, top, anything does not show high load. [04:01] Running a apt-cache search sensors (example) , is SLOW.. after a system reboot, everything is peachy again, up until sometime later, especially after copying a lot of files to it (its a backup fileserver) [04:20] Why does mail from apt-listchanges go to /var/mail/mail when it says it is emailing root? I don't see an alias for user "mail" in /etc/aliases [04:24] Daviey: just FYI for your morning: today's ISO was missing python-psutil; as such, euca does not install from ISO [05:01] Hi, When i am loging to my server its taking long time...? [05:02] How do i reduce this login time [05:15] why do ppl think irc is the fastest technology? [05:15] you can't just go in a crowd and start yelling about things [05:16] amstan: *I* can [05:16] especially if the crowd is afk [05:16] lol [05:18] the trick is to take all your clothes off [05:18] then start yelling [05:22] lifeless: In the IRC context, clothing removal would seem somewhat pointless and be unlikely to have much effect -- unless IRC clients are secretly enabling webcams these days?? :) [05:24] Only xmms2-irc [05:24] jmarsden: I was refusing the rl analogy;) [05:24] "real" life... who needs analogue signalling anyway [05:52] anyone around that could help me evaluate is Ubuntu Server would be a right fit for what I want to do? [05:56] gemclip: It might help to know first what plans you have [05:59] I have been a windows server guy for years and I am trying to migrate over and test *nix servers at home. I would like to setup a NAS server and a Multimedia server (streaming ect.) but I havent worked with nix much [05:59] I will be setting them up under vmware [06:01] Only stuff ive worked with in linux is BackTrack at work [06:02] I have a mac book, a ubuntu desktop and 4 pc's rnning windows 7 in the lab env [06:04] the machine I am setting up the vmware on is a i7 980x with 12gb ram and 10tb divespace [06:04] If you just want a NAS anything will do [06:04] streaming is a biggie I would like to find an open source solution === dendrobates is now known as dendro-afk [06:06] IMO the main benefit of Ubuntu is what it gets from Debian -- a strong Policy document which requires packages to "play nice" with one another, and infrastructure (dpkg and apt) to reliably distribute binary packages and their dependencies and updates. [06:07] Ubuntu also has better security than Debian (thanks to Kees Cook et al) and a "release train" model rather than a "when it's ready" release model. [06:07] The latter is useful for strategic planning in corporate environments. [06:09] I really like what Ive seen so far with ubuntu. The learning curve is a little rough buit I have been trying to stay away from the gui til I get comfortable under the hood [06:10] theres a gui ? [06:10] No server should have a GUI [06:11] lifeless:you can start one lol [06:11] We advise against doing so [06:11] I was talking linux in general [06:11] not just for server. I am trying to move over to linux a little at a time [06:14] but ultimately server is the goal [06:14] servers are a good start for learning linux [06:16] do you have any good websites you would point a newbie who wanted to setup a NAS and webserver. Any how-to's? [06:16] !RUTE [06:16] documentation is to be found at http://help.ubuntu.com and http://wiki.ubuntu.com - General linux documentation: http://www.tldp.org - http://rute.2038bug.com [06:16] thank you. Ill start my searching there === michael is now known as Guest32588 [06:20] gemclip: you should look into sftp, samba, and software raid [06:20] gemclip: that should get you covered for a nas [06:20] unless you wanna get fancy with nfs [06:21] I bought a block of static IP's I wanted to setup a server to be able to access away from home and Let some of my neighbors connect to it as well [06:22] gemclip: static ips are overrated, i just do with noip [06:22] I use 2 for work. I have to have them but it leaves me 3 that I can use personally [06:23] plus it only 10 bucks a month [06:25] I have a exchange server running 2008 right now but I have to keep that on its current ip [06:25] I could setup a 2008 server but I wanted to broaden my horizons [06:26] lol getting bored with windows and its expensive [06:26] yes [06:28] is there anything windows 2008 can do that ubuntu server cant? functionality wise [06:29] unix people like to do admining in different ways [06:29] usually custom sollutions [06:29] so there's not a lot of polish in those areas [06:29] (read: active directory) [06:31] is there group policies and user level control in U/S? equiliv to win servers? [06:32] well.. on unix you have groups and users, you can associate permissions for those on files [06:32] Better [06:32] and since everything is a file, ...yes [06:33] cool. I'll do some reading. Thanks for all your time [06:33] gemclip: are you a developer as well? [06:34] yeah but mostly .NET stuff [06:35] well.. regardless [06:35] it's a lot nicer deving under unix [06:36] with automated package management, you need a new library.. you can install it and have it end up in the proper place for compiling [06:36] what what Ive read there is alot of C compliling tyoe stuff [06:37] oh yeah, that stuff is pretty simple [06:37] fat fingers tonight [06:41] is there diffrent types of Ubuntu servers or just what you install on the server? [06:43] gemclip: Usually you have a base install of something like LAMP ( linux-apache-mysql-php) and then customise [06:44] gemclip: it's pretty much the same OS, with different software preinstalled [06:44] you can switch between them at will [07:06] Should I read up on UEC or wait? [07:13] Do you care about cloud stuff? [07:14] Not right now. It will be nice to know for any clients I may work with down the road. But I think I should just start with basic server install and go from there [07:15] I found the server guide so Im reading that atm [07:17] just waiting for the ISO to finish downloading [07:25] gemclip: which version? === skaet is now known as skaet_afk === genii-around_ is now known as genii-around === WinstonSmith_ is now known as WinstonSmith === genii-around_ is now known as genii-around [10:00] Im looking to get my server set up with LVM/RAID in a mirrored config. can someone guide me through this? === morpeth_ is now known as alandddd === twister004_ is now known as twister004 [11:13] jamespage: would you have some time for questions and general info on hudson and automated server testing later today? [11:15] TeTeT: I do; TBH have time now or can do later if you are busy [11:17] jamespage: I'm about to leave for lunch (5 mins), so later would be better. I need to get myself up to speed with the blueprints and would love to do some testing myself. So I guess right now I'm looking more for guidance on getting started than real questions [11:18] OK; lets catchup later then - say 1500 UTC? [11:20] jamespage: 1500 UTC is fine, just after my 1:1 [11:20] TeTeT: great - speak then === twister004_ is now known as twister004 === twister004_ is now known as twister004 [11:37] every now and then i would see only 5M free ram with a lot of it eaten by the buffers, sometimes even around 1G. Is there a way i can limit the buffers? [11:42] it's not much of an answer, but I don't believe that's really a problem. buffers will be dropped as soon as the memory is required for something more useful [11:42] but 'free' is the least useful state for ram [11:43] buffers are for your benefit === twister004_ is now known as twister004 [12:10] How does one set a static IP from the CLI? [12:11] moin \sh [12:12] jussi: do you want to set it up permanent? [12:12] yes [12:12] <\sh> moins [12:12] this server needs to sit at the same place always [12:13] so check http://www.cyberciti.biz/faq/setting-up-an-network-interfaces-file/ [12:13] ahh, excellent, thank you [12:13] welcome [12:17] binBASH: what are the => Network ID: 192.168.1.0 => Broadcast IP: 192.168.1.255 for? [12:20] jussi: read this http://en.wikipedia.org/wiki/Broadcast_address [12:21] and Network id is the subnet your pc is on ;) [12:21] New bug: #754522 in autofs5 (main) "autofs not starting if no nfs-mount in fstab" [Undecided,New] https://launchpad.net/bugs/754522 === twister004_ is now known as twister004 [13:16] I'm having some really strange issue. I rebooted my machine and get a "mountall: mount / [xxx] terminated with status 32" === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [13:48] New bug: #749402 in mysql-dfsg-5.1 (main) "package mysql-server-5.1 5.1.37-1ubuntu5.5 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/749402 [13:56] New bug: #752298 in bind9 (main) "package bind9 1:9.7.0.dfsg.P1-1ubuntu0.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/752298 [14:02] a bug on the ISO (package not installable) should be opened against which package? [14:06] hggdh: What's not installable? [14:06] ScottK: python-psutil [14:06] Do you know why? [14:06] it seems to be missing from the ISO [14:07] Why is it needed on the ISO? [14:07] eucalyptus -- eucalyptus-common depends on it [14:08] It's in Universe. That's why it's not on the ISO. [14:08] Looks like it's never been in Main, so it needs a Main Inclusion Report. [14:08] hallyn: ive made some progress, yes [14:09] jbernard, do you need any help? [14:09] hallyn: i was a bit ambitious and thought I could squash several other bugs in addition to the security issue, but that quickly turned into a small nightmare [14:09] hggdh: So the bug should be filed against python-psutil and ubuntu-mir subscribed. [14:10] jbernard: are all the bugs listed in lp? [14:10] hallyn: i have a package ready for 0.37.1 with the security issues fixed, but there is still a problem with upgrade [14:10] hallyn: between lp and bts, yes [14:11] hallyn: i now understand why tmpfs is needed [14:11] ScottK: duh. I really should have done my homework :-( thank you [14:11] ah, yes. oh, right, i'd forgotten - is one of the things you're adding my upstart job? [14:13] hallyn: yeah, that is another thing i need to do. As I understand, dh_installinit in debian was updated to include knowledge of upstart jobs, but it seems it's not fully supported yet [14:13] hallyn: vorlon posted an explaination in debian-devel, ill see if I can dig it up [14:15] jbernard: oh boy [14:15] hallyn: my plan now is to upload 0.37.1 tonight, and then tackle the other issues in a subsequent upload(s) [14:15] what does .1 address? [14:15] hallyn: mostly the security issues [14:16] hallyn: and a couple of other smaller patches [14:16] hallyn: nothing major [14:16] ppetraki: anything in particular you want to ask for for information in bug 644489 [14:16] Launchpad bug 644489 in multipath-tools "constantly changes /dev/disk/by-id/{scsi,wwn}-* LUN symlinks with multipathing" [High,Confirmed] https://launchpad.net/bugs/644489 [14:16] jbernard: all of the security issues though? [14:16] yep, both of them [14:16] cool [14:17] * ppetraki looking [14:17] jbernard: someone on #lxcontainers yesterday had a problem with libcgroup and lxc [14:17] jbernard: libcgroup was mounting each cgroup separately iiuc. is that expected? [14:18] jbernard: separately in different locations? It should just create the dir strucure under /sys/.. but it would depend on which version they're on [14:19] they said they were on natty [14:20] i've not been able to (and am not able to) reproduce right now. [14:20] hallyn, I wish they would have pointed out where in the udevadm monitor log the "legitimate event" was observed [14:20] ppetraki: yeah, i'm going to ask for a whole new set of data. starting with running his stuff with -v4 and posting daemon.log [14:21] ppetraki: i noticed it was another netapp so figured you migth have ideas :) [14:21] hallyn: that sounds very odd, ive not seen that myself. do you remember their nick? [14:21] hallyn, can't wait to get one or two :) [14:22] jbernard: it was dhubbard_ if i recall right [14:23] hallyn: cool, thanks [14:23] hallyn: i have to run to work, ill get .1 uploaded tonight [14:23] hallyn: and then we can go from there [14:23] great, thanks. ttyl. [14:23] hallyn: cheers [14:29] hallyn, those additional logs should be enough, udev monitor alone can never tell you why the kernel emitted this event [14:30] hallyn, if I had to guess, I would say a LIP occurred, or something on the FC bus that forced discovery [14:30] all right, let me ask for those logs now before i get back into what i was trying to do [14:31] cool, thanks [14:34] morning all [14:35] RoAkSoAx, o/ [14:35] Daviey: o/ [14:36] RoAkSoAx, Having fun? [14:36] Daviey: well my day is just starting but I guess I will :) [14:36] RoAkSoAx, python-psutil needs a MIR :) [14:36] hggdh noticed it. [14:37] * RoAkSoAx checks [14:38] Daviey: alright I'll add it to my todo for today [14:39] RoAkSoAx, awesome [14:51] zul: argh [14:51] zul: lxcguest bug - containers wn't shut down all the way [14:52] hallyn: do you have a fix for it? [14:52] to anyone using either puppet or chef: why choose one over the other? [14:52] hi all [14:52] zul: yeah [14:52] hallyn: bzr branch me baby [14:53] zul: will do. creating a bug first. [14:53] I just installed this box, and at the end of the installation, it asks if it should update security updates automatically - I was a bit quick and chose no automatic updates - how can I change that after installation? === rizzuh_laptop_ is now known as rizzuh_laptop [15:02] RoyK, https://help.ubuntu.com/10.04/serverguide/C/automatic-updates.html [15:03] Daviey: MIR's need FFe? [15:03] Daviey: thanks [15:05] RoAkSoAx, I believe so. [15:06] Daviey: does this look good enough to you then? bug #754661 [15:06] Launchpad bug 754661 in python-psutil "[FFe] [MIR] python-psutil" [Undecided,New] https://launchpad.net/bugs/754661 [15:07] RoAkSoAx, Yeah... the thing i am missing is how it got introduced as a new depends. [15:07] Ahh.. i see [15:08] scrub that [15:08] zul: bug 754655, bzr tree attached. [15:08] Launchpad bug 754655 in lxc "lxc guests on natty are not shutting down" [High,In progress] https://launchpad.net/bugs/754655 [15:08] s/attached/linked/ [15:08] hallyn: cool...gimme a sec [15:09] Daviey: hjehe yeah it was a bbug filed against it by hggdh :) I guess we all assumed it was in main as the installation didn't really fail [15:09] Daviey: I now see it in component-mismatches though [15:11] RoAkSoAx: I was also assuming it -- I needed ScottK to delicately cattle-prod me to the MIR [15:12] freaking firefox [15:12] hggdh: hehe yeah fortunately it is not really a difficult one as it is a small package [15:13] RoAkSoAx: yes, I was starting to look at it, seems clean [15:13] indeed [15:19] Daviey: is the problem with upstart and the IFACE thing still present? === skaet_afk is now known as skaet [15:22] hallyn: uploaded [15:24] RoAkSoAx, yes [15:25] RoAkSoAx, SpamapS proposed a solution, which i've not yet tested. [15:25] Daviey: ok, so apparently bug #726769 bus closed by mistake :) [15:25] Launchpad bug 726769 in eucalyptus "package eucalyptus-common 2.0.1 bzr1255-0ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [High,Fix released] https://launchpad.net/bugs/726769 [15:25] RoAkSoAx, yup :( [15:26] Daviey: I think hggdh did the testing with not very succesful results [15:27] indeed. From not upgrading, to not starting [15:27] hggdh, hmm.. you didn't test SpamapS proposed solution, did you? [15:28] Daviey: SpamapS told me to add 'env IFACE=manual-start' in the eucalyptus-network.conf. I added it, and *then* upgraded [15:29] hggdh, Ahh... i thought it would need adding after dpkg craps out [15:29] Daviey: end result was the upstart processes went in to what looked like an infinite wait [15:29] well, not really infinite, just a long sleep [15:29] then sudo apt-get upgrade again [15:29] then this is not a solution [15:30] hggdh, no, it is - that would mimic it being in a new package. [15:30] ah well. I did ask -- a few times, mind you -- about what was to be done, and how. I worked on what I got back [15:32] hggdh, Ah, ok.. Once we've got the present issue resolved.. we'll revisit upgrades i think. [15:32] Daviey: indeed. We go back to beta1, and try again [15:33] and even better, we go from maverick [15:33] zul: thanks! [15:34] np === Pici is now known as Guest95063 === Pici` is now known as Pici === zz_carcinogen75 is now known as carcinogen75 [15:57] New bug: #754699 in eucalyptus (main) "Eucalyptus node (nc) fails to install, when done from PackageInstall" [Undecided,New] https://launchpad.net/bugs/754699 === michael_ is now known as minorbug [16:04] jamespage: back to get more info on jenkins/hudson and the automated testing :) [16:04] TeTeT: fire away:-) [16:05] jamespage: so I installed a Natty vm and saw that apt-cache search hudson | jenkins does not show anything - it has not landed in Natty, what were the problems? [16:06] TeTeT: so challenge has been around the size of this piece of work. [16:06] jamespage: it's massive I take from the look at the PPA ppa:hudson-ubuntu/testing [16:06] TeTeT: :-) [16:07] jamespage: so simply to much work for a release cycle to get this into the archive? Or was the change from hudson to jenkins also an issue? [16:07] TeTeT: yes it is quite big; however not far off original estimates; the intent for natty was to deliver alongside main release in PPA [16:08] TeTeT: so that we could be in a good place for Oneiric. [16:08] jamespage: as it is Java, did you have to package all the specific versions of the build dependencies (I guess through maven)? Or was there a shortcut somehow? [16:09] jamespage: yeah, wearing my corp services hat I count on 12.04 for delivering this tech to customers :) [16:09] TeTeT: no shortcuts; some of the versions of dependencies are slightly different, some are major versions different. [16:10] TeTeT: however it is hanging together OK; we have been using the package in the testing PPA for all of the ISO and ec2 testing for Natty and its been solid [16:10] jamespage: ok, let's stay on jenkins for a bit, will query on automated testing in a bit [16:11] TeTeT: the newer versions of maven-debian-helper (not yet in Ubuntu) are much better and automate most of the package production; [16:12] TeTeT: TBH the rename from Hudson to Jenkins did not create to much work [16:12] jamespage: I now have a jenkins webpage on port 8080 on my vm open. If I want to test something with jenkins, what are my next steps? URL is sufficient - is this one a good start or are there better tutorials? https://wiki.jenkins-ci.org/display/JENKINS/Use+Jenkins [16:13] TeTeT: thats a good place to start; the thing to understand about Jenkins is that although there is alot of functionality in the core product (enough for most basic requirements) [16:13] TeTeT: the really neat stuff is all managed through plugins [16:14] jamespage: ok, not sure if I understand right now, but most likely will when I had a closer look at it [16:15] jamespage: how much effort is it to test something? E.g. I have a couple of python scripts for testing UEC that themselves could use testing when I develop them. As an alternative I would have an open source board gaming tool that I would want to test - however jenkins can assist in testing graphical apps [16:16] TeTeT: so for your python scripts do you have unit tests? [16:17] jamespage: I fear not [16:17] jamespage: would this be a first step to implement them? [16:18] TeTeT: absolutely [16:18] TeTeT: All of the work I've done using Jenkins has been centred on integration [16:18] jamespage: ok - do I see this right that jenkins will help me to a) build some software, b) run built-in tests, c) run integration tests [16:19] TeTeT: whatever Jenkins is doing under the hood should be executable by the developers who are developing the code. [16:19] TeTeT: It will help you automate those steps and report on results - yes [16:19] TeTeT: lemme just dig out a tutorial for working with python - that might help as well. [16:21] TeTeT: http://www.rhonabwy.com/wp/2009/11/04/setting-up-a-python-ci-server-with-hudson/ [16:22] jamespage: thanks, guess that will give me more than enough to get started on jenkins. Onto automated testing for Natty [16:22] TeTeT: OK [16:22] jamespage: so the PPA is used for driving the automated iso testing for Natty? [16:22] TeTeT: URL first: http://jenkins.qa.ubuntu-uk.org/ [16:23] jamespage: he he, thanks, need a folder for jekins and automated testing now ;) [16:23] TeTeT: so I've helped put together 2 other projects which actually do the work under the hood [16:23] jamespage: so this is the test results page for Ubuntu 11.04 testing? [16:23] TeTeT: filtered for 11.04 ISO tests - http://jenkins.qa.ubuntu-uk.org/view/ISO-server-Natty/ [16:24] TeTeT: but broadly yes it is. [16:24] TeTeT: Jenkins provides us with all of the glue to make this stuff happen automatically [16:24] TeTeT: Monitoring for new images, triggering jobs when they appear and reporting on results. [16:25] TeTeT: the actual work is done by two other projects: ubuntu-server-iso-testing and ubuntu-server-ec2-testing [16:25] TeTeT: https://launchpad.net/ubuntu-server-ec2-testing and https://launchpad.net/ubuntu-server-iso-testing [16:25] jamespage: these projects are about defining and implementing the tests I take? [16:26] TeTeT: thats right - the -iso-testing project uses libvirt and KVM to do automated installs of Ubuntu, and then run tests within those installs once completed. [16:27] TeTeT: Its all written in python; we used normal unittest's to actually implement the tests we wanted to run [16:27] TeTeT: they are executed using subunit and the subunit2junitxml filter [16:27] jamespage: so when I look at http://jenkins.qa.ubuntu-uk.org/job/lucid-server-amd64_lvm/1/testReport/test/LvmTest/ I see that 3 (unit?) tests passed [16:27] RoAkSoAx: can you push your koan/ubuntu patch upstream [16:27] TeTeT: Jenkins is good at understanding Junit XML output - thats what you see in that link [16:27] jamespage: where would I find the actual test definition for testLvs, testRootMount, testVgs? [16:28] zul: I already did [16:28] k [16:28] zul: fw 5 patches yesterday to the cobbler ML [16:28] TeTeT: http://jenkins.qa.ubuntu-uk.org/job/lucid-server-amd64_lvm/ - the unittest file is called 'lvm' [16:29] It gets archived there for future references - its actually part of the underlying ubuntu-server-iso-testing package [16:29] jamespage: awesome [16:29] jamespage: do you collect any statistics on how many tests fail during a release cycle and how many pass? Just for getting an overview [16:30] TeTeT: not really - its more used as a healthcheck throughout the dev cycle and for reporting on candidate image testing prior to milestones [16:30] jamespage: and how do I find out what's going wrong when it fails: http://jenkins.qa.ubuntu-uk.org/job/natty_server_ec2/ looks like all is broken? [16:30] TeTeT: however you can see this [16:31] TeTeT: http://jenkins.qa.ubuntu-uk.org/job/natty-server-i386_samba-server/ [16:31] this one runs daily [16:31] jamespage: so 5 tests are run, most time one fails, at a few times two? [16:32] TeTeT: yes - there is a know issue with the unittest which causes the one failure - the others are actual failures [16:32] TeTeT: http://jenkins.qa.ubuntu-uk.org/job/natty-server-i386_samba-server/55/testReport/ (click on an error in the graph to see this view) [16:33] TeTeT: nmbd failed to start for some reason [16:33] TeTeT: So this project [16:33] http://jenkins.qa.ubuntu-uk.org/job/natty_server_ec2/ [16:33] jamespage: great, thanks for all the infos! [16:33] Is using a different type of job 'Matrix' === dendrobates is now known as dendro-afk [16:34] used for testing different combinations of setup; [16:34] jamespage: is it testing the deployment rather than the execution? [16:34] TeTeT: not sure I understand the question [16:35] * popey takes "ubuntu-uk" off hilight :) [16:35] jamespage: nevermind [16:35] sorry popey :-) [16:35] not your fault :) [16:35] I'll stop pasting links in a minute :-) [16:35] * popey glares at Daviey :) [16:36] popey needs a better IRC client that lets him exclude regexes :) [16:36] jamespage: I'll need to take a look at all this now, will most likely come back with some questions next week [16:36] TeTeT: no problem [16:37] jamespage: thanks a ton so far, much easier to get an insight than reading the blogs/wikis/documentation - overall it's huge [16:40] TeTeT: So the Ubuntu use is quite simple at the moment; other projects are doing more in-depth stuff - take a look here http://jenkins.openstack.org/ [16:42] jamespage: nice, there are builds happening [16:45] TeTeT: thats quiet - look at Drizzles - http://jenkins.drizzle.org/ [16:45] I'm following along here: http://www.howtoforge.com/perfect-server-ubuntu-10.04-lucid-lynx-ispconfig-2 and can't start chrooted bind. This is in messages... : operation="open" pid=16789 parent=16787 profile="/usr/sbin/named" requested_mask="r::" denied_mask="r::" fsuid=103 ouid=103 name="/var/lib/named/etc/bind/named.conf" [16:46] hggdh, seeing S3 failure? [16:46] jamespage: guess we're really lightweight users in comparison [16:46] hggdh, I tested the IFACE upstart change, and it seems to resolve the issue... [16:46] Daviey: OK [16:48] Daviey: back to kernel sru testing [16:53] hggdh, Did you see S3 failing tho? [16:53] Daviey: I saw failures on walrus -- what causes the instances to fail [16:54] hggdh, This is *NEW*... wtf. [17:16] zul: we do not have xen kernels right? [17:17] RoAkSoAx: depends on what type of xen kernels you are talking about [17:17] zul: in archive.ubuntu.com/ubuntu/dists/maverick/main/installer-amd64 for example [17:17] zul: if I were to try to netboot into xen using virt-install [17:19] isn't the linux-ec2 package a xen kernel? [17:19] RoAkSoAx: there are the kernels in http://uec-images.ubuntu.com -- they are DomUs [17:21] hggdh: right, but talking about this :) https://www.redhat.com/archives/virt-tools-list/2011-April/msg00036.html [17:27] RoAkSoAx: as semiosis points out, the linux-ec2 are DomU kernels === dendro-afk is now known as dendrobates [17:30] RoAkSoAx: what are you trying to do then i can provide you with a better answer [17:31] zul: basically, the doubt came when I was looking into forwarding a patch to virtinst, and I read this: https://www.redhat.com/archives/virt-tools-list/2011-April/msg00036.html [17:33] RoAkSoAx: the virtual kernels i think are domU kernels we dont support dom0 (yet) [17:33] smoser: do we run the virtual kernels or the server kernels on ec2? [17:34] virtual [17:34] also, there is a difference between "xen kernel" and dom0 [17:35] dom0 kernel is a linux kernel that runs under the xen kernel as the priviledged guest named 'dom0' [17:35] for that you need dom0 support in your kernel (which is what zul is asking for for 'o') [17:35] smoser: using xen on an ubuntu host? [17:36] where should I save variables that I need to run each time I login to my server? [17:36] smoser: but virtual/server is/was the same kernel so it shouldnt matter [17:36] to run xen, you need xen (a package that installs a hypervisor on your computer that then runs ubuntu as dom0) [17:36] MetaJake: .bashrc [17:36] virtual/server is *close* to being the same kernel. [17:36] smoser: yeah that I understand, and I didn't mean dom0 :) I guess I should have phrased my question correctly [17:36] thanks royk thats what I thought. I'll try this again. something didn't work last time I tried that. [17:36] smoser: right server just has a few more modules ;) [17:37] smoser: but yeah In my question was xen kernel as a guest (not as dom0) [17:37] last I checked, Xen wasn't very well supported on ubuntu - use kvm [17:37] RoAkSoAx, you *need* dom0 support in your kernel, or "ubuntu and xen" dont really work [17:37] (which is the case right now) [17:37] kvm works very well, though [17:37] zul, its not exactly that any more [17:37] far less hassle [17:37] -virtual is no longer a "sub flavour" [17:37] it is a flul "flavor" [17:38] smoser: ah ok [17:38] smoser: since when? [17:38] what's the difference between linux-virtual & linux-ec2 kernels? even ec2 instances use linux-virtual [17:38] smoser: right right. But what I mean was "If I have a xen running (let's say Debian), do Ubuntu have kernels to run as guests?" [17:38] natty and maverikc. [17:38] some doc I can read up on? [17:38] -ec2 is gone [17:39] ohhh ok :) [17:39] it was a heavily patched ubuntu kernel [17:39] we dropped that in maverick [17:39] and moved to using the '-virtual' kernel, which has 'pv_ops' support [17:39] which is what enables the generic -virtual kernel to run as a xen domU [17:39] this has proven somewhat buggy [17:40] * RoyK wonders why people use xen when kvm is so much easier to get around [17:40] RoAkSoAx, i tried this once, and had some issues. but short answer is "yes, they should" [17:40] and even "yes, they do" [17:40] smoser: ok thanks :) [17:40] ubuntu kernels (-virtual) run on EC2, which is CentOS-ish xen install. === carcinogen75 is now known as zz_carcinogen75 [17:44] i've got a centos 5 xen server that i host several ubuntu VMs on (lucid & maverick) it works very well, but i've only been able to boot linux-ec2 kernels, never could get linux-virtual working [17:44] going to upgrade it to UEC sooner or later ;) [17:45] thx for explaining that smoser [17:45] semiosis, well the -virtual kernels *do* work on xen [17:45] i'm not going to pretend that amazon's xen is "pure xen" [17:46] i've had some issues to, and have not even been able to do an install from CD into an hvm xen instance the one time i tried (centos 5.5) [17:46] which, in theory, should have "just worked" [17:46] yeah hvm didnt work too well for me either [17:58] RoAkSoAx: the linux-ec2 kernels caused much grief and sorrow..they should die in a firey death [17:59] zul: ok :) thanks for the info [18:02] * semiosis takes note of that === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [18:45] is there a Sip/VOIP pbx server that ubuntu supports or will run on ubuntu server? [18:46] shaggy2: asterisk should work [18:46] SpamapS: around? [18:47] shaggy2: just don't ask me about issues there - the code isn't very good, or at least, it wasn't, last time I was using it http://karlsbakk.net/fun/asterisk_architecture.jpg http://karlsbakk.net/fun/asterisk-installation.wav [18:48] zul: yes , OTP, but will be off in 10 minutes [18:50] has anyone ever had a problem with cups? [18:50] I installed it and something happened to it, so I went to remove it and reinstall it and now it hangs when I apt-get remove cups or even apt-get purge cups [18:50] and when I do apt-get install cups it hangs on setting up cups [18:51] NoqturnalX: can you try to strace that? [18:52] what do I strace? [18:52] i've only used strace once and it was a couple years ago so i'm not very familiar with it [18:52] how do i configure PAM to not prompt for a password during the login of a specific user? [18:53] pmatulis_: eeeeerm - why do you want some users to login without a password? [18:54] RoyK: i'd rather not say [18:54] if it's for ssh logins, just use ssh keys [18:54] RoyK: it's not SSH [18:54] if it's for local logins, don't [18:54] RoyK: it's to automate jobs basically === dendrobates is now known as dendro-afk [18:56] i just killed everything that had cups in it lol [18:58] so I want to reinstall cups, is there a way I can make sure it's completely gone before I try it? [18:58] NoqturnalX: purge i guess [18:59] I had to interupt purge [18:59] it locked up [18:59] so I had to kill the tasks [19:00] so I tried apt-get purge cups, then I had to kill the process stop cups and a few others, rm -rf /etc/cups, /usr/lib/cups, /usr/share/cups & did dpkg --configure -a [19:00] what should be my next step? [19:02] should I try apt-get install cups? [19:02] or is there something I should try before that? [19:22] apt-get install cups, made it to "Setting up cups (1.4.4-6ubuntu2.3) and is just sitting there now [19:26] :( === dendro-afk is now known as dendrobates [19:29] did you try to strace it? [19:30] strace -f apt-get install .... [19:30] i'll try that right now [19:33] brb switching computers [19:38] protip : don't use apple ipod hard-disks as server disks [19:39] strace -f apt-get install cups [19:39] right? [19:40] yes [19:40] strace output is sent to stderr [19:40] so 2>somefile [19:40] ok, I did "strace -f apt-get install cups | tee -a strace" [19:40] will send the data there [19:40] 2>&1 | tee [19:41] well it's stuck on "read (0, " [19:41] erm... that's reading from stdin [19:42] *gulp* [19:43] ok so instead I should do "strace -f apt-get install cups 2>&1|tee -a strace" ??? [19:46] dpkg -l cups shows "Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend" [19:46] is there a url of all the man pages as html pages? [19:47] Err?=(none)/Reinst-required (Status,Err: uppercase=bad) [19:48] still get the same thing with 2>&1|tee [19:49] duh a little google goes a long way [19:53] Rawr Can I kill it === dendrobates is now known as dendro-afk [20:00] Hi, does anyone have any good personal experience using Ubuntu as a backup server for Windows clients? [20:13] FoolsRun: I had 4-5 Windows boxes here backing themselves up to an Ubuntu box using a Windows version of rsync. Worked pretty good [20:14] there's supposed to be a /var/run/cups/cups.sock but it doesn't exist [20:15] could this be part of the problem? [20:38] gemclip: manpages.ubuntu.com === dendro-afk is now known as dendrobates [20:47] how do you creat a .sock? [20:53] NoqturnalX: i think the daemon (cups in this case) would usually do that for you [20:53] I think I may have accidentally removed it [20:53] and I can't get the daemon to even run :( [20:55] NoqturnalX: just scrolled back to catch up... did you even get the package installed ok? [20:56] Nope, It always hangs up starting or stopping cups [20:57] so when I install cups it gets to setting up cups and just freezes up on start cups process [20:57] when I try to remove it same thing but with stop cops [20:59] NoqturnalX: anything in /var/log/messages, daemon.log, or syslog about cups? [20:59] NoqturnalX: or /var/log/cups [20:59] /var/log/cups is completely empty folder there are no logs at all [21:00] NoqturnalX: ok what about the system logs? any cups messages? [21:00] daemon.log has init: cups pre-start process (958) terminated with status 1 and init: cups post-start process (1684 killed by TERM signal [21:01] if `sudo lshw -C network` reveals both my NICS are disabled, and thus they don't appear in the list output of ifconfig, what do I do to enable them ? [21:02] /var/log/messages has Apr 8 12:28:18 FLHS-SERVER kernel: [40271.443951] type=1400 audit(1302290898.847:53): apparmor="STATUS" operation="profile_replace" name="/usr/sbin/cupsd" pid=1736 comm="apparmor_parser" [21:02] a bunch of times [21:03] airtonix: network interface manual: http://manpages.ubuntu.com/manpages/maverick/en/man5/interfaces.5.html (also 'man interfaces' in your shell) [21:03] name= either "/usr/sbin/cupsd" or "/usr/lib/cups/backend/cups-pdf" [21:04] theres 25 of each of those messages in /var/log/messages [21:05] could someone give me a hand setting up my resolv.conf I am not understanding what I am doing wrong [21:05] I set my server to a static ip [21:05] semiosis: already tried that, entering desired ifname in /etc/network/interfaces has no effect [21:06] I am running under vmware [21:06] gemclip: have you looked over the resolv.conf man page? it's pretty straightforward, what are you trying to do? [21:07] i see in my gateway my dns ip numbers but im not sure about the search line [21:07] NoqturnalX: sorry i'm not too familiar with cups wish i coudl be more help. wondering if you're able to install other packages, or if this is really just a cups issue. if the cupsd binary is installed, can you launch it by hand? on my system its running as "/usr/sbin/cupsd -C /etc/cups/cupsd.conf" hope that helps you out at least a little === dendrobates is now known as dendro-afk [21:08] do i just enter my gateways ip address? [21:09] gemclip: what are you trying to do? what do you expect to happen that's not happening? [21:09] resolving outside addresses [21:10] and seeing named servers [21:11] following along on the page said once I change my server ip to static I needed to change the resolv.conf file [21:11] gemclip: can you ping other hosts by IP, specifically, can you ping the DNS server IP address you'd like to use as your resolver? [21:11] sec [21:12] airtonix: have you tried 'ifup eth0' or whatever the interface name is you want to use? [21:12] semiosis: yes, no effect [21:12] said "network is unreachable" [21:12] semiosis: I'd provide a pastebin but the computer is obviously not network connected [21:13] everything in ifconfig looks right. I cleared the resolv.conf file [21:16] gemclip: generally speaking, it's not enough for everything to 'look right' in ifconfig, there also has to be an appropriately configured network on the other end of the wire attached to the interface. sounds like you've got connectivity issues. [21:17] gemclip: can you even ping your default gateway address? [21:17] before switching over to static ip's I was able to get a dhcp address ok [21:17] checking [21:18] yep [21:21] gemclip: you may be missing a default route... does the command 'route -ne | grep ^0.0.0.0' return a line beginning with 0.0.0.0 ? [21:21] seems if I try to ping anything outside my network i get "network unreachable" internal; pinging works [21:22] airtonix: what kinda network interface is this you want to get running? ethernet, wireless, something exotic? [21:24] semiosis: it's two gigabit ethernet onboard NICS that worked before with lucid 10.04.2 on a different hard-drive. but switching to another lucid 10.04.2 install on a different hard drive now reveals they are disabled. [21:24] no but i did a route -ne i get Destination 108.78.41.40 gateway 0.0.0.0 genmask 255.255.255.248 flags U MSS 0 Window ) irtt 0 iface eth0 [21:25] semiosis: /etc/NetworkManager/nm-system-settings.conf contains (among other things) : [ifupdown] managed=false [21:26] * NoqturnalX wonders what just happened [21:26] airtonix: are you using ubuntu desktop or server? [21:27] semiosis: desktop functioning as a server (bind, sshd, apache, gitolite, ldap, dovecot, nfs-kernel-server) [21:28] gemclip: you need a default route. i suspect you're missing (at least) a 'gateway' line in /etc/network/interfaces [21:28] ok looking [21:28] airtonix: i dont know about networkmanager, someone else here might be able to help with that, or possible another #ubuntu channel geared toward desktop [21:29] I'm not convinced it an issue with NetworkManager [21:29] it's* [21:29] airtonix: cool, neither am i :) [21:29] its there here is my interfaces: [21:29] iface eth0 inet static [21:29] address 108.78.41.45 [21:29] netmask 255.255.255.248 [21:29] network 108.78.41.0 [21:29] broadcast 108.78.41.7 [21:29] gateway 108.78.41.6 [21:31] my resolv.conf is empty [21:31] gemclip: first things first, you need a default route to reach any system outside your local subnet, then adding the DNS server to resolv.conf is easy [21:32] I recently upgrade my DSL, but now I only have a "sticky" IP address, not a static IP address. What's the best way to automatically update my domain's A record when my IP changes? [21:33] gemclip: you say the 'route -en | grep ^0.0.0.0' returned nothing, that means you dont have a default route, but interfaces file is configured with one... maybe 'service network-interface restart INTERFACE=eth0' will reload the config from file [21:33] semiosis: should I look at the man pages for 'route' sorry im really new at this [21:33] gemclip: i'd rather get it loaded from the config file, which is how it will work at boot... if you use route cmd directly (which is an option) it will not persist across reboot [21:34] gemclip: try that network-interface restart command, replace eth0 with the name of the interface if it's not actually eth0 of course [21:36] yep tried no luck [21:36] let me bring it down and back up [21:37] blah gateway still missing [21:37] gemclip: oh i just noticed something about your interfaces file... your IP address is not in the subnet, but your default gateway is. thats not going to work [21:38] oops [21:38] gemclip: your subnet is ...41.1 - ...41.6, and .6 is the gateway, so you can only use IPs 1-5 [21:38] grr nice catch [21:38] gemclip: ...assuming your network & netmask & broadcast are correct that is [21:39] yeah I set it wrong [21:40] gemclip: well fix the file, do the network-interface restart cmd, and you'll probably be able to ping your DNS server's IP address... verify that, then setting up the resolv.conf is super easy [21:41] gemclip: all you really need is one line, 'nameserver ' [21:43] gemclip: the most popular optional parameter after that is 'domain', then sometimes also 'search' check out the resolv.conf man page for info about those. [21:44] trimeta: Probably to have a script which replaces the date/serial number in format yyyyMMddss in your zonefile with the current moment, does same for the A record, then restarts rndc [21:44] on restart it will push the new record out since the timestamp is newer than the cached versions out there [21:49] trimeta: check out 'dynamic dns' on google or wikipedia [21:50] airtonix: idk what to say, maybe check out the /etc/network/interfaces on the old HDD and compare, if it worked on the old system, then that system has all the answers, hopefully you can still read the disk [21:51] good luck, and have a nice weekend everyone :) [21:54] hello [21:54] I have a question about license of ubuntu server [22:00] I want to use ubuntu server on a server on my company , I should buy it or it's completly free ? === hallyn is now known as hallyn_afk === dendro-afk is now known as dendrobates [22:24] semiosis: got the interface fixed and I can ping the dns server [22:28] woohoo i can ping yahoo! the little things in life that matter lol [22:47] RoAkSoAx: hey [22:47] RoAkSoAx: you still need me to review/sponsor anything for you this week? [22:57] kirkland: now that you are here, sure :) [22:57] kirkland: bug #751979 [22:58] Launchpad bug 751979 in virtinst "virt-install fails to install Ubuntu ISO when it is located in an HTTP location" [Medium,New] https://launchpad.net/bugs/751979 [22:59] kirkland: i proposed another patch first, which was uploaded, but didn't really fix the problem as enabled something and disabled something else. This new patch (which is actually an update to the same patch) hanldes both situations [23:01] kirkland: and my cobbler patches have already been applied upstream :) [23:09] who can help me to run an embedded web server? [23:14] * RoAkSoAx -> EOW [23:24] who can help me to run an embedded web server? [23:24] samira-t: what do you mean? [23:25] amstan: something like this http://www.gnu.org/software/libmicrohttpd/ [23:25] i can't run it