=== lordievader is now known as Guest80543 [02:47] Hwo do I change the username of Ubuntu server? [02:47] To change username (it is probably best to do this without being logged in): [02:47] Hwo do I do that if I am logged in? [02:48] changing usernames is hard [02:48] far easier is to create a new user with the name, group memberships, etc., that you want [02:58] I get this message [02:58] usermod: user mitraj is currently used by process 1861 [03:00] what do I do? [03:00] uxfi: the user will have to not have any processes running to change the name [03:01] ah [03:01] bigjazzsound how do I do that? [03:03] uxfi: you can do something like `ps aux | grep username`. Check the processes that are running under that user. The do `kill processid` to kill them off. [03:03] ok [03:03] uxfi: some of the processes might be inadvisable to kill, which is why sarnold suggested to make a new user instead with the same permissions, groups, etc [03:03] right [03:03] like SSH [03:04] Exactly [03:04] so I should make a new user i guess? [03:04] bigjazzsound well actaully if I can change the name without needing to make a new user is htat possible? [03:05] the thing is it's more than just changing /etc/shadow /etc/passwd and their home directory -- you may need to modify mail spool names, crontab names, maybe atd names, etc [03:05] uxfi: then you run into the issue with processes running under that user, no? [03:06] if you've set up any per-user access controls on databases then those too [03:06] etc [03:06] bigjazzsound which isnt many [03:06] jsut ZNC [03:06] and systemd [03:08] bigjazzsound ill make a new user [03:09] Sorry I could not be of much help here [03:10] its ok bigjazzsound [03:11] i dont want ot lsoe ZNC settings but I can rengegerate them [03:11] you can of course move the data by hand.. [03:12] creating a new user and moving the stuff you need might not be a huge deal [03:12] ^^ [03:13] ah [03:13] I forget where it is stored hmm [03:21] Well, I must admit to being very confused. Migrated a very old conf to a new server, and I have postfix-pcre installed, but I'm still getting "warning: pcre:/etc/postfix/header_checks is unavailable. unsupported dictionary type: pcre" [03:22] keithzg: check output of dpkg -l '*pcre*' and make usre it makes sense [03:23] sarnold: only installed packages look to be libpcre3 and postfix-pcre [03:25] ok nwo i gave the temeproary account sudo access can I delete the origianl one? [03:25] keithzg: then i'm out of ideas.. [03:25] ok now i gave the temeproary account sudo access can I delete the origianl one??? [03:25] uxfi: (a) test the new account first (b) make sure you copied over -everything- from the original that you want to keep. if youv'e done that then go ahead. [03:26] I did sarnold I think [03:26] I tested sudo su [03:26] I tested sudo apt-get update [03:26] seems to work [03:26] I dont have anything in the user folder sarnold the only important thign I ahve is ZNC settings [03:26] sarnold also myw ebsite stuff for some reason its not letting me access /var/www? is there a reason? [03:26] and also will I lose mysql install? [03:29] uxfi: what permissions did you set on /var/www? if you changed owner or set acls you may need to do that again [03:29] uxfi: and probably not lose mysql but if your useraccount had privileges within the mysql database that other users didn't have you may need to redo that too [03:29] I forget what permissions [03:29] sarnold but I made a copy of my /var/www anwyays [03:30] bash: cd: /var/www/: Permission denied [03:30] check with something like find /var/www -ls [03:30] hmmm [03:31] bash: /var/www/: Is a directory [03:31] sarnold [03:31] got any ideas? [03:31] uxfi: namei -l /var/www/htdocs [03:32] ah [03:32] mitraj owns that sarnold the root user [03:32] the account im trying to delete [03:33] sarnold https://i.imgur.com/QoHJ4ln.png [03:33] uxfi: aha, then you'll want to chown /var/www and subdirs to the new user [03:33] ah [03:34] a1berto so log into the old account? [03:34] oops [03:35] sarnold chown what? chown 770 or? [03:35] chown -R newmitraj /var/www [03:36] time ot make dinner :) have fun uxfi [03:36] ah [03:37] what is newmitraj vaishali ? [03:37] sarnold ? [03:39] sarnold: honestly the header checks are the smallest part of my postfix conf so I've just disabled that for now, heh. Frankly a bit astonished that this config, which was set up on a *Trustix* server until today, is even working! :D [03:42] am I doing this right? [03:42] what od I do to take contro of my directory when I make a new account? [03:43] ah nvm [03:43] got it [03:44] Question when I delete a account from Ubuntu (the original admin account) and make a new account will I still get the default ifnormation when I log in (Packages need to be updateD) ? [03:57] Question when I delete a account from Ubuntu (the original admin account) and make a new account will I still get the default ifnormation when I log in (Packages need to be updateD) ? === Guest is now known as uxfi [05:28] here we go [05:28] good morning [05:32] :D [05:32] wooo === JanC_ is now known as JanC === lordievader is now known as Guest54373 [07:46] mwhahaha: I've pushed 4.1.1 to updates, but I think that magnum problem might be related to the newer sqlalchemy [07:46] this is a bit of a problem for projects which are not following the main release cadence for Openstack [07:46] dpkg: feil: dpkg-statusdatabase is locked by another process. / N: Fila «50unattended-upgrades.ucf-dist» in map «/etc/apt/apt.conf.d/» are ignored, because of not guilty file end. / E: Sub-process /usr/bin/dpkg returned an error code (2) [07:46] what I do ? [08:00] blueking: it seems your message is shortened [08:01] it was in norwegian had to translate [08:01] blueking: the message "not guilty file end" reads odd - but in general anything that is not .list in /etc/apt/apt.conf.d is ignored [08:01] ah fine then blueking [08:02] so the message tells you that this file is ignored - the ending "ucf-dist" suggests on which upgrade it was disabled [08:02] yes [08:02] as upgrades usually invalidate old PPA/apt.conf setups they renae them [08:03] The other message about dpdk DB being locked is usually another process installing/updating something at the same time [08:03] dpkg: fail/error: dpkg-statusdatabase is locked by another process [08:03] so if you in a console wrestled with packages, but the auto-updater triggered that might occur (not sure if it has collision detection in unattaneded upgrades) [08:04] I did 14.04 -> 16.04 distro upgrade.. a few apt upgrades later, the last two I have had this dpkg error [08:05] so your real issue is the locking then, just to close the ucd-dist here a reference if you want more on that https://askubuntu.com/questions/829370/n-ignoring-file-50unattended-upgrades-ucf-dist-in-directory-etc-apt-apt-con [08:05] blueking: I think one can check and force to unlock - let me look if I find what I think I remember :-) [08:05] I'll look into that :) TY :) [08:07] blueking: for the locked dpkg it could be an aborted upgrade (aborted harder than ususal to cause this) [08:07] blueking: please follow this https://askubuntu.com/questions/219545/dpkg-error-dpkg-status-database-is-locked-by-another-process [08:07] blueking: and let us know if it helps [08:09] hmm lock file empty ? [08:11] ok I deleted lock file and that part gone [08:11] about dpkg locked by another process [08:13] ok error gone deleted 50unattended-upgrades-ucf-dist [08:14] blueking: did you check if there was a process still holding it? [08:16] yes was none process [08:16] checked with lsof [08:16] ok good [08:16] lock file was empty [08:16] blueking: now does a new update work now? [08:16] yes [08:16] great [08:16] good to have zero errors :) [08:16] "flott" [08:17] norwegian ? [08:17] no trying to go intercultural with google translate, but I'd have got the meaning of "flott" in german as well :-) [08:20] ah I see :) [08:20] 'flott' not youth language :P [09:33] rbasak: if you are on SRUs today could you take a look on the reasoning why trusty isn't through unapproved on bug 1690730? [09:33] bug 1690730 in postgresql-9.6 (Ubuntu) "New upstream microreleases 9.3.17, 9.5.7 and 9.6.3" [High,Triaged] https://launchpad.net/bugs/1690730 [09:35] cpaelzer: probably worth checking with bdmurray (not here right now). [09:36] cpaelzer: FWIW, in the SRU procedure as a deviation from the norm we mark "in unapproved" as In Progress and only Fix Committed once accepted. [09:36] cpaelzer: from https://wiki.ubuntu.com/StableReleaseUpdates#Procedure [09:36] cpaelzer: so perhaps he missed it for that reason? [09:52] Good morning [09:55] rbasak: thanks I'll sned him a mail to avoid forgetting it === tinwood is now known as tinwood_afk === led2 is now known as led1 === tinwood_afk is now known as tinwood [13:41] my apologies for not being at the meeting yesterday [13:42] rbasak: nacc: thoughts? https://wiki.ubuntu.com/NGINX/ReleaseNotes/Artful (this is the 'current working draft' [13:42] and i'm tired heh... *sips coffee*) [14:15] teward: looks good. Thanks! [14:18] I have a systemd question, and it boils down to "how to disable a service from a package maintainer script (postinst)", considering that in debian/ubuntu by default if you install a service it will be started [14:18] and here is the current example: samba's samba.postinst: http://pastebin.ubuntu.com/24644329/ [14:18] the service in question is samba-ad-dc.service [14:19] it does a "ln -s /dev/null /etc/systemd/system/samba-ad-dc.service" [14:19] and via echo asks the user/admin to ignore an error that will show later on [14:19] is that the only way? [14:21] the error you get is just on screen, exit status is zero: http://pastebin.ubuntu.com/24644378/ [14:39] So....anyone here have experiance setting up 802.1x port secrity on a wired network ? [14:41] rbasak: you're welcome. I wanted to write it *somewhere* so I just stuck it two levels underneath the NGINX page. If we need to revise it, let me know, but if not we can use *that* as the release notes for Artful. [14:41] for the nginx subsection. [14:41] the full list of changes was *long* so I only grabbed a few of the pertinent points from the nginx blog post :P [15:17] hey all, I have a network share mounted to an NFS mount point in Ubuntu but when I try to rsync files within that share I get permission denied [15:18] even when I do sudo or use root user on the system [15:18] would that be an indication of how that network share is being mounted [15:19] teward: will review today [15:20] Hi all [15:20] Can an ethernet connection be disconnected due to file download ? [15:45] arunpyasi_: if the router or a proxy sends a connection termination signal, possibly. [15:45] why? [15:46] If the download uses all of the available capacity on any part of the end-to-end link a network device might have to drop it. [15:47] As in it has to get a quart into a pint pot and your Ethernet connection might be part of the spillage. [15:49] jge: You get a permssions result for some odd cases in network shares and exports. Do you have permission to write within the export at all on both ends of the link? [15:56] Hmm OK thanks ! [16:07] jamespage: around? [16:07] nacc: yup [16:08] jamespage: coreycb mentioned syncing up with you to see if you had a chance to test openstack with the django in my PPA for LP: #1605278 (1.11 based) [16:08] Launchpad bug 1605278 in python-django (Ubuntu Artful) "Merge python-django 1:1.11-1 from Debian unstable" [Wishlist,In progress] https://launchpad.net/bugs/1605278 [16:08] nacc: I have not - but lemme check in horizon (which would be the main impact point) [16:08] jamespage: thanks! [16:09] nacc: Django<1.9,>=1.8 # BSD [16:09] hmmm [16:09] jamespage: hrm, prior test was with 1.10.3 and worked (per c#12 in the above bug) [16:12] nacc: I'll take a look and see [16:12] nacc: the whole of openstack pike b1 is backed up in proposed atm which makes things tricker [16:13] jamespage: understood [16:14] jamespage: if you can put it on your probably already busy plate, i'd appreciate it :) [16:15] nacc: gah - artful ain't to happy under lxd today... [16:44] jamespage: it's not? was working decently for me yesterday. *spins a new container* [16:52] forgive me for asking this here since it might be off topic but can anybody recommend a web based open source accounting software for business that runs on ubuntu server or at least where i can find such info except google [16:54] jamespage: seems to work fine for me under an LXD container, what're you testing? [16:54] (under 'pure' LXD command line made by me) === JanC_ is now known as JanC [17:40] forgive me for asking this here since it might be off topic but can anybody recommend a web based open source accounting software for business that runs on ubuntu server or at least where i can find such info except google [17:42] redvic: sorry, have no idea [17:43] thx [17:54] keithzg: _wow_, that's stability :) trustix all the way through to modern. impressive work from postfix team and your management efforts :D === mhcerri_ is now known as mhcerri [19:38] Anyone here with corosync knowledge? I have questions on using postgresql with corosync/pacemaker. [19:40] \? [19:41] paxatron4034: you may get better results with a 'concrete' question [19:42] sarnold: Did you know that concrete was used as far back as ancient Rome? [19:42] FWIW: https://en.wikipedia.org/wiki/Roman_concrete [19:43] mason: crazy. I guess it makes sense that our accomplishments are built onthe knowledge and experience of previous generations but you just sort of assume that olden times were insanely primitive, right? [19:44] @sarnold well. I am running postgresql as a resource in corosync. and there is a monitor process that kicks off every few seconds to see if postgresql is alive (it runs a select now). My immediate conserning is that it is generating alot of logs in auth.log, and I am wondering of this is normal behavior https://pastebin.com/UnR3tBdi [19:45] paxatron4034: eww, i'm surprised they don't re-use a single long-lived process [19:45] paxatron4034: quite often 'large' deployments will shuttle logs to a centralized log server, reducing the impact of logging on disk IO on the individual servers [19:48] \r sarnold okay, but is this normal behavior for postgresql corosync resource? [19:49] paxatron4034: probably [19:54] sarnold: thanks [19:55] paxatron4034: itm ight be worth a report to the project to let them know the logspam is annoying :) maybe they'd be willing to re-do the test.. [19:55] or maybe that's just the way their tests work out. [20:10] Hello everyone, i updated the packages on my Ubunu Server 14.04 yesterday and now i am experiencing some of my cron.d jobs no longer running. Can someone help me diagnose this issue? [20:34] henk717: are you able to determine which packages were updated? [20:36] smoser: ping [20:36] nacc, henk717: Does 14.04 have /var/log/apt/history.log ? [20:36] That might help. [20:37] mason: yeah, i believe it should [20:37] Ah, looks like it should be. [20:37] mason: if not the dpkg log would [20:37] It was a huge list, the server was quite behind on updates [20:38] Another server of mine runs on 16.04 with slightly newer packages and has no issues [20:38] The cron jobs can run if i run them by hand [20:38] They do not get automatically executed [20:38] henk717: do you have an example cron entry that fails? [20:39] henk717: also, do you get an e-mail about any errors [20:39] Yes, i do it seems to be one entire file. No errors are present in the logs as far as i have seen. [20:39] henk717: well, it often won't be logged, but if there was an error, I believe root (or the configured user for cron mails) will receive one locally [20:39] henk717: but it soudns like, it's not failing, it's just not even running? [20:40] henk717: which file? [20:41] It looks like it is not even running indeed, cron runs though and some other cron jobs did ran [20:41] File is /etc/cron.d/overwatch which contents i am currently posting on pastebin for you to see [20:43] https://pastebin.com/GUE1704S [20:43] um, i don't think you can have users in the crontab like that [20:43] It always worked [20:44] The same file runs perfectly on a 16.04 machine [20:44] only /etc/crontab (I believe) supports that format (per `man 5 crontab` [20:45] It would be quite specific if the lower version handles this fine and the higher version handles it fine [20:45] nacc, here. [20:45] nacc: the system /etc/crontab can have user definitions [20:45] user level crontabs can't [20:45] and user level crontabs are *usually* how you do things [20:46] smoser: hey, wanted to bounce something off you about the git tooling, if you have a minute for a HO? [20:46] teward: right, what i said -- but is /etc/cron.d/* a 'system' crontab? [20:46] In my experience it has been [20:46] sure [20:46] nacc: yes. [20:47] henk717: ah ok, it is considered the same, sorry [20:47] nacc: No problem, i apprechiate you thinking along [20:47] Fact remains its a proven cron setup prior to the update which runs on a newer cron version as well [20:47] nacc: http://paste.ubuntu.com/24647476/ <- example, from mdadm's automatic crontab it installs [20:47] teward: yep, i'm seeing that now [20:50] What makes the issue more difficult is that there is no error anywhere to be seen, and i can run any of the scripts manually without any failure [21:21] henk717: i'm not sure how to debug it. I would start with adding a new crontab at the same level, with a test cronjob and see if it triggers. Then start adding lines from the non-working to the test one and see when/if it breaks [21:25] nacc: Currently planning to upgrade to 16.04 if there isn't to much risk involved [21:26] smoser: re: LP: #1569925, I think one of the two iscsi services is supposed to kill all sessions on shutdown [21:26] Launchpad bug 1569925 in systemd (Ubuntu) "Shutdown hang on 16.04 with iscsi targets" [High,Confirmed] https://launchpad.net/bugs/1569925 [21:26] smoser: so if that's not happening, probably that's what needs debugging [22:04] sarnold: Yeah, Trustix to Ubuntu 16.04; I'm even more impressed now with the Postfix folks than I already was! [22:04] :D [22:38] Got good news and i got bad news, the upgrade to 16.04 was succesfull. Bad news is cron is still doing the same thing. [22:39] henk717: seems highlgy likely , if there is an identically configured server somewhere else at the same rev, that something else is going on :) [22:40] Got more good news [22:40] 16.04's cron dumps errors [22:40] The newer versions didn't like the fact the reboot line had no user specified [22:40] Its all fixed now :D [22:40] henk717: probably that was required on 14.04 too [22:41] ? [22:41] On the new cron, yes totally [22:41] But it threw no error [22:41] On the old cron the line likely just got ignored [22:41] right, i'm saying that there is probably an improvement to be made to the error handling, but that might be why they weren't running [22:41] the crontab failed to parse [22:49] The improvement is already made apparently, its just not in the 14.04 version since its out of support [22:53] henk717: hrm? 14.04 is still supported [22:53] henk717: perhaps file a bug, or see if one already exist [23:45] Hmm, trying to set up a UEFI-capable PXE boot largely via https://wiki.ubuntu.com/UEFI/PXE-netboot-install, but on the (very new and high end!) PC I'm trying it on I'm getting the Intel Boot Agent throwing a fit, complaining "PXE-E79: NBP is too big to fit in free base memory" before it can even load the Grub menu. [23:46] Any easy way to chainload into Grub for that?