[00:00] wonder how you ended up with that installed [00:00] Dunno [00:00] Know anything about setting up VNC? This server is on my DMZ, i can connect to it from another linux install on the home network [00:01] can't access via WAN even though I have port forwarding on the incoming router [00:01] if you can connect to it from another system on the LAN, then VNC isn't the problem, but networking ;) [00:01] I have to go, hopefully someone else can help. good luck [00:03] thank you tarpman [00:39] hi [00:39] :))) [00:39] I have to say vulr is awesome [00:39] they run whole lot on qemu [00:39] can simply attach new iso and say null windows user passwd in case u forgot it [00:39] and snapshots are free [00:41] is there some web stats [00:41] as to when channel is most active? [00:41] :D [00:42] :))) [00:42] it wouldn't be too hard to build a scraper for https://irclogs.ubuntu.com/2017/06/21/ [00:42] yes [00:42] and least active users are klined for 3 days :D [00:42] would be fun [00:43] sarnold: is your friend here sometimes too? the one who can explain very well [00:44] I was reading some material on sales and terms they use influence how one approach it, I say same applies to coding [00:44] terms used to visualise stuff can influence how one operate with it [00:45] for example if u use this https://sitecdn.adespresso.com/wp-content/uploads/2016/02/sales-funnel-chart.png in a street fight u may get badly beaten [00:45] as it implies some pause, stages [00:45] :) [00:45] hehehe: who needs vulr when you can use lxd. :) [00:45] what is lxd? [00:46] https://linuxcontainers.org/lxd/introduction/ [00:46] yes oki === MAbeeTT_ is now known as MAbeeTT [00:46] greatly simplifies my day-to-day of testing linux distros, etc [00:47] I think subgraph uses them? [00:47] lxd or similar [00:47] vultr is online vps provider, sure local server is cheaper [00:48] sure, not the same [00:48] dpb1: what about co location? [00:48] that is probably cheap [00:49] and them other folks can access machine easier [00:49] then [00:49] yup, I use lxd corrently more for dev/test [00:50] hehehe: no, but some blog posts http://bert-hubert.blogspot.com/ [00:50] also while I am yet to be good at coding - lxd visual also directs approach in a certain way [00:52] since often `complex` tasks are approach using a specific framework, instead of intuitively enacting most efficient solution at that moment [00:52] *approached [00:53] hehehe: you may enjoy http://www.paulgraham.com/avg.html [00:55] quote `When you choose technology, you have to ignore what other people are doing, and consider only what will work the best. [00:55] replace technology with life [00:55] :) [00:57] sarnold: well I can reply to that article - few points - many words are used to make proposed worldview appear as fact - economics - economise , do we have to economise or not? :D [00:58] the whole idea with constant distruptions may not be what it seems @ startups [01:02] Blub is good enough for him, because he thinks in Blub. [01:02] sarnold: yes idea is to think not in C or Python - simply see what is wanted and tool [01:05] what about go? [01:06] many say go is cool [01:06] it's a pretty complex game [01:07] attempting to capture territory and all that [01:07] level 1 is too easy and level 2 is too hard [01:08] the whole thing about competition - money supply is limited by issuers - competition need not to be present, customers demand can simply select who is wanted :D [01:09] sarnold: I think coding is only hard cause its taught in a hard way [01:09] for example one man goes to toilet to pee, he pees and go - another is though that in order to pee you have to bow to sun 20 times, then sing a song [01:09] and then pee :D [01:10] mechanics veiled in mental constructs [01:11] no wonder so many programs smell like someone peed all over them [01:13] what about folks who create new languages? they probably are more fluid with coding? [01:14] yet programming often is collective sport so one dude writing fluid genius something yet can next coder easily to extend this code and modify it/ [01:16] btw who here likes cooking? I plan to dry tomato and store them in oil for entire year :D [01:17] cooking is a programming too - you have to select ingredients and combine them well [01:17] else all you make is spaghetti :) [01:22] hehehe: oh yeah. I like cooking. :) or more to the point I love eating good food... [01:23] well eating is more like apt-get install :D [01:23] albeit I admit u can eat in many creative ways [01:31] any ideas why windows updates take ages to install? :D === mikal_ is now known as mikal === JanC is now known as Guest88964 === JanC_ is now known as JanC === frickler_ is now known as frickler [05:34] hehehe: because it is a very bad OS :)) [06:15] Good morning === andol_ is now known as andol [07:00] almost forgot, but still it is morning - so good morning everybody [07:07] Hey cpaelzer [07:07] How are you doing? [07:16] hi lordievader, busy but good [07:16] lordievader: how are you? [07:16] Doing good here :) [07:17] lordievader: I read yesterday you have 40Gbit up on your university link - I'd with I could get 1/40th of that here :-) [07:18] s/with/wish/ [07:19] Theoretical ;) The university has a 40Gbit up. The link to my server is a 1Gbit, still nice though :) [08:36] is it normal that smtp port 485 needs to be manually opened? [08:39] zetheroo: That all depends on your firewall policy, if it's all open or all closed by default. [08:40] sorry 465 [08:40] smtp ssl port [08:40] well I haven't done any manual configurations of the firewall [08:40] zetheroo: Same answer, even if the questions makes a bit more sense now :) [08:42] zetheroo: Perhaps you could do a pastebin of the output from the command "iptables -L -v" ? [08:42] sure [08:42] Since you haven't done any manual configuration it shouldn't be any sensitive in that output. [08:42] https://pastebin.ubuntu.com/24915684/ [08:43] zetheroo: Everything is wide open there, so nothing which needs to be open. [08:43] ok [08:43] You might need to configure your MTA to explictly listen on port 465. [08:44] it's just that I am getting a 'Could not open SMTP Port' error from a Dokuwiki plugin, and I have port 465 selected as the outgoing mailserver port [08:48] I have configured ssmtp on this server [08:49] Now I'm confused :) Is the plan to have the dokuwiki-plugin use the ssmtp on localhost? [08:50] me too ... usual protocol is to have ssmtp setup on our servers so they can mail out ... but then the admin of this wiki instance wants their wiki to send mail from another account ... so I to am confused now :P [08:52] The dokuwiki smtp plugin has it's own config in dokuwiki ... I didn't think it had anything to do with the servers MTA setup [08:53] apparently this plugin 'will contact a configured SMTP server directly to send emails' [08:55] perhaps I can configure an additional account in ssmtp which matches the dokuwiki plugin account info === BlackDex_ is now known as BlackDex [11:44] hello, SRU for this https://bugs.launchpad.net/ubuntu/+source/cinder/+bug/1692446 is in progress? it is released for trusty-kilo, but no news for trusty [11:44] Launchpad bug 1692446 in cinder (Ubuntu Trusty) "cinder upload-to-image doesn't copy os_type property" [High,In progress] [11:46] nacc: any idea who I can ask to test the pacemaker security updates I have in the security team PPA? [12:05] cpaelzer: got a minute? I have a question about #1657646. Or are you having lunch? [12:05] already back [12:05] giev me 2 min to close out something [12:05] ok [12:05] then I'll read into the bug [12:06] cpaelzer: tl;dr I changed the title, and then read comments #13 and #14 [12:11] ok more thna 2 min :-/ but with you soon ahasenack [12:11] np :) [12:26] ahasenack: now I'm with you [12:26] ahasenack: I already read the comments this morning [12:26] cpaelzer: did you read those comments? [12:26] ok [12:27] so here is my question [12:27] it feels like a bug the fact that I can create thin pools without the package [12:27] but after a reboot they are not there if I do not have the package [12:27] ack [12:27] either refuse to create or work [12:27] right [12:28] do you know the history of that package split? [12:28] or any reason why it's split? [12:28] you seemed to know from memory that you had to install it [12:29] just did more with it in the past, but I don't know why it is split right away [12:30] ahasenack: well I know [12:30] ahasenack: component mismatch [12:30] ahasenack: lvm2 is in main [12:30] thin-provisioning-tools is not [12:30] does it come from the same source? [12:30] no [12:30] I see [12:31] is that also why we can't make it a Recommends? [12:31] yes [12:31] you sometimes find delta that downgrades recommends to suggests because of that [12:31] but [12:31] would this be a Foundations bug? [12:31] in this case even Debian only holds a suggest [12:32] I'm tempted to rephrase the bug, and assign it to lvm2 [12:32] but who would get it then? [12:34] ahasenack: https://launchpad.net/ubuntu/+source/lvm2 see the subscriptions on the right [12:34] foundations [12:34] isn't too important who will get it - is it? [12:35] if it is the correct further triage then do it [12:35] yep, thx [12:35] you can still subscribe(assign us if we want to handle it [12:35] ahasenack: I'm parsing through history of the package if there could be a hint how to improve [12:36] things that come to mind other than changing the dependency are too intrusive [12:36] something to spit out a message when a thin pool is provisioned, essentially, if the tools package isn't installed [12:37] config options sound interesting --with-thin=internal --with-thin-check=/usr/sbin/thin_check [12:37] what would happen without the check defined [12:38] don't know, I haven't used thin pools before coming across this bug [12:43] cpaelzer: I'll rewrite the bug description and change the package, and maybe do some experiments [12:43] cpaelzer: thx for the discussion [12:50] abtraction FTW - I lost track of the thin check on like the 8th level [12:50] anyway it is certainly not "trivial" [12:50] heh [12:51] ahasenack: did you check which part misses the thin check? [12:51] ahasenack: is it a init script or the binary itself? [12:51] and if you found it - if it is by Debian/Ubuntu we own to fix it [12:51] ahasenack: otherwise it might be a valid upstream bug report [12:52] ahasenack: with the focus on "please reject to create if not all req's are around" [12:52] yeah [12:52] I just saw the think_check error when I tried to run "vgchange -a y" manually [12:52] that is what complained about the missing script [12:52] I think it's fair to say some initscript/systemd job calls vgchange -a y at some point [12:53] yes [12:54] ahasenack: the init script does /sbin/lvm vgchange -aay --sysinit [12:54] is that equivalent? [12:54] let me check the second "a" [12:54] would it provike the same error [12:55] it is also the vm binary with vgchange as an arg [12:55] that is ok (lvm vgchange) [12:55] The -aay option should be [12:55] also used during system boot so it's possible to select which volumes to activate using the activation/auto_activation_volume_list settting. [12:56] "a" is auto [14:10] nacc: before I file a bug you might point me diretcly to what I'm missing [14:10] nacc: git merge-changelogs lp1604010/old/debian lp1604010/old/ubuntu lp1604010/new/debian [14:10] nacc: works fiine from git [14:10] git+ssh://git.launchpad.net/usd-importer [14:10] nacc: but fails from the snap complaining there is no merge-changelogs [14:11] first confusion almost got me as there also is "sudo apt install git-merge-changelog" but that again is something different [14:12] I couldn't find it anywhere under the "git ubuntu" scope - maybe you renamed it or did a bigger change to it I missed [14:12] ? [14:13] nacc: uh that could now be "hidden" under git ubuntu merge finish - is it? [14:18] nacc: yeah that was it, the wiki really needs an update in that regard [14:18] nacc: I'll file you a bug listing what I think to be changed instead of messing it up right away [14:34] cpaelzer: ok, lvm2/docker bug reworded and with a simpler test case [14:34] cpaelzer: the moment you create a thin pool LV, vgchange stops working, both to activate or deactivate a VG [14:34] unless you install thin-provisioning-tools [14:35] ahasenack: yeah much better test than reboot [14:35] ahasenack: thanks! [14:35] ahasenack: did you check if it is reasonable to report upstream? [14:35] sounds that way [14:35] not yet, just finished updating the bug [14:35] well they might reflect to distro and say "if you enable make it a depends" [14:36] but still worth to ask [14:36] a safety check on creation would still be nice [14:36] the binary could be broken/unavailable for whatever reasons [14:39] creating an actual thin LV also fails [14:39] many failure cases [14:44] EmilienM: nearly there - there is a bug in neutron @ pike b2 related to not having the port-security extension enabled; this was blocking testing - I've picked the fix and its working it way through the build system atm [14:45] cool [14:46] cpaelzer: should I change the bug status to "new" after having changed the package, so foundations is better aware of it? [14:53] ahasenack: yes I thnik so, but add a comment why you do so (for proper re-tirage from their POV) [14:54] done [14:58] ahasenack: thanks for the updated description - much more readable [14:59] what is this bugproxy thing anyway? :) [14:59] some sort of shared account? [14:59] ahasenack: the "obvious" blunt solution would be to MIR thin-provisioning-tools and then make it a recommends [14:59] yep [14:59] ahasenack: yet since none of us knows the package I can't speak for their code quality [15:00] and that is part of the MIR pre-req [15:00] at least it has no further non-main dependencies [15:00] ahasenack: TL;DR bugproxy is a bot on a companies buzilla that maps their internal bugzilla posts to Launchpad [15:00] ahasenack: and to other projects tracking tools [15:00] ahasenack: a bit like LP track remote bug [15:00] ok [15:01] unfortunately it sometimes gets confsed - posts on wrong bugs, ... [15:15] cpaelzer: the snap should have both subcommands now (e.g. git-ubuntu.merge-changelogs) [15:15] mdeslaur: the openstack folks? I honestly don't know :/ [15:15] nacc: I had version 60 I think and there was no refresh [15:15] available [15:17] cpaelzer: hrm, one sec [15:19] nacc: thanks [15:19] jamespage: any idea who I can ask to test the pacemaker security updates I have in the security team PPA? [15:20] mdeslaur: erm [15:20] mdeslaur: which ubuntu release? [15:20] jamespage: trusty, xenial, yakkety [15:20] jamespage: https://launchpad.net/~ubuntu-security-proposed/+archive/ubuntu/ppa/+packages [15:22] cpaelzer: http://paste.ubuntu.com/24917530/ [15:22] cpaelzer: not git-merge-changelogs, git-ubuntu.merge-changelogs [15:30] nacc: I used what the doc said which is with spaces still [15:30] nacc: but since I got it to work I'm fine [15:30] nacc: I filed bugs for suggested updates to the wiki and man pages to resolve that [15:31] cpaelzer: right, that last bit (merge-changelogs missing) was my realization that you need that command for the wiki :) === danpawlik is now known as danpawlik_ [16:55] hehe [16:55] https://launchpad.net/~james-page/+related-packages [16:55] "Timeout error" [17:19] jamespage: you work too quickly :-P === tomreyn_ is now known as tomreyn [19:29] hm, where are the pgsql-auth bugs I triaged yesterday... [19:30] got them [19:37] oh? netplan only supports systemd-networkd and network-manager? === oerheks_ is now known as oerheks [21:39] Is there anywhere I can still get ub12 .udeb packages? [22:02] !eolupgrade [22:02] End-Of-Life is the time when security updates and support for an Ubuntu release stop, see https://wiki.ubuntu.com/Releases for more information. Looking to upgrade from an EOL release? See https://help.ubuntu.com/community/EOLUpgrades [22:02] maybe that old versions mirror, but i am not even going to look [22:03] ubuntu 12 is EOL === hehehe is now known as hehehe_off [22:09] yeah i know - but udebs should still be downloadable [22:10] i was able to install the scsi module for the netboot kernel on a baremetal ub12 and rebuilt my initrd.gz with the driver i needed, seems to be working [22:50] Is there something more than just 'do-release-upgrade' you're supposed to do to move from 14.04 to 16.04 on a server? I ask because every time, I lose all in-kernel encryption and luks encrypted drives, and have to manually install linux-generic-lts-xenial, bounce again, etc to get it going [22:51] kyle__: why would you install 'linux-generic-lts-xenial' on 16.04? [22:51] kyle__: that's a 14.04 package [22:51] kyle__: also what do you mean by 'in-kernel encryption'? [22:52] kyle__: oh i see, it's there to allow for clean upgrades from the hwe stack, i suppose -- sorry [22:52] The parts of th kernel luks uses for encryption. cat /proc/crypto. That stuff. [22:53] kyle__: but you shouldn't need to manually install it, it should pick it up automatically during hte upgrade [22:53] nacc: Probably. It was that and linux-image-generic that I had to do when I first came across it. [22:53] It's not a show stopper for me, just an annoyance, so I thought maybe I missed some vital step. [22:54] kyle__: i think there's something else going on (possibly a bug). /proc/crypto's presence should not be dependent upon which kernel you have installed (i believe they all have crypto support) [22:55] kyle__: well, linux-generic-lts-xenial and linux-image-generic are rather different packages [22:57] nacc: True. It could be the linux-generic-lts-xenial package was just an artifact from the server I first saw this on, and not neccesary. [22:58] kyle__: yeah -- but in any case, do-release-upgrade *should* be sufficient, presuming you weren't dependent on some 3rd party packages, etc. [22:59] * kyle__ shakes his head [22:59] We keep that to the barest minimum. [23:01] kyle__: seems prudent :) [23:01] kyle__: i guess i'd need to see it in a bit more detail -- that is an upgrade that's in the failed state and see if we can debug it [23:02] I'll pop back in tomorrow with one hopefully. [23:02] It's quittin time, and I nursed the one I was testing on through [23:02] kyle__: sounds good, I should be around [23:03] groovy. G'night! === Epx998- is now known as Epx998