| darkxst | pitti, hi, more systemd woes ;( not getting any filesystems (from fstab) mounted, seemingly triggered by glib on ubuntu-desktop ppa | 06:18 |
|---|---|---|
| pitti | Good morning | 06:19 |
| pitti | hey darkxst | 06:19 |
| pitti | darkxst: oh? would you mind booting with "debug", and put the "journalctl" output to a bug report? | 06:20 |
| darkxst | hey | 06:20 |
| darkxst | ok | 06:21 |
| pitti | darkxst: glib is rather unlikely, but it may have triggered some timing stuff | 06:21 |
| pitti | s/stuff/difference/ | 06:21 |
| pitti | darkxst: oh, not in #u-desktop any more? | 06:22 |
| darkxst | pitti, on laptop | 06:22 |
| pitti | darkxst: so, by the release I plan to fix that "adm" users can access the journal | 06:22 |
| darkxst | most normal users are in that group? | 06:23 |
| pitti | darkxst: debian bug 771980 | 06:24 |
| ubottu | Debian bug 771980 in systemd "systemd: /run/log/journal is not readable by the adm group" [Minor,Open] http://bugs.debian.org/771980 | 06:24 |
| pitti | darkxst: no, only the first one (which also gets sudo); other than that, that should be limited to administrators | 06:24 |
| pitti | darkxst: exactly the same as syslog, BTW | 06:24 |
| pitti | -rw-r----- 1 syslog adm 809053 Feb 9 07:24 /var/log/syslog | 06:25 |
| darkxst | pitti, right, so there is really no way to get logs then from apport? | 06:25 |
| pitti | darkxst: you mean now, or after fixing the above? | 06:27 |
| pitti | darkxst: nothing regressed under systemd, you can still get syslog etc.; but of course accessing the per-unit journal would be nicer | 06:27 |
| pitti | darkxst: if you aren't in "adm", then the hook could still ask for it with root privs (you'll get a polkit prompt then, and an admin has to authenticate) | 06:28 |
| darkxst | pitti, well currently we grab the upstart user logs for say gnome-session | 06:28 |
| pitti | darkxst: oh, user | 06:28 |
| pitti | darkxst: nothing has changed for the user session | 06:28 |
| darkxst | maybe its a GNOME thing? but no user logs with that stuff now | 06:30 |
| pitti | darkxst: I don't know, did GNOME Ubuntu ever use upstart sessions? | 06:31 |
| pitti | I thought all of our DEs would | 06:32 |
| darkxst | pitti, yes | 06:32 |
| pitti | and session init is still upstart | 06:32 |
| darkxst | pitti, I am 100% we do not get logs in .cache/upstart | 06:32 |
| darkxst | and also a couple of wierd bugs due to upstart user sessions disappear | 06:33 |
| darkxst | under systemd init | 06:33 |
| pitti | darkxst: oh, indeed; I only have one recent file there, .cache/upstart/unity-panel-service.log | 06:33 |
| pitti | but also some rotated stuff from yesterday | 06:33 |
| darkxst | pitti, bug 1419623 | 06:35 |
| ubottu | bug 1419623 in systemd (Ubuntu) "systemd init is no mounting drives from fstab" [Undecided,New] https://launchpad.net/bugs/1419623 | 06:35 |
| darkxst | and bug 1385572 is the reason I think there is no upstart user sessions under systemd init | 06:44 |
| ubottu | bug 1385572 in upstart (Ubuntu) "gnome-session not shutting down cleanly" [Low,Confirmed] https://launchpad.net/bugs/1385572 | 06:45 |
| === darkxst is now known as darkxst__ | ||
| === darkxst_ is now known as darkxst | ||
| pitti | darkxst__: I suppose logging into a VT works and puts you into the root dir; there sudo mount -a should work? | 07:02 |
| pitti | darkxst__: (just making sure it's not something serious with util-linux or the actual partitions) | 07:02 |
| darkxst | pitti, yes that is what I have been doing | 07:02 |
| pitti | cyphermox: oh, nothing wrong with python-dbusmock -- it just detected gnome bug 734466 :) | 07:05 |
| ubottu | Gnome bug 734466 in nmcli ""nmcli dev wifi" crashes with multiple wifis" [Normal,Resolved: fixed] http://bugzilla.gnome.org/show_bug.cgi?id=734466 | 07:05 |
| dholbach | good morning | 07:26 |
| pitti | infinity: hm, the latest "patch" release fixes the dangling symlink problem for most packages, but apparently not glibc yet; I wonder what kind of funky thing it tries to do there, symlink outside the build tree or so? | 07:52 |
| Mirv | can a core-dev "ack" https://ci-train.ubuntu.com/job/ubuntu-landing-006-2-publish/lastSuccessfulBuild/artifact/packaging_changes_ubuntu-ui-toolkit_1.1.1403+15.04.20150206-0ubuntu1.diff ? both new build-deps are in main. | 07:52 |
| infinity | pitti: The new glibc should be fine, no? | 07:58 |
| infinity | pitti: It unpacks and builds locally, at least. | 07:59 |
| infinity | pitti: Seems to be going happily on the buildds too. | 07:59 |
| pitti | Mirv: LGTM | 07:59 |
| infinity | pitti: No idea, TBH, why -13ubuntu1 is/was sad, but I think we're past that. | 07:59 |
| pitti | infinity: hm, I still see it on http://d-jenkins.ubuntu-ci:8080/job/vivid-adt-glibc/27/ARCH=i386,label=adt/console which ran on Feb 6 | 08:00 |
| pitti | glibc_2.19-13ubuntu3.dsc | 08:00 |
| pitti | infinity: ok, if it works locally now I'll investigate, thanks | 08:00 |
| infinity | pitti: -15ubuntu1 is the one that works. | 08:01 |
| LocutusOfBorg1 | hi people | 08:02 |
| pitti | hm, why didn't that run | 08:02 |
| infinity | pitti: It only uploaded 43m ago. | 08:02 |
| pitti | oh! just from 33 minutes ago :) | 08:02 |
| Mirv | thanks martin | 08:06 |
| === mfisch is now known as Guest22817 | ||
| pitti | wgrant: hey William, how are you? | 08:41 |
| pitti | wgrant: not sure if you saw my email about RTM langpack exports the other day, but would we have enough capacity to build them more often? like twice a week? | 08:42 |
| === kickinz1|afk is now known as kickinz1 | ||
| === work_alkisg is now known as alkisg | ||
| alkisg | mvo: Good morning! About the update-manager SRU, and a possible re-upload without the other changes, last Friday you said "I look at this, in a call right now" - did you? Should I wait for a re-upload, or just try to make SRUs out of the other changes included in the same upload? | 09:00 |
| wgrant | pitti: Ah, missed that. RTM exports only take about 90 minutes, so we can do them daily, particularly since the new DB servers are slowly coming online. I'll have to add a new flag for the persistent full exports, though. How pressing is this? | 09:00 |
| alkisg | https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1415785 | 09:01 |
| ubottu | Launchpad bug 1415785 in update-manager (Ubuntu Precise) "Please remove all ltsp* blacklisting" [High,In progress] | 09:01 |
| pitti | wgrant: vrr knows more about how pressing it is, but I suppose it's the kind of "the earlier the better", but "not world breaking" class | 09:01 |
| pitti | wgrant: cjwatson proposed to just add the "export full pack" to the API so that we can set it via cron | 09:01 |
| wgrant | pitti: If you're happy with that then that's fine too. | 09:02 |
| wgrant | It's a little easier from our end. | 09:02 |
| pitti | wgrant: yes, that'd work well enough for me | 09:03 |
| pitti | le huh? my mouse pointer just became invisible; VT switching and switching between different windows (with different cursor shapes) doesn't help | 09:04 |
| pitti | is there a trick to make it visible again? | 09:04 |
| ogra_ | wiggle it | 09:05 |
| ogra_ | :) | 09:05 |
| pitti | mlankhorst, tjaalton ^ do you happen to know? | 09:05 |
| alkisg | Unplug it from the usb port and plug it again? :) | 09:06 |
| pitti | alkisg: no; the mouse clearly works (FFM, marking text, clicking, etc.) | 09:07 |
| pitti | I just don't see the cursor | 09:07 |
| pitti | (unpluggging, VT switch etc. doesn't help) | 09:07 |
| * ogra_ hands pitti some http://i.stack.imgur.com/DVuWk.png | 09:08 | |
| pitti | :) | 09:08 |
| pitti | well *shrug*, I'll just reboot as soon as I can | 09:08 |
| tjaalton | try restarting compiz? | 09:08 |
| tjaalton | gtk can hide it, maybe it got stuck | 09:08 |
| tjaalton | this happens frequently for me when logging in, but it comes back once everything is loaded | 09:09 |
| pitti | nope, restarting compiz didn't help either | 09:11 |
| pitti | anyway, probably sun rays / rare graphics driver bug | 09:11 |
| mlankhorst | pitti: no idea, I know gsd hides the cursor if no mouse is found, annoying with synergy :P | 09:12 |
| pitti | oh wow, and all of a sudden it popped up again | 09:12 |
| pitti | mlankhorst: ah, perhaps some weirdness with the thing that hides the touchpad and/or mouse cursor on inactivity | 09:13 |
| pitti | syndaemon | 09:13 |
| mlankhorst | sounds likely | 09:13 |
| tjaalton | is that even used anymore? | 09:14 |
| tjaalton | guess it shouldn't | 09:14 |
| pitti | yeah, if you enable the option in control-center | 09:14 |
| pitti | now I remember that I enabled it in in the train last weekend | 09:14 |
| tjaalton | sounds legacy | 09:14 |
| tjaalton | anyway :) | 09:14 |
| pitti | but I thought that was for disabling the touch pad during typing, not disabling the mouse pointer on inactivity | 09:14 |
| pitti | we once had something for the latter in the past ("unclutter"), but I thought that has long been obsolete | 09:15 |
| pitti | and/or went into GTK or whereever | 09:15 |
| tjaalton | yep | 09:15 |
| === kickinz1 is now known as kickinz1|afk | ||
| === kickinz1|afk is now known as kickinz1 | ||
| === work_alkisg is now known as alkisg | ||
| LocutusOfBorg1 | hi people does anybody know if bug 1417716 is affecting virtualbox or sysvinit? | 09:36 |
| ubottu | bug 1417716 in virtualbox (Ubuntu) "package virtualbox 4.3.18-dfsg-2ubuntu1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,Incomplete] https://launchpad.net/bugs/1417716 | 09:36 |
| LocutusOfBorg1 | I can't figure it out | 09:36 |
| LocutusOfBorg1 | seems more bug 1385817 | 09:36 |
| ubottu | bug 1385817 in Release Notes for Ubuntu "initscripts package fails to upgrade if there are local init scripts on the system with no LSB headers" [High,Triaged] https://launchpad.net/bugs/1385817 | 09:36 |
| === kickinz1 is now known as kickinz1|afk | ||
| === kickinz1|afk is now known as kickinz1 | ||
| xnox | LocutusOfBorg1: most of the time it's neither. | 09:46 |
| xnox | LocutusOfBorg1: it's duplicate of "initscripts must have sysv headers" | 09:47 |
| xnox | bug | 09:47 |
| === Tribaal_ is now known as Tribaal | ||
| LocutusOfBorg1 | thanks xnox :) | 09:58 |
| LocutusOfBorg1 | I was so tempted to close as duplicate | 09:58 |
| infinity | pitti: That calibre/powerpc failure that references 'Pool(cpu_count)' ... It's not losing its mind because sagari has 24 CPUs, perhaps? | 10:02 |
| pitti | oh, I didn't even know about it | 10:03 |
| pitti | I guess I don't get ftbfs mail for autosynced packages | 10:03 |
| infinity | pitti: Yeah, cause you don't get mail from LP... That. | 10:03 |
| pitti | thread.error: can't start new thread | 10:04 |
| seb128 | bdmurray, hey, is there a known issue with e.u.c and crossing of lines corresponding to fixed bugs? it seems it stopped doing that since the "... seen" columns indicate series rather than versions? | 10:04 |
| infinity | pitti: I'm betting that if I re-run it on one of the <= 8 CPU buildds, it'll work fine. But that's just a wild guess. | 10:04 |
| pitti | sounds a bit like debian bug 760865 | 10:04 |
| ubottu | Debian bug 760865 in src:calibre "calibre: FTBFS on mips: thread.error: can't start new thread" [Serious,Fixed] http://bugs.debian.org/760865 | 10:04 |
| pitti | I coudl extend the workaround to powerpc | 10:05 |
| pitti | but that was also just a workaround without truly understanding the issue | 10:05 |
| infinity | pitti: Yeah, the workaround definitely sounds incorrect. :P | 10:06 |
| pitti | $ python -c 'from multiprocessing import cpu_count; print(cpu_count())' | 10:06 |
| pitti | 4 | 10:06 |
| infinity | pitti: And I bet I can reproduce on other machines with sufficiently large numbers of cores/threads. | 10:06 |
| pitti | so that would print 24, but it cannot actually handle 24 threads? | 10:06 |
| infinity | pitti: It can handle 24 threads fine, but I suppose it depends on what "handle" means. | 10:07 |
| mvo | alkisg: hi, sorry for the slow reply, I added a sru header to the bugreport now where it was missing (the https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1402706) | 10:07 |
| ubottu | Launchpad bug 1402706 in update-manager (Ubuntu Precise) "More explicit wording needed for HWE updates" [High,In progress] | 10:07 |
| infinity | pitti: 24 threads consuming massive amounts of memmory each could explode, for instance. | 10:07 |
| alkisg | mvo: hi, I think there are 2 more involved bug reports that lack SRU headers? Let me check... | 10:07 |
| alkisg | mvo: LP #1415785, #1402706, #1341320, #1343296 | 10:08 |
| ubottu | Launchpad bug 1341320 in update-manager (Ubuntu) "empty apt-get install command suggested by hwe-support-status" [Undecided,Confirmed] https://launchpad.net/bugs/1341320 | 10:08 |
| ubottu | Launchpad bug 1415785 in update-manager (Ubuntu Precise) "Please remove all ltsp* blacklisting" [High,In progress] https://launchpad.net/bugs/1415785 | 10:08 |
| ubottu | Launchpad bug 1343296 in update-manager (Ubuntu) "hwe-support-status does not output all necessary commands for upgrade" [Undecided,New] https://launchpad.net/bugs/1343296 | 10:08 |
| mvo | alkisg: let me check | 10:08 |
| infinity | pitti: It would be better to find out *why* we're seeing "thread.error: can't start new thread" | 10:10 |
| infinity | pitti: Cause it's nonsensical to make a blanket statement like "arch $foo can't handle $bar threads", given than a single-core machine can run thousands of threads. | 10:10 |
| infinity | pitti: But if it's trying to eat a ton of RAM with each, or hitting some bizarre internal counter or something, that would be worth knowing. | 10:11 |
| infinity | pitti: I figure that replacing the autodetection with a static "200" and testing locally would probably show the same bug. | 10:11 |
| infinity | pitti: Or, from a random stackoverflow hit: "You're using a 32bit system and running out of virtual memory. One of your libraries is likely spawning threads and not reclaiming them correctly. As a workaround, try reducing the default thread stack size with threading.stack_size." | 10:15 |
| infinity | pitti: So, maybe just wise to limit your thread count to "NUM_CORES || 8", whichever is smaller, or do so only on 32-bit systems... | 10:15 |
| infinity | pitti: Seems like python's just exhibiting a bit of suck here. (not surprising, from how much I know about python threading sucking) | 10:16 |
| === work_alkisg is now known as alkisg | ||
| pitti | infinity: ah yeah, trying with "200" on a low-ram VM or so could work | 10:16 |
| pitti | (to try and reproduce) | 10:17 |
| infinity | pitti: Note that that machine is hardly "low RAM". It's 16G, but yeah, also 32-bit, so python's only getting 2G (or 3G, I can't remmeber what the user/kernel split on ppc32 is). | 10:17 |
| pitti | ah ok, so "200" on an i386 VM should be reasonably close then | 10:18 |
| infinity | pitti: So, reproducing on i386 with a large thread count should be easy, I'd think. | 10:18 |
| infinity | pitti: Yeah. | 10:18 |
| popey | pitti: seems something changed recently in udev (perhaps) on vivid, as I can no longer "adb devices". adbd has to be started as root. This seems to be a regression over the last week or so. | 10:20 |
| pitti | popey: i. e. your /dev/usb/... is no longer accessible to you? | 10:21 |
| pitti | popey: you should get an automatic ACL to it | 10:22 |
| ogra_ | pitti, he can see the device when he runs "adb devices" as root | 10:22 |
| ogra_ | (after adb kill-server on the PC) | 10:22 |
| pitti | (check lsusb for bus/device number, then getfacl /dev/bus/usb/001/002 and change the numbers accordingly) | 10:22 |
| infinity | pitti: Anyhow, it would make sense if mips is seeing a similar issue, cause Debian's mips buildds are Cavium CPUs with a ridiculous number of cores, and 32-bit userspace. | 10:23 |
| ogra_ | http://paste.ubuntu.com/10140289/ | 10:23 |
| ogra_ | this is what we ship with the android-tools-adb package | 10:23 |
| infinity | pitti: Which is something no other port has. I don't think any other 32-bit port goes over 8 cores in any Debian or Ubuntu buildd configuration. | 10:23 |
| popey | pitti: http://paste.ubuntu.com/10140305/ | 10:25 |
| * ogra_ sees infinity's lanst android upload and thinks we really need to get rid of these win32 build deps at some point | 10:25 | |
| pitti | user:alan:rw- | 10:25 |
| pitti | popey: ^ LGTM | 10:25 |
| popey | hm, so I wonder why adb devices returns no devices then. | 10:25 |
| ogra_ | well, adb didnt change since november | 10:26 |
| infinity | ogra_: I'm not convinced mingw32 is actually used in the build (a grep in the build log didn't seem to show it being called), but I didn't feel the urge to experiment either. | 10:26 |
| popey | http://paste.ubuntu.com/10140312/ | 10:26 |
| ogra_ | yeah | 10:26 |
| pitti | hard to tell; perhaps stracing adb as user reveals some EPERM thing someplace else? | 10:26 |
| pitti | (FTR, works fine here with nexus 4 on current vivid) | 10:27 |
| popey | hmm, ok. thanks. | 10:27 |
| pitti | popey: can you run adbd in the foreground? perhaps that already spits out some error message | 10:27 |
| popey | will try | 10:28 |
| ogra_ | not adbd ... adb | 10:29 |
| ogra_ | adbd is the bit running on the phone | 10:29 |
| === work_alkisg is now known as alkisg | ||
| pitti | ogra_: ah, ok; I mean the server bit that it forks off into the background | 10:30 |
| pitti | I don't see a CLI switch to run it in foreground, so I guess strace it is :/ | 10:30 |
| ogra_ | right, not sure you can run it easily in the fg | 10:31 |
| popey | :( | 10:35 |
| popey | aha! | 10:37 |
| popey | I think some path has changed and it's found another version of adb? http://paste.ubuntu.com/10140430/ | 10:38 |
| pitti | /home/alan/tools/android/eclipse-adt/sdk/platform-tools/adb | 10:38 |
| popey | \o/ "/usr/bin/adb devices" works! | 10:38 |
| pitti | so you have that in your $PATH? | 10:38 |
| popey | yeah, blame didrocks ㋛ | 10:38 |
| popey | umake :) | 10:39 |
| popey | looks like it was added to my bashrc | 10:39 |
| didrocks | popey: that's what you gain by listening to users! :) | 10:39 |
| popey | I know right!? | 10:39 |
| didrocks | why the adb version here doesn't work for you? (should I backlog?) | 10:39 |
| popey | ogra_: did we patch adb in android-tools-adb? | 10:40 |
| ogra_ | popey, sure | 10:40 |
| didrocks | ah, here we go :) | 10:40 |
| popey | :) | 10:40 |
| ogra_ | popey, adb abd adbd are both a cmpolete patch | 10:40 |
| didrocks | popey: I purified your bashrc with upstream goodness :) | 10:41 |
| popey | I feel blessed. | 10:41 |
| ogra_ | (not sure we make any changes to it though ) | 10:41 |
| ogra_ | (for adb that is, adbd is heavily changed) | 10:41 |
| * popey modifies his bashrc | 10:42 | |
| * popey has learned things today. thanks chaps :) | 10:42 | |
| === didrocks1 is now known as didrocks | ||
| alkisg | mvo: should I help with the SRUs for the other 2 bug reports? | 11:09 |
| alkisg | (sorry if you talked to me and I didn't receive it, laggy network today here...) | 11:09 |
| caribou | tkamppeter: Hi, remember the cups fix you sponsored for me recently ? | 11:12 |
| caribou | tkamppeter: I have just uploaded the debdiffs for Trusty & Utopic. want to sponsor those as well ? | 11:12 |
| === alkisg is now known as work_alkisg | ||
| tkamppeter | caribou, Utopic I have uploaded now, now I will do Trusty. | 11:27 |
| === _salem is now known as salem_ | ||
| === MacSlow is now known as MacSlow|lunch | ||
| caribou | tkamppeter: perfect! I've already built & tested on both; I'll add the SRU team once you're done | 12:07 |
| tkamppeter | caribou, Trusty is now done, too. | 12:09 |
| caribou | tkamppeter: great, thank you ! | 12:09 |
| tkamppeter | caribou, thanks for setting up the CUPS SRUs. | 12:10 |
| === damon is now known as ngaio | ||
| === darkbasic_ is now known as darkbasic | ||
| === ara is now known as Guest94549 | ||
| === MacSlow|lunch is now known as MacSlow | ||
| === darkbasic_ is now known as darkbasic | ||
| tkamppeter | caribou, the CUPS SRU path for Trusty is still blocked by bug 1386241. I will look into what can be done here. The Utopic SRU can already be processed though. | 14:13 |
| ubottu | bug 1386241 in system-config-printer (Ubuntu Trusty) "Add the full IPP Everywhere support from Utopic to Trusty" [High,Triaged] https://launchpad.net/bugs/1386241 | 14:13 |
| caribou | tkamppeter: ok, thanks for checking that out | 14:13 |
| === marcusto_ is now known as marcustomlinson | ||
| === negronjl_ is now known as negronjl_afk | ||
| bdmurray | seb128: not, that I'd heard of | 15:09 |
| === darkbasic_ is now known as darkbasic | ||
| ricotz | mvo, hi, please take a look at the attached debdiff here https://bugs.launchpad.net/ubuntu/+source/aptitude/+bug/1399582 | 15:38 |
| ubottu | Launchpad bug 1399582 in aptitude (Ubuntu) "aptitude fails to fetch changelogs because of illformed URL request" [High,Confirmed] | 15:38 |
| === Guest22817 is now known as mfisch | ||
| === sabdfl_ is now known as sabdfl | ||
| sabdfl | ScottK, wow, that was a fun memory! | 15:56 |
| sabdfl | or, sort of memory :) | 15:56 |
| sabdfl | the sea survival training was a bit of a blur | 15:57 |
| === kickinz1 is now known as kickinz1|afk | ||
| === a7med is now known as Neo31 | ||
| bdmurray | seb128: I'm looking into it though, thanks | 16:47 |
| seb128 | bdmurray, can you confirm? want me to report a bug? | 16:48 |
| bdmurray | seb128: I can confirm, it works better on a filtered page like https://errors.ubuntu.com/?release=Ubuntu%2015.04&package=media-hub&period=day and a bug would be helpful | 16:51 |
| seb128 | bdmurray, against what component? | 16:52 |
| bdmurray | launchpad.net/errors/ | 16:52 |
| seb128 | bdmurray, https://bugs.launchpad.net/errors/+bug/1419878 | 16:54 |
| ubottu | Launchpad bug 1419878 in Errors "default view doesn't seem to cross lines corresponding to fixed bugs" [Undecided,New] | 16:54 |
| === rickspencer3_ is now known as rickspencer3 | ||
| alexbligh1 | Dependency question. Package A provides, conflicts, and replaces B. Package C conflicts with package B. The idea here was for an upgrade from A(old) to A to remove package B, but to optionally let a legacy version be reinstalled (package C shares some of the same files as package B). What happens in practice is package C can't be installed without removing A. Any ideas? | 18:14 |
| cjwatson | alexbligh1: Make the Conflicts from C to B versioned, so that it won't match the Provides. | 18:15 |
| cjwatson | E.g. <= max-version-that-ever-existed, or even just >= 0 | 18:15 |
| alexbligh1 | cjwatson, aha! very clever. Thanks | 18:16 |
| rharper | hallyn: stgraber: is there a non-interactive form of lxc remote add <url> when the password is set? currently it prompts when password is set, would like to do this automatically | 19:00 |
| hallyn | rharper: I work around it by adding the client certificsate in the server's certs dir | 19:01 |
| hallyn | then server doesn't ask for pwd | 19:01 |
| rharper | hallyn: hrm... for some reason I thought I needed to do the pass bits before the certs were generated | 19:02 |
| rharper | I'm doing: 1. start lxd locally, 2) lxc config set password <pass> 3) I want to add the remote -- 4) copy client certs ... but you're saying I can swap 4 and 3 ? | 19:03 |
| * rharper gives it a go | 19:03 | |
| hallyn | rharper: yeah just do a 'lxc remote list' first to generate th client certs, copy them over, restart lxd | 19:04 |
| rharper | cool | 19:04 |
| hallyn | i've got a juju charm that tries to do this. (but fails at some point, and i've neglected it) | 19:04 |
| rharper | I get fingerprint prompt | 19:05 |
| rharper | hallyn: oh, I should look at what you have then | 19:05 |
| * rharper is charming lxd/lxc stuff as well | 19:05 | |
| hallyn | rharper: oh excellent then i may not have to :) what i've got is at... oh lemme re-push | 19:07 |
| rharper | hrm, looks like somethin with config trust list might help fill out the client/servercert/local.crt bits... | 19:07 |
| hallyn | lp:/~serge-hallyn/charms/trusty/ovs-lxd/trunk | 19:07 |
| hallyn | ? | 19:08 |
| rharper | http://paste.ubuntu.com/10146613/ | 19:09 |
| rharper | hallyn: I noticed that once I do the remote add, I get a new file in ~/.config/lxc/servercert/<name>.crt | 19:09 |
| rharper | I assume once that's present (from accepting the fingerprint) that we don't get prompted again | 19:09 |
| hallyn | right | 19:09 |
| rharper | but I still want a way to accept the server cert without prompting | 19:09 |
| hallyn | but 'config trust', if that was implemented, i missed the pull request to od that :) obviously shouldn't crash though | 19:09 |
| hallyn | raise a github issue :) | 19:10 |
| rharper | heh | 19:10 |
| hallyn | that tych0 chap will get on it in a jiffie | 19:10 |
| hallyn | uqestion is how would we make that safe? | 19:10 |
| rharper | right, so what would let me pull down the server cert into the local config ? | 19:10 |
| rharper | I assume the same way ssh does it (you put your hands in your ears and supply a parameter that says I don't care) | 19:11 |
| hallyn | i understnad on a trusted network with auto-instlal it's what you want. but how do we keep users from then using that to their ec2 host | 19:11 |
| rharper | but seriously it'd need to be done via some published fingerprint ... maybe gpg keyring publishing ? | 19:11 |
| hallyn | yeah i think best ot discuss this on a github issue | 19:11 |
| * rharper is way out of his comfort zone | 19:11 | |
| hallyn | with certs, or with github issues? | 19:12 |
| hallyn | btw what are you doing, testing nc-lxd? | 19:13 |
| rharper | yes, nc-lxd, charming it up while zul works on kilo stuff; | 19:15 |
| rharper | hallyn: I see now how yuo get the servercert | 19:15 |
| rharper | in the nc-lxd case, it's eaiser since the server/client are the same machine | 19:16 |
| rharper | I just need to fish out the lxd server cert file | 19:16 |
| rharper | copy it to the client config location | 19:16 |
| rharper | and won't get the prompt | 19:16 |
| zul | rharper: im going to be using the unix socket daemon in the next iteration | 19:16 |
| hallyn | don't make him cry | 19:17 |
| zul | hallyn: i dunno...rhaper is a big boy | 19:17 |
| rharper | sometimes | 19:17 |
| === roadmr is now known as roadmr_afk | ||
| === pgraner is now known as pgraner-afk | ||
| === roadmr_afk is now known as roadmr | ||
| israel | does anyone know how to set _NET_WM_ICON easily in C++ I haven't been able to easily find documentation on this | 20:48 |
| dobey | israel: in qt? | 20:59 |
| israel | No dobey in FLTK... sorry I should have specified this.. | 21:00 |
| israel | I was going to try to simple set it via xkb stuff (or xlib) but I am not very familiar with that | 21:00 |
| israel | the documentation for xkb is not very good for someone unfamiliar with it | 21:01 |
| dobey | israel: i think you'd use Fl_Window::iconlabel("exe-name") and it will pull the icon from the theme and set the appropriate x window hints | 21:02 |
| dobey | xkb is for keyboards | 21:02 |
| dobey | so i can see how that would be confusing :) | 21:02 |
| israel | heheh sorry xcb | 21:02 |
| israel | too many x acronyms | 21:02 |
| israel | i will try it dobey and let you know | 21:03 |
| dobey | at least, that's what i gathered from a simple "fltk set window icon" search :) | 21:04 |
| israel | not quite working.... dobey thanks for the try :) | 21:08 |
| dobey | israel: if fltk doesn't provide working facilities for it, and you don't care about portability, then you'll need to use Xlib indeed; best place to look at that is probably the gdk-x11 code in gtk+ | 21:10 |
| Noskcaj | cyphermox, What are we still waiting on for bluez5? | 21:11 |
| israel | thanks dobey it is GNU/Linux only (JWM specificly) I will check it out! | 21:15 |
| cyphermox | Noskcaj: just coordinating things, I'm not convinced that we are waiting on much. You should ask seb though | 21:32 |
| cyphermox | there's already a PPA with the transition in progress: ppa:ubuntu-desktop/transitions | 21:32 |
| arges | hallyn: hey | 21:34 |
| arges | hallyn: almost have this patch done for disabling ksm in vm. found an issue when testing. | 21:34 |
| hallyn | arges: ok, cool. i'm doing the merge right now. head spinning a bit, will bbiab | 21:34 |
| arges | hallyn: yea no hurry, i'll just post it to the bug when its ready | 21:35 |
| hallyn | arges: ok, thanks | 21:37 |
| rharper | hallyn: stgraber: where do you get updated golang package for trusty ? | 21:47 |
| roaksoax | ScottK: do you have any plans of updating psycopg2 to 2.6 in debian anytime soon? | 22:05 |
| ScottK | roaksoax: I hadn't noticed it was out. I can look at doing an upload to experimental. Possibly as soon as Friday. | 22:07 |
| roaksoax | ScottK: yeah, I just noticed today. And thanks! Let me know if you do so we can make sure it makes it into vivid | 22:07 |
| hallyn | rharper: https://launchpad.net/~ubuntu-lxc/+archive/ubuntu/lxd-daily | 22:11 |
| rharper | hallyn: thanks! | 22:11 |
| stgraber | rharper: ubuntu-lxc/lxd-daily | 23:02 |
| rharper | stgraber: thanks! | 23:03 |
| === mwenning is now known as mwenning-rr5 | ||
| aeoril | How do I use /debian/rules to configure a make environment for an Ubuntu package? Or, where would I find that info on the web? | 23:55 |
| aeoril | Should I set up my build environment using this tutorial, or is it out of date? https://wiki.ubuntu.com/BeginnersTeam/FocusGroups/Development/Devbeginnings | 23:56 |
| aeoril | (I think the BeginnersTeam was disolved some time ago, IIRC) | 23:57 |
| sarnold | aeoril: I suggest this instead of the pbuilder: https://wiki.ubuntu.com/SimpleSbuild | 23:57 |
| aeoril | sarnold hey! Thanks! | 23:57 |
| sarnold | aeoril: but e.g. Unit193 said he uses pbuilder, so it's not like pbuilder is wrong; I've heard though that sbuild does a better job approxmiating the buildds | 23:58 |
| aeoril | sarnold I remember Unit193 - I used to interact with him a lot on #ubuntu-beginners-team or whatever ... | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
