[00:57] <zhul> what is the program used to install "meta packages" ?
[00:57] <sarnold> either apt-get install <foo>^  or tasksel
[00:58] <zhul> huh taskel ty :)
[00:58] <zhul> tasksel*
[01:01] <zhul> sarnold :  I get this error : tasksel: debconf failed to run
[01:02] <sarnold> zhul: is that the entire error? there's not much to work with there
[01:03] <zhul> sarnold: yup it is , i tried aptitude reinstall debconf but it doesn t change a thing
[01:03] <zhul> "aptitude reinstall debconf" *
[01:04] <sarnold> zhul: try apt-get install instead. maybe it gives better error mesages than tasksel
[01:04] <zhul> sarnold: i tried "aptitude reinstall tasksel" and it worked
[01:05] <sarnold> zhul: sorry, I meant, try apt-get install <tasknamehere>^
[01:06] <zhul> sarnold: tasksel is now running ty
[01:06] <zhul> sarnold: it doesn't work as I expected although :)
[01:09] <sarnold> zhul: oh?
[01:11] <zhul> sarnold : I wanted to use it to remove properly all lxc-lxd / virtual stuff but but it wasn't selected in tasksel, so I used apt-get insted
[01:15] <sarnold> zhul: ah. if you want to remove things, indeed apt-get purge is probably a better tool :)
[01:36] <infinital> got a crontab job that is running at startup on 16.10 server, I had scheduled but somehow I'm stuck with my server not booting up.  eventually it just sits on a memory error.  the cron job was:  rm contents of a folder, zip files from other folder, upload to s3,  running as  a bash script, ran fine on CL, but now can't get server to boot.  any thoughts?
[01:36] <infinital> thanks
[01:37] <infinital> like right now it's stuck at the last line of adding files to the zip, which next line was the aws s3 upload.  have left it sitting for over an hour, and the size of the zip file should have uploaded in about 30 minutes
[01:39] <sarnold> I'm surprised a crontab job could halt bootup
[01:39] <sarnold> the usual problem with crontab entries not working right is a difference in environment between a normal login shell and the cron env, 98% of the time it's PATH not being set as expected
[01:40] <infinital> it was weird.  was having path issues with the 's3 aws' command, so was fixing that, then went to crontab -e, to check something, then it just fired off trying to run the script
[01:40] <infinital> sarnold: right, i had the script logging out so i could see why it was running
[01:41] <infinital> but now just trying to figure out how to get this cronjob to stop, booted into a ubuntu 16 desktop live cd, to try and delete the scripts, but the one I moved into the /bin/local (i think) is got an X and I can't access it
[01:44] <infinital> where I'm at currently with it https://usercontent.irccloud-cdn.com/file/eKdhgeAG/file%20server%20issue.png
[01:44] <infinital> the python stuff is from the AWS cli
[01:46] <sarnold> ah.
[01:46] <sarnold> the errors man, the errors.
[01:46] <sarnold> (a) looks like some disk is full. nothing works right when that happens.
[01:46] <sarnold> (b) looks like you've run it way out of memory too. That can happen when swap files fill or ..
[01:47] <JohnMcClain> I always end running out of inodes before I know what's happening.
[01:47] <sarnold> JohnMcClain: oh that's a classic.
[01:47] <JohnMcClain> "Look, I can scrape the web!" "Wait, I still have 90% disk space..."
[01:47] <infinital> well it's on exsi, so I'll take it down and allocate more memory to it, as it's just on 1gb atm
[01:48] <JohnMcClain> Then I end up cronning it all away, then I cry when my DB takes 40s/query
[01:49] <infinital> disk space is 500GB allocated, only using about 100, probably little less
[01:50] <infinital> don't remember size of swap
[01:50] <infinital> i'll let this run, and see what happens.  went from 1gb to 4gb for time being
[01:50] <infinital> bbiaf
[02:06] <zhul> should I use / attach  a virbr0 / bridge device to my lxc container ? or rather my eth0 / physical device ?
[02:13] <sarnold> zhul: I think this is one of the better descriptions of the networking types https://linuxcontainers.org/lxc/manpages/man5/lxc.container.conf.5.html
[02:15] <zhul> ty
[02:21] <zhul> I had made a typo mistake :) lxc improvement are quite impressive lately
[02:23] <zhul> sarnold: do you know if there is any benefit to use virbr0 rather than physical interface to attach to containers ? what is the "best" / common practice ?
[02:24] <sarnold> zhul: I'm sorry to say that I really don't udnerstand the tradeoffs well :/ I think if you pass through a physical interface you won't be ablwe to use it on the host, or any other containers
[02:24] <sarnold> zhul: if I'm right, that might be a deal breaker, or that might be exactly what you wanted :) it's hard to guess..
[02:25] <sarnold> zhul: and I understand that the maclan option may look nice at first, but there's no way for the host to talk to the containers. Or maybe it's no way for any container to talk to any other container or the host. it felt like a stiff penalty.
[02:31] <zhul> sarnold: well I just realized in recent Ubuntu it seems you attach rather to a network than to a network device, so my question was misleading I guess
[02:32] <zhul> sarnold: lxc network attach <network> <container> [device name]
[02:36] <patrick_> Anybody else here using Ubuntu Mate?
[02:39] <sarnold> zhul: ah if you're doing lxd there's some more docs here https://github.com/lxc/lxd/blob/master/doc/configuration.md
[03:41] <protn> hi folks
[03:42] <protn> whats the easiest way to back up and restore list of all installed packages and repositories keys?
[03:47] <sarnold> protn: dpkg --get-selections | dpkg --set-selections -- and /etc/apt/trusted*
[03:49] <protn> sarnold: and where dpkg --get-selections stores text file?
[03:50] <protn> on a new box I have to somehow tell dpkg also where the list of selections is
[03:52] <sarnold> protn: it just dumps it to standard output, and --set-selections reads from standard input.. save it whereveer :)
[03:53] <protn> hmm
[03:53] <protn> maybe simply apt-clone clone foo ?
[03:54] <protn>  dpkg --get-selections | dpkg --set-selections -- and /etc/apt/trusted*
[03:54] <protn> dpkg: error: --set-selections takes no arguments
[03:55] <sarnold> sorry, I shouldn't have used punctaion that's also common on the command line :)
[03:55] <sarnold> you can manage the package list with dpkg --get-selections and dpkg --set-selections
[03:55] <sarnold> and you can manage the repository trust through the /etc/apt/trusted* files
[03:56] <protn> so dpkg --get-selections  and somehow save them to file and then set selections from that file?
[03:56] <sarnold> yeah
[05:09] <hallyn> hm, md127_raid1 has been running for 48 hours, at high cpu
[05:09] <hallyn> doesn't seem like a "reboot" kinda situation, but...
[07:17] <sonu_nk> i am getting service unavailable for my domain..which log i need to check ?
[09:16] <Capprentice> Help Needed: Im trying to set up an Squid Box in bridge mode using Ubuntu 16.04 with two interfaces with TPROXY. Following is the firewall conf -https://paste.ubuntu.com/23526212/ . I can see packet hits in all chains but not in the tproxy redirect chain. Can you explain what Im doing dumb? Why the traffif is not hitting SQUID at all? Any info you need on this, pls ask :) - http://imgur.com/a/dvvz
[09:27] <cpaelzer> Capprentice: sorry I see nothing obvious, the issue seems to exceed my iptables-foo
[09:31] <Capprentice> hmmm ^
[09:31] <Capprentice> cpaelzer:
[09:31] <Capprentice> thanks for looking.
[09:43] <Capprentice> imgur.com/a/dvvzm
[09:45] <sonu_nk> i am getting service unavailable for my domain
[09:45] <sonu_nk> some time some time it works
[11:53] <jonah> hi can anyone help. I'm struggling with NUT / upsd. Although my servers share the same cabinet and UPS they are on different networks and not the same LAN. My NUT slave/client won't connect the master/server whatever I try. I can get them to work together and connect if on the same LAN but not over WAN. Can anyone please help how this can be done?
[14:31] <zul> jamespage: https://code.launchpad.net/~zulcss/charm-helpers/swift-ocata/+merge/311743
[14:44] <jamespage> zul, -> #openstack-charms
[14:44] <zul> sorry
[19:01] <Fieldy> hello, how can I change the server that apt uses from us.archive.ubuntu.com to ca.archive.ubuntu.com ? i haven't been able to reach any of the IPs of the us one for a few days on port 80. mtr shows it falls into a black hole after 91.189.88.162 but i can manually connect to the ca. ones
[19:02] <Fieldy> "after 38.104.186.42" mispaste
[19:04] <Fieldy> ah sources.list that was easy. working now :)