Wednesday, 2019-04-10

*** tjgresha has quit IRC00:00
*** lbragstad has quit IRC00:45
*** whoami-rajat has joined #openstack-tc01:12
*** mriedem has quit IRC01:18
gmannTC office hour time though01:18
*** jamesmcarthur has joined #openstack-tc01:20
*** jamesmcarthur has quit IRC01:23
fungia quiet one, once more01:49
*** adriant has joined #openstack-tc02:38
*** ricolin has joined #openstack-tc03:31
*** diablo_rojo has quit IRC04:29
*** openstackstatus has quit IRC04:35
*** openstackstatus has joined #openstack-tc04:36
*** ChanServ sets mode: +v openstackstatus04:36
*** Luzi has joined #openstack-tc05:17
*** ricolin has quit IRC05:30
*** e0ne has joined #openstack-tc06:28
*** e0ne has quit IRC06:48
*** e0ne has joined #openstack-tc06:57
*** tosky has joined #openstack-tc07:28
*** e0ne has quit IRC07:33
*** e0ne has joined #openstack-tc07:40
*** jpich has joined #openstack-tc07:48
*** e0ne has quit IRC07:48
*** e0ne has joined #openstack-tc07:53
*** evrardjp has quit IRC08:18
*** evrardjp has joined #openstack-tc08:19
*** Luzi has quit IRC08:21
*** ricolin has joined #openstack-tc08:30
*** e0ne has quit IRC08:47
*** dtantsur|afk is now known as dtantsur08:58
*** e0ne has joined #openstack-tc08:58
*** e0ne has quit IRC09:06
*** e0ne has joined #openstack-tc11:12
*** e0ne has quit IRC11:15
*** e0ne has joined #openstack-tc11:19
*** e0ne has quit IRC11:20
*** e0ne has joined #openstack-tc11:24
*** sapd1 has quit IRC11:29
*** sapd1 has joined #openstack-tc11:29
asettleAfternoon o/11:31
smcginnisMorning o/11:35
evrardjpall those timezones!11:49
smcginnishttps://www.youtube.com/watch?v=PWcZKrP2sDU11:52
*** sapd1 has quit IRC12:00
*** sapd1 has joined #openstack-tc12:02
ricolino/12:05
*** e0ne has quit IRC12:09
*** e0ne has joined #openstack-tc12:12
*** e0ne has quit IRC12:13
*** ijolliffe has joined #openstack-tc12:25
*** e0ne has joined #openstack-tc12:28
*** ricolin has quit IRC12:50
*** ricolin has joined #openstack-tc12:51
*** lbragstad has joined #openstack-tc12:58
*** mriedem has joined #openstack-tc13:02
*** irclogbot_2 has joined #openstack-tc13:04
*** altlogbot_3 has joined #openstack-tc13:07
dtantsurjust use UGT: http://www.total-knowledge.com/~ilya/mips/ugt.html13:16
*** dklyle has joined #openstack-tc13:17
*** e0ne has quit IRC13:20
*** e0ne has joined #openstack-tc13:25
*** e0ne has quit IRC13:54
*** e0ne has joined #openstack-tc13:57
*** jrosser has joined #openstack-tc14:19
mnaserdoes anyone know who manages/maintains cirrus?14:21
mnasercirros*14:21
mnasergiven that it's a bit of a fundamental tool in our CI, I've had contributors working on trying to get it working under aarch64 (and have OSA deploy on that as a target)14:21
dtroyerdoes smoser (Scott Moser) still own that?14:21
mnaserbut they've been blocked for a long time due to a bug https://bugs.launchpad.net/cirros/+bug/181331914:22
openstackLaunchpad bug 1813319 in CirrOS "Arm64 image requires a kernel and grub update to work" [Undecided,Confirmed]14:22
mnaserdtroyer: according to launchpad, yes14:22
dtroyerThere is an OpenWRT image in Nova's git history, too bad that isn't still useful :)  It runs on many architectures14:25
dtroyer(not that specific image)14:25
*** sapd1 has quit IRC14:27
*** sapd1 has joined #openstack-tc14:27
mugsiemnaser: is there an issue using the image jrosser produced?14:28
mugsieor, potentially, taking the mods he did, and pushing it to somewhere like tarballs.o.o ?14:28
jrosserthere was a fairly reasonable -1 on that as it was hosted somewhere randomly14:28
mnasermugsie: nope, that one works well and it was built with help of code think which are pretty much arm experts afaik.  I like the idea of making it reproducible and pushing it to tarballs.o.o14:29
mnaserperhaps we can pick up cirros into opendev and use zuul to test/build new images across different architectures..?14:29
mugsie(just thinking of a way to unblock you, I would love to see a new arch supported :) )14:29
mnaserthat's a lot of work im not sure who would pick up though :P14:29
mugsietrue14:29
mnaserI'm more trying to view this from a wider POV14:29
mnasersay if: something major was broken in cirros overnight and all our CI is blocked forever :)14:30
mugsieyeah. looks like the  last merge proposal that was actually merged was in 2017?14:30
dtroyerthat has happened (measured in hours IIRC) and there are quite a number of things that have that potential for us…14:31
jrosserihmo the path to contributing a patch is quite opaque - and the codethink guys seem to be being funded to move aarch64 forward generally14:31
mugsiehttps://code.launchpad.net/cirros/+merges14:31
mugsieoh, looks like frickler may have merge rights?14:32
mnaserjrosser: perhaps we can try to push a merge proposal there?14:34
jrossermnaser: sure, it appears that most of the work has been done already on fixing things14:35
mnaserjrosser: do we really need to release 0.4.1 for example or is there changes to be done?14:36
jrossermnaser: i don't know - i'd have to defer to the guys at codethink, perhaps we should engage a bit more directly with them and get a patch published, decisions on versioning can follow that i guess14:37
mnaserI think getting a merge request might be a good next step14:40
*** nowster has joined #openstack-tc14:41
nowsterhi14:41
mnasernowster is from codethink and seems to mention that cirros didn't build at all under aarch64 -- nowster, is there any code changes that we could do to make it build-able ?14:42
mnaserI'm thinking we can push https://code.launchpad.net/cirros/+merges14:42
nowsterI spent about a day on trying to get it to build... deeper and deeper into the rabbithole of Buildroot stuff.14:42
mnaserso it looks like inherently cirros hasn't really thought of any arches I assume14:43
nowstermy memory is hazy on it now. I did work back in December.14:44
mugsiethey claim to have aarch64 for 0.4.0 - http://download.cirros-cloud.net/0.4.0/14:44
nowsterThat has too old a kernel/grub-efi combination to work with QEMU-KVM14:44
mtreinishmnaser: it also supports ppc14:45
mnasermtreinish: oh I see14:45
nowsterWhat happens is that the kernel doesn't find the device tree that's passed via UEFI14:45
mnaserI wonder if that image was ever tested then, I guess probably not14:46
mtreinishnot that anyone really cares about it, I just remember some of the work years ago getting it to work there14:46
mtreinishthe ppc one is used in the power CIs iirc14:47
nowsterI was only able to diagnose the problem by manually forcing the console port registers via grub command line. Otherwise, no console output with that image.14:49
*** e0ne has quit IRC14:50
clarkbwe pin the cirros version we use. It shouldnt unexpectedly break one day as long as qemu/kvm maintain rough compatibility14:52
clarkband if we want to come up with an alternative I think it would be worth considering a dib built image14:53
*** e0ne has joined #openstack-tc14:54
fricklerclarkb: yeah, there have been thoughts about building something suited for e.g. neutron testing some years ago already. so if there's another blocker now, looking into doing something with dib would seem the best approach to me, too14:56
*** ianychoi has quit IRC14:56
fricklerI don't actively do anything with cirros atm14:56
fricklerexcept consuming the released images of course ;)14:56
mugsieah, I saw your name in a commit, with no related mrege request, and assumed :)14:57
fricklermugsie: maybe I still have merge rights, but building and publishing a release isn't automated, so that would be smoser's task anyway14:58
fricklerfolks interested in that should join us in #cirros and talk to him directly14:58
mugsieopenstack and the release team has spoiled me :)14:58
frickleryeah, our release automation should really be praised. in particular given that it just seems to have been running extremely smoothly15:01
nowster(I'm going to retry a build now and report back.)15:01
nowster(However, it builds the world, and may take some time.)15:01
*** diablo_rojo has joined #openstack-tc15:02
mnasercould we maybe built some super stripped down version of Debian using dib?15:04
clarkbmnaser: yes, however super stripped down debain isn't very small :/15:06
mnaserclarkb: yeah, I can imagine that cirros is very tiny and it's pretty good at what it does15:07
clarkbjust the kernel alone is bigger than a cirros image iirc15:07
clarkbbut dib is a flexible tool that we can build something around if necessary15:07
*** sapd1 has quit IRC15:09
*** sapd1 has joined #openstack-tc15:09
*** openstackgerrit has joined #openstack-tc15:12
openstackgerritMerged openstack/governance master: Add a link for Zun technical vision reflection  https://review.openstack.org/64888115:12
ttxICYMI we have a  release !15:27
smcginnisAll aboard the train!15:28
ttxsmcginnis: the name that keeps on giving15:28
smcginnis:)15:29
*** jamesmcarthur has joined #openstack-tc15:32
mnaserfeels like just a few weeks ago we were figuring out a name..15:35
mnasertime goes by fast15:35
nowstercirros buildroot failed whilst building ncurses.  Looks like a similar problem to the one fixed by f33bf60d8fa05fba2c08ed0047eae93891f6663515:37
nowster(Build environment is Ubuntu 18.04.2 LTS)15:40
mnasernowster: so looks like we can't even build cirros at all?15:46
nowsterit probably needs to be updated for gcc 715:48
nowsterhttps://stackoverflow.com/questions/37475222/ncurses-6-0-compilation-error-error-expected-before-int15:49
nowsterThe build continues now.15:50
nowsterand now it's dying somewhere else in automake/autoreconf15:52
nowster...whilst autoconfiguring util-linux 2.2715:54
mnaserit sounds like cirros needs work to at least _build_ :\15:55
nowsterThat was the conclusion I came to back in December.15:55
*** jamesmcarthur has quit IRC16:06
*** jamesmcarthur has joined #openstack-tc16:08
*** dtantsur is now known as dtantsur|afk16:15
*** sapd1 has quit IRC16:32
*** sapd1 has joined #openstack-tc16:32
*** altlogbot_3 has quit IRC16:46
*** e0ne has quit IRC16:55
*** jpich has quit IRC17:00
*** sapd1 has quit IRC17:29
*** sapd1 has joined #openstack-tc17:30
*** jamesmcarthur has quit IRC17:42
*** ricolin has quit IRC17:49
*** tjgresha has joined #openstack-tc17:59
*** sapd1 has quit IRC18:00
*** sapd1 has joined #openstack-tc18:00
*** tjgresha has quit IRC18:09
*** jamesmcarthur has joined #openstack-tc18:12
*** tosky has quit IRC18:21
*** ijolliffe has quit IRC18:29
-openstackstatus- NOTICE: Restarting Gerrit on review.openstack.org to pick up new configuration for the replication plugin19:06
*** jamesmcarthur has quit IRC19:59
*** sapd1 has quit IRC20:10
*** sapd1 has joined #openstack-tc20:16
*** sapd1 has quit IRC20:34
*** sapd1 has joined #openstack-tc20:37
fungitc-members: please see and follow up on jroll's awesome message about the git namespace plan: http://lists.openstack.org/pipermail/openstack-discuss/2019-April/004920.html20:46
fungiand forgive his nonsequential footnote reference indices ;)20:47
jrollwat!20:49
jrollah dang20:50
fungi;)20:50
jrollI blame you for the suggestion to move that line20:51
fungii'm used to being blamed. doesn't faze me20:51
jroll:)20:52
mtreinishsigh, the day the git:// protocol dies is the day I have to update remotes in a gazillion git repos on multiple machines21:09
*** sapd1 has quit IRC21:14
*** sapd1 has joined #openstack-tc21:19
*** ijolliffe has joined #openstack-tc21:19
*** mriedem has quit IRC21:40
*** sapd1 has quit IRC21:50
*** sapd1 has joined #openstack-tc21:54
notmynameto whom it may concern, as swift PTL, timburke (tim@swiftstack.com) needs access to https://review.openstack.org/#/admin/groups/542,members22:05
*** ijolliffe has quit IRC22:06
notmyname(honestly, I'm not sure where to go with that request, but as it's a change in PTL, I figured I'd start here. seems like governance stuff)22:07
timburkei was torn between here and #openstack-stable, myself :-)22:08
notmynameoh, I was thinking -release ;-)22:08
gmannyeah stable core maintain by stable team. tonyb or matt can guide.22:09
notmynameok, I'll ask there22:09
notmynamegmann: thanks22:10
clarkbin general the way to know who to ask is to check https://review.openstack.org/#/admin/groups/542,info and lookup the owner listed there22:11
*** tjgresha has joined #openstack-tc22:15
*** tosky has joined #openstack-tc22:26
*** sapd1 has quit IRC22:41
*** sapd1 has joined #openstack-tc22:42
*** tosky has quit IRC22:47
*** whoami-rajat has quit IRC22:51
*** sapd1 has quit IRC22:55
*** sapd1 has joined #openstack-tc22:58
*** sapd1 has quit IRC23:23
*** sapd1 has joined #openstack-tc23:24
*** david-lyle has joined #openstack-tc23:36
*** dklyle has quit IRC23:36
tonybYup openstack-stable but I'm here so I'll look at it anyway.23:37
*** david-lyle has quit IRC23:46
*** ianychoi has joined #openstack-tc23:46
*** sapd1 has quit IRC23:48
*** sapd1 has joined #openstack-tc23:50

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!