=== darth_chaos7 is now known as chaos7theory === juliank_ is now known as juliank [02:43] bdmurray, https://news.ycombinator.com/item?id=10262719 ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster (scylladb.com) [02:43] do we need/want that for the whoopsie / daisy stuff? [02:46] http://blog.octo.com/en/scylladb-vs-cassandra-towards-a-new-myth/ [03:00] i don't understand the process for making network config changes via netplan after boot [03:01] you edit files in /etc/netplan, run /lib/netplan/generate... and restart something? [03:04] systemctl try-restart systemd-networkd.service i guess [03:04] slangasek: does this sound right? ^ [03:06] mwhudson: I was under the impression you didn't need to restart networkd because it would read its config dynamically, but ICBW [03:06] oh ok [03:06] that would seem kinda un-systemd, but certainly useful [03:17] it'd suck to coordinate changes to N interfaces that way though; you'd have to stage them all outside the directory and move them in in one go, and hope to hell that you could win the race.. [03:18] google suggests that you have to restart it [03:18] systemd is not dynamic, you have to manually reload for it to pickup changes to service files. Chances of it re-reading here would be minimal. === chaos7theory is now known as darth === darth is now known as chaos7theory [06:07] good morning [07:25] mardy: ping? is mesa in xenial-proposed now good for arm64? [07:25] mardy: because I need to start another sru asap [07:28] tjaalton: hi! I'm not sure how to test it... [07:28] tjaalton: the issue was found when trying to build some package on the citrain [07:29] tjaalton: I wonder, should I make a citrain request just to test this? [07:30] mardy: perhaps? [07:30] tjaalton: uh... and how can I tell the citrain to pull from xenial-proposed? [07:31] sil2100: any idea? ^ [07:31] sil2100: speaking of which, xorg-server in xenial needs verification === hikiko is now known as hikiko|bbl [07:57] dholbach: around? [07:57] ginggs, yes [08:06] mardy: hey! All citrain silos by default use -proposed [08:07] tjaalton: ok, I can verify our part there, thanks for the reminder [08:07] sil2100: excellent, thanks [08:09] sil2100, tjaalton: ok, so I'll prepare a silo to test it === JanC is now known as Guest88920 === JanC_ is now known as JanC === hikiko|bbl is now known as hikiko [10:46] * LocutusOfBorg just uploaded sbuild in yakkety, lets see how many things broke! [12:18] hum, deleted a preinstalled windows partition and now booting xenial takes 60s extra [12:35] tjaalton: ... wat? [12:37] well, it's been bugging me for a while on this machine.. uefi, installed ubuntu, deleted (corrupted) win8 install and recovery partitions, and now booting up takes quite a while. there's a 60s timeout somewhere [12:38] is that partition still listed in fstab? [12:38] tried creating new partitions on them, next I'll put an fs there [12:38] no [12:41] tjaalton, go to grub, edit boot line to include systemd.debug-shell to the kernel cmdline, boot, ctrl-alt-f9 and systemctl status [12:41] that should tell you what jos is waiting [12:41] likely timeouting on a mount unit [12:41] but unsure where it's coming from if it's not fstab [12:41] ok, thx [12:41] yw [12:41] jos->job [13:03] the shell started after lightdm was up [13:10] tjaalton, the shell? [13:10] debug shell, vt9 was empty until too late [13:10] like you couldn't go to vt9 during that 1 min hang? [13:10] weird :-/ [13:11] I'll try again [13:11] tjaalton, try to systemd-analyze blame ? [13:12] doesnt that just print "pitti" all the time ? :P [13:12] good that he's on holidays ;-) [13:12] that's git blame :) [13:12] (sometimes "lennart" if it is an upstream issue ) [13:17] blame shows apparmor on the top with ~1.5s [13:17] state is degraded, three failed units [13:20] which ones? [13:23] mdeslaur hi, willing to sponsor a curl merge? [13:23] or anybody else [13:23] LocutusOfBorg: sure [13:24] thanks, grabbing it now from incoming [13:24] lots of cve fixes [13:26] " Drop dependencies not in main:" [13:26] is that still needed after the universe archive reorg? [13:26] * LocutusOfBorg is not sure anymore [13:27] The change is that you can build depend on stuff in universe. [13:27] THere is no change to the rules about binary dependencies. [13:28] so, while I can pull from universe on B-D, I can't runtime-depend on it [13:28] right? [13:29] Right. Though this means that you could split and add a binary package that lives in universe and then depends on stuff in universe. That wasn't possible before. [13:32] thanks [13:35] I re-enabled stunnel4, but the openssh removal is still needed [13:35] mdeslaur, https://launchpad.net/%7Ecostamagnagianfranco/+archive/ubuntu/locutusofborg-ppa/+sourcepub/6772977/+listing-archive-extra :) [13:35] I hope that dsc is ok for you? [13:41] or mdeslaur there is also an openssl merge FYU [13:41] FYI [13:41] https://launchpad.net/%7Ecostamagnagianfranco/+archive/ubuntu/costamagnagianfranco-ppa/+sourcepub/6765471/+listing-archive-extra [13:42] seb128: how do I tell? [13:42] systemctl status doesn't show [13:42] ah, just systemctl [13:43] tjaalton, systemctl list-units --failed [13:46] LocutusOfBorg: looking at curl now, thanks [13:47] thanks to you [13:48] seb128: ok, systemd-modules-load.service, the other two were for nfs mounts [13:49] and no wifi before login [13:49] tjaalton, what does systemctl status -l systemd-modules-load.service says? [13:50] main process exited... failure [13:53] [ 9.723679] mei_me 0000:00:16.0: wait hw ready failed [13:53] could be related [14:01] nope [15:14] LocutusOfBorg: curl uploaded, thanks [15:15] LocutusOfBorg: I'll take a look at the openssl merge in a few days, I need to get the fips patches tested === Guest69044 is now known as ahoneybun [15:45] thanks mdeslaur I admit I'm not 100% sure about it [15:45] but there is an open merge bug where I explained my worries [15:48] anyone else seen errors in sbuild environment due to http://stackoverflow.com/questions/6033599/oserror-38-errno-38-with-multiprocessing ? [15:48] i'm running flake8 (python3 -m flake8) [15:49] and getting a stack trace due to http://paste.ubuntu.com/22042094/ [15:50] oh. i see. [15:50] i had /dev/shm commented out because of bug 1438942 [15:50] bug 1438942 in schroot (Debian) "Host's /dev/shm is mounted over when entering 14.10 and older sbuild schroots" [Unknown,New] https://launchpad.net/bugs/1438942 [16:28] hallyn: around ? wanted to chat about SRU for https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1541902 ; looking at the debdiff. it's a bump to 2.6 (merge from debian) and then the upstream patch [16:28] Error: Could not gather data from Launchpad for bug #1541902 (https://launchpad.net/bugs/1541902). The error has been logged === JanC is now known as Guest94529 === JanC_ is now known as JanC [16:45] rharper: sru into xenial? [16:45] yeah [16:46] that's where they ultimately want it; so we likely need to push it into yakkety first, and then SRU the bump and merge, right ? [16:46] this is the GPU stuff for ppc64 [17:38] I finally have my Chromebook (CYAN) booting yakkety from its internal hard drive. things that don't work are keyboard (need to use external), touchscreen and sound. === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [22:36] something is still wrong with apport tests, the new apt upload brought it from failures=1,errors=8 to failures=1,errrors=1 - but I have exactly no idea why the same code works those 7 tests now, but fails in one... [23:29] Mmh, I should have apport's testsuite fixed now, so APT should go in tomorrow === _salem is now known as salem_