| arooni-mobile | hey folks; im on an ubuntu 14.04 laptop and id like a bitter vnc viewer than vinagre. when i connect to my mac mini that has two 1920x1080 displays i am either too far zoomed in or out. i'd like to view only one display at a time via vnc. how can i do this? | 02:30 |
|---|---|---|
| === athairus is now known as afkthairus | ||
| === _degorenko|afk is now known as degorenko | ||
| devster31 | on a live server I should only install new updates if they're security upgrades? | 10:34 |
| coreycb | jamespage, beisner: hello, 1:2.2+dfsg-5expubuntu9.7~cloud5 is ready for promotion to kilo-proposed when you have a moment | 11:46 |
| jamespage | coreycb, ok got it | 13:13 |
| coreycb | jamespage, thanks. AMZ_ ^ | 13:14 |
| coreycb | AMZ_, it'll need some time to build and publish binaries then qemu will be available to test in kilo-proposed | 13:15 |
| AMZ_ | jamespage: coreycb: thanks! i'll keep an eye on it | 13:15 |
| jamespage | done | 13:16 |
| coreycb | jamespage, we're going to leave paramiko at >= 1.16.0 for b1. there were missing directories in the released tarball that d/watch points at. the maintainer said he'll be updating to 2.0.0 soon so we can pick it up in b2. | 13:24 |
| coreycb | ddellav and I were chatting with him | 13:25 |
| jamespage | coreycb, ok | 13:26 |
| ddellav | coreycb jamespage though his latest email says he's almost done with updating it | 13:26 |
| coreycb | ddellav, hmm. ok. | 13:27 |
| coreycb | ddellav, alright well nova and cinder are just about ready to upload so I think I'll just upload with >= 1.16.0 and we can bump d/control afterward for b2. | 13:29 |
| ddellav | coreycb: ok sounds like a plan | 13:30 |
| coreycb | ddellav, looks like monascaclient needs an MIR (for heat) | 13:35 |
| ddellav | coreycb: ok, I'll add that to my todo list | 13:36 |
| coreycb | ddellav, ok thanks | 13:36 |
| coreycb | ddellav, this can be a good reference to understand why something is stuck in proposed: http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html | 13:36 |
| === Bruh is now known as ily | ||
| coreycb | ddellav, jamespage: nova and cinder uploaded | 14:03 |
| nixnothing | morning | 14:42 |
| === ShaRose_ is now known as ShaRose | ||
| jamespage | ddellav, coreycb: dealing with python-babel being stuck in proposed - openerp6.1 need a fixup | 14:48 |
| coreycb | jamespage, ok need a hand with that? | 14:51 |
| jamespage | coreycb, nah is just using old pkg names | 14:51 |
| coreycb | jamespage, ok | 14:52 |
| jamespage | coreycb, btw I did a bit of a re-jib on the ubuntu theme in the dashboaard | 14:52 |
| jamespage | if i've got it right, its always installed, and is end-user selectable | 14:53 |
| jamespage | its the default all the time unless someone changes their mind of reconfigures things | 14:53 |
| coreycb | jamespage, awesome I'll take a look. selectable in local_settings.py? | 14:56 |
| jamespage | coreycb, yes | 14:58 |
| jamespage | DEFAULT_THEME | 14:58 |
| coreycb | jamespage, ok. cool, looks like the changes also limit the static assets compression to once per install too. | 14:59 |
| jamespage | coreycb, yes - thus avoiding that trigger thing I was talking about | 14:59 |
| coreycb | jamespage, great | 15:00 |
| coreycb | jamespage, I'd like your opinion on bug 1574144 when you have a sec | 15:00 |
| ubottu | bug 1574144 in horizon (Ubuntu) "package openstack-dashboard-ubuntu-theme 2:9.0.0-0ubuntu2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Medium,Triaged] https://launchpad.net/bugs/1574144 | 15:00 |
| jamespage | coreycb, hmm I wonder | 15:02 |
| jamespage | coreycb, might be related to https://bugs.launchpad.net/ubuntu/+source/horizon/+bug/1584040 | 15:02 |
| ubottu | Launchpad bug 1584040 in horizon (Ubuntu Yakkety) "Horizon wsgi can run in the wrong WSGIProcessGroup" [Undecided,New] | 15:02 |
| coreycb | jamespage, hmm not sure. basically it's a result of installing openstack-dashboard when apache2 is already installed but apache2 is not running at install time. | 15:04 |
| coreycb | jamespage, then openstack-dashboard.postinst fails here if apache2 is down: apache2_invoke enconf $CONF || exit $? | 15:05 |
| jamespage | coreycb, hmm | 15:06 |
| nacc | sdeziel: would you be able to help me verify if LP: #1457957 is the same underlying issue as the puppet bug you're working? | 16:24 |
| ubottu | Launchpad bug 1457957 in puppet (Ubuntu Vivid) "puppet uses upstart for service status checks in vivid" [High,Triaged] https://launchpad.net/bugs/1457957 | 16:25 |
| nacc | sdeziel: on cursory reading so far, it seems like it must be :) | 16:25 |
| sdeziel | nacc: yes, looks like the same thing to me. Let me try to reproduce in a Vivid container | 16:27 |
| nacc | sdeziel: thanks! | 16:27 |
| nacc | sdeziel: it might need a trusty -> vivid upgrade (so you had upstart at some point), but not sure | 16:28 |
| rbasak | magicalChicken: please can I have a progress update from you on bug 869017, bug 1394403 and 1511222? No progress is an acceptable report. | 16:45 |
| ubottu | bug 869017 in kbd (Ubuntu) "Ubuntu server enables screenblanking, concealing crashdumps (DPMS is not used)" [Medium,Triaged] https://launchpad.net/bugs/869017 | 16:45 |
| ubottu | bug 1511222 in apache2 (Ubuntu Trusty) "Incorrect trusted proxy match test in mod_remoteip" [Medium,Triaged] https://launchpad.net/bugs/1511222 | 16:45 |
| ubottu | bug 1394403 in apache2 (Ubuntu Trusty) "RewriteRule of "^$" is broken" [Medium,Incomplete] https://launchpad.net/bugs/1394403 | 16:45 |
| coreycb | jamespage, ddellav: new oslo.messaging uploaded. I failed to merge the ubuntu delta on my last upload. should be fixed now. | 16:49 |
| === afkthairus is now known as athairus | ||
| magicalChicken | rbasak: Hey, I have not made any progress on the bugs yet, sorry about that. I got sidetracked with curtin work. I see the patch for 1511222 and I can get that patched into the apache package this afternoon | 16:51 |
| magicalChicken | As for 1394403 I was not able to reproduce the bug after the original patch that was uploaded was applied using the method provided in the comments. I'll try again with the method from the original report though | 16:51 |
| magicalChicken | For the kbd bug I am kind of at a loss as to how to deal with that, aside from setting consoleblank=0 in the default grub configuration, but modifying boot seems a bit overkill for a minor bug | 16:52 |
| sdeziel | nacc: so the Vivid issue really looks the same as the Xenial/Yakkety one | 16:54 |
| nacc | sdeziel: great, thanks! can you put a comment in and mark it as a dupe? I trust your judgment in this regard. And we're not going to be fixing vivid anyways, so it makes sense to indicate it is fixed in xenial, at least | 16:55 |
| nacc | (once it is) | 16:56 |
| sdeziel | nacc: but the problem is not fixed in Xenial yet | 16:56 |
| nacc | sdeziel: yes, hence the parenthetical caveat | 16:57 |
| nacc | sdeziel: the bug is the same, we'll fix it in one place (bug) | 16:57 |
| sdeziel | nacc: ah, sorry | 16:58 |
| nacc | sdeziel: np, i was being overly verbose -- i agree with your assessment, if you could put a comment in and indicate it's a duplicate of the other bug, that'd be great (or i can later today) | 16:58 |
| sdeziel | in https://bugs.launchpad.net/ubuntu/+source/puppet/+bug/1495853, paelzer mentioned: "There is a known issue in case upstart is installed - not 100% a dup, that is tracked in bug 1457957" | 16:59 |
| ubottu | Launchpad bug 1495853 in puppet (Ubuntu) "puppet service enable broken on ubuntu vivid with debian provider" [Medium,Fix released] | 16:59 |
| ubottu | bug 1457957 in puppet (Ubuntu Vivid) "puppet uses upstart for service status checks in vivid" [High,Triaged] https://launchpad.net/bugs/1457957 | 16:59 |
| Odd_Bloke | jackdpeterson: o/ | 17:00 |
| sdeziel | I'll dig deeper as apparently Debian now always use the "service" wrapper and it seems to work for them | 17:00 |
| Odd_Bloke | jackdpeterson: What version of Vagrant are you using? | 17:00 |
| jackdpeterson | Latest (1.8.1, virtual box also at latest) | 17:00 |
| Odd_Bloke | jackdpeterson: On Ubuntu, or...? | 17:01 |
| jackdpeterson | Actually, on Mac OS-X ... but my colleagues could reproduce this issue across host OS's | 17:01 |
| nacc | sdeziel: ah ok, thanks | 17:02 |
| Odd_Bloke | jackdpeterson: I've updated that bug with what I'm seeing. | 17:03 |
| jackdpeterson | Glad you could reproduce the issue :-) | 17:04 |
| Odd_Bloke | jackdpeterson: Umm, I don't think I am reproducing the issue there; the box works fine for me. | 17:04 |
| jackdpeterson | hmm, let me give it a shot on one of my ubuntu boxes to see if I can reproduce here in my strange alternate reality | 17:04 |
| Odd_Bloke | jackdpeterson: Sure; I'm heading out now (I'm UK-based), but if you update the bug and ping me again tomorrow then we can keep working through it. :) | 17:05 |
| jackdpeterson | Sure thing | 17:05 |
| Odd_Bloke | jackdpeterson: (One thought: I have a pretty hefty machine so my boot might be happening before some timeout causes a failure to reported?) | 17:06 |
| Odd_Bloke | Anyway, now I'm gone. | 17:06 |
| jackdpeterson | Ah, well the other part of this is that it *should* be using the vagrant user and not ubuntu as the user to be consistent with that target environment. | 17:07 |
| === InfoTest1 is now known as InfoTest | ||
| === iberezovskiy is now known as iberezovskiy_afk | ||
| voot | Anyone have experience with FreeRADIUS? I'm writing some unlang, and my %{Group} variable expansion is returning an empty list, even when "id <username>" returns a full list of group membership | 17:27 |
| === ily is now known as lrn2whois | ||
| genii | voot: Now to wait until someone knowledgable on the subject responds :) | 17:31 |
| * genii makes more coffee | 17:31 | |
| etropes | hi there | 17:43 |
| jelly | or ask in #freeradius | 17:43 |
| === alexisb is now known as alexisb-afk | ||
| cpaelzer | sdeziel: nacc: I just thought them to be related at least - that is why I just put it in as a note | 17:47 |
| cpaelzer | nacc: sdeziel: I lack some of your contect - did I accidentially misclassify something there? | 17:48 |
| nacc | cpaelzer: no, i think you're good | 17:49 |
| cpaelzer | nacc: ok, thanks | 17:49 |
| coreycb | ddellav, need any uploads sponsored? | 18:04 |
| ddellav | coreycb did you do the neutron-* packages? i dont remember seeing confirmation on those | 18:04 |
| coreycb | ddellav, nope, I saw your name by them | 18:05 |
| coreycb | ddellav, I sponsored an upload for neutron-fwaas | 18:05 |
| ddellav | coreycb ok lbaas and vpnaas are done too, let me post the links | 18:05 |
| coreycb | ddellav, ok working on lbaas, let me know when vpnaas is ready | 18:15 |
| ddellav | coreycb ok, pushing it now but launchpad seems to be really slow | 18:17 |
| ddellav | either that or my connection is bad | 18:17 |
| coreycb | ddellav, it might be, there was an outage today | 18:17 |
| ddellav | yea, i know IRC was down for awhile | 18:18 |
| ddellav | didnt know that spread to launchpad | 18:18 |
| coreycb | ddellav, lp seems ok for me now | 18:18 |
| ddellav | coreycb lp:~ddellav/ubuntu/+source/neutron-vpnaas ready for review. I couldn't build due to mock | 18:18 |
| coreycb | ddellav, we decided not to bump mock for b1 | 18:19 |
| ddellav | coreycb oh i thought that was only paramiko | 18:19 |
| ddellav | i'll drop it back down and rebuild | 18:19 |
| coreycb | ddellav, k thanks | 18:19 |
| coreycb | ddellav, neutron-lbaas pushed/uploaded | 18:20 |
| ddellav | coreycb rebuilding vpnaas and manila after dropping mock back down to 1.2 | 18:22 |
| coreycb | ddellav, sounds good | 18:22 |
| ddellav | coreycb oh i gotta do designate too, i'll get that cooking as well | 18:22 |
| coreycb | ddellav, ok | 18:22 |
| coreycb | ddellav, next time around for the *aas we should bump python-neutron to the current yakkety level | 18:32 |
| ddellav | coreycb ok noted | 18:33 |
| ddellav | coreycb lp:~ddellav/ubuntu/+source/manila lp:~ddellav/ubuntu/+source/designate lp:~ddellav/ubuntu/+source/neutron-vpnaas ready for review | 19:01 |
| coreycb | ddellav, taking a look, they build ok? | 19:06 |
| ddellav | coreycb yep, all green | 19:06 |
| coreycb | ddellav, greaT | 19:06 |
| coreycb | ddellav, ok all are pushed/uploaded | 19:21 |
| ddellav | coreycb ok great, thanks | 19:22 |
| coreycb | ddellav, yep np. now on to ensuring nothing is stuck in proposed and we'll get everything backported to the Newton UCA, then test. | 19:23 |
| EmilienM | coreycb: did you update neutron packaging in Mitaka/Xenial? | 20:34 |
| EmilienM | tempest fails to run now :( | 20:34 |
| EmilienM | http://logs.openstack.org/98/326698/1/check/gate-puppet-openstack-integration-3-scenario001-tempest-ubuntu-xenial/bc5b5c2/console.html#_2016-06-07_19_37_25_274 | 20:34 |
| coreycb | EmilienM, we did a stable release for neutron but none for neutron-*aas | 20:38 |
| coreycb | EmilienM, is neutron-fwaas installed? | 20:38 |
| EmilienM | yes | 20:38 |
| EmilienM | coreycb: do you run tempest in your CI? | 20:39 |
| EmilienM | you might have seen the same thing maybe | 20:39 |
| coreycb | EmilienM, we run tempest smoke tests before we release | 20:39 |
| EmilienM | coreycb: do you deploy fwaas ? | 20:40 |
| coreycb | EmilienM, mind giving me a sec while I look into this? | 20:41 |
| EmilienM | coreycb: sure | 20:42 |
| Belldandu | Guys There have been NUMEROUS bug reports for this same issue https://bugs.launchpad.net/ubuntu/+source/postfixadmin/+bug/1321955 that have been confirmed OVER and OVER again for the last 2 years about postfixadmin removing mariadb even though its a dropin replacement for mysql | 20:47 |
| ubottu | Launchpad bug 1321955 in postfixadmin (Ubuntu) "postfixadmin dependencies not well defined (requires mysql-server or postgresql, removes MariaDB)" [Undecided,Confirmed] | 20:47 |
| Belldandu | Its getting rediculous at this point] | 20:47 |
| Belldandu | the fix is as simple as adding mariadb as another alternate dependency aside from mysql and postgressql | 20:48 |
| Belldandu | and yet | 20:48 |
| Belldandu | IT HASNT been done | 20:48 |
| Belldandu | whats worse is | 20:49 |
| Belldandu | it hasnt been done through SEVERAL ubuntu releases | 20:49 |
| tgm4883 | Belldandu: please don't use enter as punctuation | 20:49 |
| Belldandu | how has no one caught and fixed this by now with so many confirmed bug reports? | 20:50 |
| Belldandu | I actually murdered my own mariadb installation a few minutes ago because i completely forgot about this crap. | 20:52 |
| Belldandu | I fixed it but its annoying because i now have to install postfixadmin from source (which was the main reason i didnt do it before) | 20:52 |
| rattking | I suppose if you are building from source you could add that dep to the package and build it from the src deb | 20:54 |
| tgm4883 | Belldandu: and in the process, you could submit a patch | 20:54 |
| === ily is now known as Bruh | ||
| Belldandu | im about to do that because im tired of seeing it like that | 20:56 |
| tgm4883 | Belldandu: good, I look forward to seeing that patch | 20:57 |
| coreycb | EmilienM, I'm not sure what the problem is. did this ever work on mitaka? | 21:04 |
| EmilienM | coreycb: yes, like 1 hour ago it worked | 21:04 |
| coreycb | EmilienM, hmm, ok. for the neutron packages, we've just released the stable version of neutron 8.1.0. | 21:04 |
| coreycb | EmilienM, xenial or trusty? | 21:04 |
| EmilienM | coreycb: xenial | 21:05 |
| EmilienM | I haven't tested trusty but let me kick off a CI test | 21:05 |
| EmilienM | let's see if Ubuntu Trusty jobs fail too now https://review.openstack.org/326727 | 21:06 |
| coreycb | EmilienM, neutron 8.1.0 was released to xenial-updates on 5/30/2016 so it can't be that then if it started an hour ago | 21:07 |
| EmilienM | we pin Tempest since Friday | 21:08 |
| EmilienM | maybe something between, idk | 21:08 |
| coreycb | EmilienM, btw it may not be related but mixing pip and apt packages is not a recipe for success, unless the pip packages are isolated in a virtualenv | 21:12 |
| bulldawgden | Is Zoneminder the best solution for security camera recording? | 21:14 |
| bulldawgden | For Linux of course | 21:15 |
| K4k | What's the Ubuntu equivalent of `chkconfig --list`? | 21:26 |
| K4k | I know adding and removing services is update-rc.d but it doesn't look like it can just list the services and their run levels? | 21:27 |
| sarnold | K4k: service --status-all is a good start if you're on an ubuntu with upstart | 21:29 |
| sarnold | K4k: if you want to see what's been disabled, probably "grep manual /etc/init/*" is a decent start point. | 21:30 |
| K4k | `ls /etc/rc3.d` got me enough to go on. Unfortunately, `service --status-all` just shows if the services are running or not currently. | 21:30 |
| K4k | `chkconfig --list` shows you each service and if it's on or off for each run level | 21:31 |
| EmilienM | coreycb: our stable/mitaka CI that pin tempest on latest stable tag is not failing so I suspect something merged in Tempest recently and the upgrade of Neutron broke something | 22:03 |
| Belldandu | well | 22:37 |
| Belldandu | i made the patch | 22:37 |
| Belldandu | but | 22:37 |
| Belldandu | every time i goto build it complains about fuzz | 22:37 |
| Belldandu | even tried copying the patch out deleting the entire dir and redownloading it and then copying the patch back in and adding it to series | 22:38 |
| Belldandu | still complains | 22:38 |
| sarnold | Belldandu: are you trying to patch the contents of the debian/ directory in a quilt patch? | 22:40 |
| Belldandu | yeah | 22:40 |
| nacc | that's not what quilt patches are for | 22:41 |
| nacc | quilt patches are for source modifications | 22:41 |
| sarnold | aha :) that won't work -- just modify the debian/ directory's contents directly | 22:41 |
| Belldandu | fun | 22:42 |
| nacc | Belldandu: and given that trusty was in sync with debian, isn't it really a debian bug? has it been fixed upstream (debian?) or the most recent versions in yakkety? | 22:43 |
| Belldandu | let me see what debian says | 22:44 |
| Belldandu | cause so far i've only run into this issue on ubuntu | 22:46 |
| nacc | Belldandu: what removes mariadb? it cna't be doing it silently, so users are saying yes and not paying attention? | 22:47 |
| Belldandu | yeah | 22:47 |
| Belldandu | postfixadmin does | 22:47 |
| Belldandu | it says that either mysql or postgressql is required | 22:48 |
| Belldandu | if mariadb is installed it installs mysql | 22:48 |
| Belldandu | which you cant go back to mysql from mariadb | 22:48 |
| nacc | Belldandu: and why is it removing mariadb? I don't see a conflicts for mysql-client or postgresql-client | 22:48 |
| Belldandu | because mariadb-client/mysql-server must be listed as an alternate dependency for mysql-client/mysql-server | 22:49 |
| nacc | Belldandu: -server is a recommends, not a depends | 22:50 |
| Belldandu | mariadb-server/mysql-server | 22:50 |
| Belldandu | yeah ik | 22:50 |
| Belldandu | sec | 22:50 |
| Belldandu | i typed it out wrong | 22:50 |
| Belldandu | because mariadb-client must be listed as an alternate dependency for mysql-client | 22:50 |
| Belldandu | there | 22:50 |
| nacc | but there's no 'postfix-mariadb' | 22:51 |
| nacc | so what would the alternative be for that? | 22:51 |
| nacc | is there a courier-authlib-mariadb? | 22:51 |
| Belldandu | thats what postfix-mysql is for. Mariadb is a dropin replacement for mysql | 22:51 |
| Belldandu | everything mysql is integrated in mariadb | 22:51 |
| Belldandu | mariadb just has more features | 22:51 |
| nacc | it would seem like mariadb should be doing some sort of provides: for mysql if that were true? | 22:52 |
| Belldandu | so there is no need for a postfix-mariadb unless you plan to use the extra features in postfix | 22:52 |
| nacc | i think i recall rbasak telling me something about this | 22:53 |
| Belldandu | as far as i know mariadb does a replaces on mysql when its installed | 22:53 |
| nacc | ah it provides | 22:53 |
| nacc | virtual-mysql-client | 22:53 |
| tarpman | possibly postfixadmin wants its Depends: mysql-client changed to virtual-mysql-client? that would be a debian bug too, and I don't see an existing one | 22:53 |
| nacc | so i think that's your fix | 22:53 |
| nacc | tarpman: jinx! | 22:54 |
| tarpman | too slow >_< | 22:54 |
| Belldandu | xD | 22:54 |
| nacc | Belldandu: simple, straightforward fix, i think ... | 22:54 |
| Belldandu | yeah | 22:54 |
| nacc | Belldandu: but yeah, should go to Debian and get sync'd down | 22:54 |
| nacc | Belldandu: note there is also virtual-mysql-server | 22:55 |
| nacc | Belldandu: also present in trusty (both virtual packages) | 22:56 |
| Belldandu | hmm | 22:56 |
| nacc | Belldandu: but we'd need to get it fixed in yakkety first, then backported per SRU policy | 22:56 |
| nacc | presuming this qualifies (which it seems like it might) | 22:56 |
| Belldandu | also | 22:57 |
| Belldandu | virtual-mysql-client | 22:57 |
| Belldandu | oh derp | 22:58 |
| Belldandu | i read that late | 22:58 |
| Belldandu | xD | 22:58 |
| tarpman | oh, there is a bug. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=778794 | 22:58 |
| ubottu | Debian bug 778794 in postfixadmin "postfixadmin: Please change the dependencies/recommendations" [Wishlist,Open] | 22:58 |
| Belldandu | yeah im not sure why i havent seen that till now | 22:58 |
| tarpman | the postgres dep looks wrong too, there doesn't seem to be a package called postgresql-server, rather postgres... | 22:59 |
| Belldandu | yeah | 22:59 |
| tarpman | *postgresql | 22:59 |
| Belldandu | i remember there being a bug report about that too | 22:59 |
| Belldandu | something about the dependency being old | 22:59 |
| Belldandu | for postgres | 22:59 |
| Belldandu | supposedly there is an upstream package that contains the fix already in debian | 23:01 |
| tarpman | ah yeah. fixed in 2.3.7-1, which is in wily | 23:01 |
| tarpman | so that would be an SRU candidate already | 23:01 |
| Belldandu | virtual-mysql-client/server is in yakkety all the way back to precise | 23:03 |
| Belldandu | although in precise mariadb doesnt exist | 23:04 |
| lamont | nacc: postfix-mysql Depends: libmysqldb${number}, it doesn't care about clients | 23:09 |
| Belldandu | yeah | 23:09 |
| lamont | so unless the library is abi-compatible, you'll want to request a postfix-mariadb. patches welcome, of course. | 23:09 |
| lamont | if there's interest, please do file the bug though, with or without the patch | 23:11 |
| lamont | and no, I don't care if you file it against debian or ubuntu | 23:11 |
| Belldandu | MariaDB is a fork of the MySQL database project that provides a drop-in replacement for MySQL. It preserves API/ABI compatibility with MySQL and adds some new features. | 23:12 |
| Belldandu | so yes its abi compatible | 23:12 |
| lamont | abi compatible to the extent of having the same library filename? | 23:13 |
| lamont | api-compatible I believe. | 23:13 |
| lamont | abi is work | 23:13 |
| nacc | lamont: ack, i was just looking at the packaging without too much knowledge ad figured we'd want the alternatives to be correct for all cases | 23:13 |
| nacc | *and | 23:13 |
| lamont | hrm... | 23:13 |
| Belldandu | mariadb was forked made by some of the original mysql developers | 23:14 |
| lamont | actually... if you grab the latest postfix, and install postfix-mysql and then smash /etc/postfix/dynamicmaps.cf to point to the mariadb library, does it even work? | 23:14 |
| lamont | forked usually means abi differences, even if api remains the same | 23:14 |
| Belldandu | sure there are some differences however they maintained compatiblity for both api and abi with mysql | 23:15 |
| lamont | Belldandu: I'd recommend going the postfix-mariadb route | 23:15 |
| lamont | and, knowing the maintainer like I do, I'm sure he'd agree. :D | 23:15 |
| Belldandu | there is no postfix-mariadb though | 23:16 |
| lamont | that would be why I said to file a bug if you want it to exist. | 23:16 |
| Belldandu | even how to forge gives "installation guides" saying to use postfix-mysql with mariadb | 23:16 |
| sarnold | lamont :) | 23:16 |
| Belldandu | Now we can install Postfix, Dovecot, MariaDB (as MySQL replacement), rkhunter, and binutils with a single command: | 23:17 |
| lamont | the abi that postfix is compiled against for xenial says that the name of the library shall be "libmysqlclient.so.20". IOW, no, it's not fully ABI compatible at the library level. | 23:17 |
| Belldandu | apt-get install postfix postfix-mysql postfix-doc mariadb-client mariadb-server openssl getmail4 rkhunter binutils dovecot-imapd dovecot-pop3d dovecot-mysql dovecot-sieve sudo | 23:17 |
| Belldandu | postfix works completely fine with mariadb | 23:18 |
| Belldandu | its only postfixadmin thats causing issues | 23:18 |
| Belldandu | and postfixadmin is strictly php | 23:18 |
| lamont | that would be a second reason I've never looked at postfixadmin | 23:18 |
| sarnold | yikes, rkhunter? didn't that have a nice user->root exploit in it? | 23:18 |
| sarnold | or was that the one that thought upstart was a rootkit? | 23:19 |
| sarnold | maybe it's both. heh. | 23:19 |
| lamont | sarnold: I've found that "php" and "root vul" tend to come in the same package, 99% of the time. | 23:19 |
| sarnold | lamont: pretty good indicator isn't it? :) | 23:19 |
| Belldandu | so as far as postfixadmin goes there should be no need for postfix-mariadb since postfix works fine with postfix-mysql and mariadb-client/server | 23:20 |
| lamont | tbf, mostly because the original example code was totally clueless in that regard (what security??? what's that?) and people keep copying it without getting the memo about it being braindead | 23:20 |
| Belldandu | i actually have an email service with postfix/dovecot/postfix-mysql/mariadb | 23:20 |
| Belldandu | and its running completely fine xD | 23:20 |
| lamont | Belldandu: so what you're telling me is that mariadb hijacks^Wreplaces mysql below the library layer? | 23:21 |
| Belldandu | yeah | 23:21 |
| Belldandu | its a complete replacement | 23:21 |
| lamont | then the postfix depends are completely correct, and there is no need for postfix-mariadb | 23:21 |
| Belldandu | yeah | 23:21 |
| lamont | so why are we having this conversation? :D | 23:21 |
| Belldandu | because postfixadmin is being a dork and uninstalling mariadb because its not a alternate dependency for mysql | 23:22 |
| Belldandu | hence why people suggested it depend on virtual-mysql-client instead of mysql-client | 23:22 |
| lamont | all of this has nothing to do with the postfix package, which (properly, you tell me) depends on libmysqlclient20 | 23:23 |
| Belldandu | yeah | 23:24 |
| * lamont goes back to the day-job | 23:24 | |
| Belldandu | we have been complaining about postfixadmin | 23:24 |
| Belldandu | not postfix | 23:24 |
| Belldandu | postfixadmin is a php layer over top of postfix | 23:24 |
| Belldandu | it doesnt come with postfix | 23:24 |
| Belldandu | its a seperate package | 23:24 |
| lamont | and will always remain such. :D | 23:25 |
| Belldandu | yeah | 23:25 |
| Belldandu | but with postfixadmin removing mariadb when postfix doesnt. It makes installing it from the repo pointless | 23:25 |
| Belldandu | postfixadmin basically breaks any postfix install thats using mariadb | 23:26 |
| Belldandu | not to mention anything else using mariadb. simply by installing it | 23:26 |
| Belldandu | the problem in itself is the postfixadmin dependencies | 23:27 |
| nacc | Belldandu: i think we get that now, and we know the fix (and there is already a debian bug), right? and it's fixed in the upstream 2.3.8? | 23:30 |
| Belldandu | is 2.3.8 Xenial? | 23:30 |
| Belldandu | if so then i'll be upgrading later | 23:31 |
| nacc | Belldandu: not even in debian yet | 23:35 |
| nacc | afaict | 23:35 |
| Belldandu | o.O | 23:47 |
| tarpman | huh | 23:48 |
| tarpman | nacc: postfixadmin's mysql-client dependency is not yet fixed in unstable, no. but it's a packaging question afaict, not an upstream one... | 23:49 |
| tarpman | s/mysql-client dependency/mysql-{client,server} dependencies/ | 23:50 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!