Home
release-team@conference.openafs.org
Friday, September 28, 2018< ^ >
Room Configuration
Room Occupants

GMT+0
[01:53:29] mvita joins the room
[02:09:46] mvita leaves the room
[06:30:00] mvita joins the room
[07:14:55] mvita leaves the room
[12:51:07] wiesand joins the room
[12:51:52] meffie joins the room
[13:00:34] <wiesand> good morning
[13:00:39] mvita joins the room
[13:00:52] <mvita> hi
[13:01:05] <kaduk@jabber.openafs.org/barnowl> greetings
[13:01:21] <mvita> I don't expect Mike will join us today
[13:01:43] <kaduk@jabber.openafs.org/barnowl> He wasn't in the room's membership list when I looked an hour ago, but
is now...
[13:01:47] <wiesand> i've been fully occupied by moving home last week again :-(
[13:02:11] <meffie> i am here, but lurking...
[13:02:24] <mvita> lurkers gotta lurk
[13:02:25] <kaduk@jabber.openafs.org/barnowl> I was pretty occupied by IESG telechat prep last week, and just got
some gerrit reviews done this morning.
[13:02:49] <kaduk@jabber.openafs.org/barnowl> I was hoping to ask Mike if there were buildbot workers that I need to
pester the owners about, since we don't seem to be getting verified
+1s
[13:03:39] <meffie> oh?
[13:04:30] <kaduk@jabber.openafs.org/barnowl> 12746/etc. don't have buildbot feedback
[13:04:44] <wiesand> thanks for 13330 - i guess we'll have a 1.6.24pre1 "soon" now
[13:04:50] <kaduk@jabber.openafs.org/barnowl> A lot of the stuff that is last updated in the past week or so, too
[13:05:36] <meffie> i'll take a look ben.
[13:05:42] <kaduk@jabber.openafs.org/barnowl> Thanks Mike!
[13:06:20] <mvita> Mike, you need to cross-train me on buildbot stuff
[13:06:35] <mvita> so you can enjoy your vacations from time to time
[13:08:13] <mvita> I meant to write a patch for the PPC crash yesterday but other events intervened
[13:08:16] <meffie> you covered the important stuff yesterday!
[13:08:27] <wiesand> looks like opensuse12-x86_64-builder is sick
[13:08:40] <mvita> one of our sites had a P1 outage
[13:08:48] <mvita> fortunately NOT an AFS bug
[13:08:53] <kaduk@jabber.openafs.org/barnowl> (Not Cisco Webex, I trust)
[13:09:35] <mvita> it was hardware related
[13:09:44] <kaduk@jabber.openafs.org/barnowl> *nods*
[13:09:55] <kaduk@jabber.openafs.org/barnowl> (The timeline at https://status.ciscospark.com/incidents/6gzfst6ls3n5
is kind of scary long)
[13:10:09] <mvita> but then we were in Salvager Hell for a while
[13:10:44] <kaduk@jabber.openafs.org/barnowl> Fun times
[13:11:25] <kaduk@jabber.openafs.org/barnowl> The ppc reporter also sent a patch for ppc64le_linux; not sure how
much demand we'd have for such a target.
[13:11:26] <wiesand> should we reintroduce fast-restart?
[13:11:32] <wiesand> (just kidding)
[13:12:37] <meffie> "death to the salvager"
[13:12:37] <mvita> oh, wow, webex
[13:12:44] <mvita> I didn't know about that
[13:12:48] <mvita> we were using Zoom
[13:14:06] <wiesand> "Due the limited resource and clear preference of the ppc64le architecture in any new developments, it was decided that ppc64 architecture will transfer into a maintenance mode where the new features (eg. Atomic Host compose) we plan for ppc64le won't be added, see email <https://lists.fedoraproject.org/archives/list/ppc@lists.fedoraproject.org/message/C23EQYITA4DQWM7CQF6LJC5ABXY2XIEM/> for details."
[13:14:19] <wiesand> from https://fedoraproject.org/wiki/Architectures/PowerPC
[13:14:37] <wiesand> so it could be worthwhile
[13:14:52] <wiesand> and the patch in TR doesn't look too scary
[13:14:57] <wiesand> RT
[13:15:24] <mvita> looking for the patch, I haven't seen it yet
[13:15:33] <wiesand> and part of it is a build fix we want to issue anyway
[13:15:36] <kaduk@jabber.openafs.org/barnowl> Yeah, the main question on it is probably just to make sure that the
syname allocations are sane, to the extent that any of ours are sane.
[13:15:46] <mvita> oh, it's in RT
[13:15:50] <mvita> sorry
[13:16:08] <kaduk@jabber.openafs.org/barnowl> https://rt.central.org/rt/Ticket/Display.html?id=134674
[13:16:47] <mvita> Oh, sysname thing
[13:17:03] <mvita> I was talking about the issue w/ rx_atomic
[13:17:10] <mvita> also on PPC
[13:17:48] <kaduk@jabber.openafs.org/barnowl> Yeah, we still have to fix the rx_atomic thing, but could also add the
arch that doesn't have endianness sadness
[13:18:06] <mvita> nod
[13:20:00] <wiesand> who decides what a sane sysname is?
[13:20:09] <meffie> (opensuse12 disabled for gerrits builds for the time being)
[13:20:25] <meffie> (i'll ping christof)
[13:20:45] <kaduk@jabber.openafs.org/barnowl> I was thinking more about the numbers, which are nominally allocated
by the GCO registrar but in  practice we just squat on whatever we
feel like.
[13:21:09] <kaduk@jabber.openafs.org/barnowl> But sometimes we keep a block of 100 for different versions of a
specific OS+arch, and sometimes we mix things up within a block of
numbers, etc.
[13:22:12] <wiesand> the numeric value is defined by the implementation?
[13:22:35] <kaduk@jabber.openafs.org/barnowl> apparently
[13:23:24] <wiesand> speaking of the build fix: the only actual topic i have today is about schedules
[13:23:38] <wiesand> 1.6.24pre1 soon, obviously
[13:23:55] <wiesand> but how to get out the fixes? 1.8.2.1?
[13:24:39] <wiesand> or go for a 1.8.3 with more churn?
[13:25:30] <kaduk@jabber.openafs.org/barnowl> Which fixes would we be thinking for 1.8.2.1 -- the TLog() build fix,
and maybe this ppc rx_Atomic thing that's not in gerrit yet?
[13:26:27] <wiesand> probably, though I'm not sure whether there wasn't another one mentioned in a recent meeting
[13:26:53] <mvita> yes, there was something else… can't remember
[13:27:01] <meffie> GCTrans?
[13:27:10] <mvita> that's it
[13:27:16] <mvita> I did the pullups for 1.8.x
[13:27:22] <wiesand> ah, right
[13:27:29] <meffie> busy volserver crashes
[13:27:47] <kaduk@jabber.openafs.org/barnowl> GCTrans would I think be a 1.8.3 with our normal versioning scheme
[13:28:01] <wiesand> right
[13:28:02] <meffie> i agree
[13:28:03] <mvita> yes
[13:28:17] <wiesand> so, go for a "lean" 1.8.3?
[13:28:53] <kaduk@jabber.openafs.org/barnowl> Somewhat lean, at least; there are some other pull-up-able commits on
master
[13:29:30] <wiesand> there are also still some remainders of the ubik stack
[13:29:51] <mvita> CTF stuff for 1.8.2.1, maybe?
[13:30:21] <kaduk@jabber.openafs.org/barnowl> CTF could be 1.8.2.1, I think
[13:30:35] <meffie> hmm, i feel that would be 1.8.3
[13:31:10] <meffie> i feel 1.8.2.1 would be for things like macos support for the new version
[13:31:13] <mvita> it is platform support, kind of
[13:31:18] <wiesand> CTF is about the solaris client?
[13:31:20] <kaduk@jabber.openafs.org/barnowl> 1.8.3 is not wrong for CTF, of course
[13:31:38] <mvita> CTF has a few commits
[13:31:46] <mvita> some for client were already merged
[13:31:53] <mvita> I think the current set is for the servers
[13:32:09] <mvita> (iirc)
[13:32:19] <meffie> changes were made to support both
[13:32:27] <wiesand> i keep forgetting that there are platforms requiring the kernel module on servers
[13:32:49] <mvita> I don't think that's true any longer….
[13:33:13] <kaduk@jabber.openafs.org/barnowl> <tiny>NFS translator</tiny>
[13:33:43] <meffie> inode fileservers? i dont think there are any left?
[13:33:44] <mvita> <deprecated to death>
[13:33:56] <mvita> also deprecated to death
[13:34:22] <mvita> we should rip out both features and see who squeals
[13:34:49] <kaduk@jabber.openafs.org/barnowl> I doubt anyone using those would be running master anyway
[13:34:52] <mvita> I think the silence would be deafening
[13:35:20] <kaduk@jabber.openafs.org/barnowl> But yeah, I'd merge removals of those to master
[13:35:30] <mvita> oooh
[13:35:37] <mvita> so masterful!
[13:36:29] <mvita> anyway, back to CTF - the client support was in 1.8.0
[13:36:39] <mvita> we are just discussing server support
[13:36:57] <mvita> the issue of servers needing clients sometimes is tangential
[13:36:58] <wiesand> ok, then it should be 1.8.3
[13:37:11] <mvita> okay
[13:38:10] <mvita> I just thought of another one - butc is documented as "requiring" a client (for obtaining tokens) but of course it can use -localauth
[13:39:17] <kaduk@jabber.openafs.org/barnowl> yup
[13:39:59] <wiesand> so, seems like we go for a 1.8.3 with most of what's on https://gerrit.openafs.org/#/q/status:open+project:openafs+branch:openafs-stable-1_8_x today
[13:40:14] <wiesand> (which includes gctrans and the build fixes)
[13:40:25] <kaduk@jabber.openafs.org/barnowl> sounds like a plan
[13:40:45] <wiesand> + the rx_atmoic fix discussed today
[13:41:10] <mvita> I will have a PPC patch for review either today or early next week
[13:41:17] <meffie> we should be in a better position for gerrits next week
[13:41:30] <wiesand> good
[13:41:51] <meffie> mark, andrew, marcio, and i will be in the same room.
[13:42:06] <meffie> and spend some time on the backlog
[13:42:16] <mvita> we're getting the band back together
[13:42:18] <wiesand> sounds promising
[13:42:35] <kaduk@jabber.openafs.org/barnowl> We're getting the band back together!!!
[13:42:36] <wiesand> i should find a bit more time for openafs next week too
[13:42:43] <mvita> !
[13:42:59] <meffie> yay
[13:43:00] <mvita> "We're on a mission from God"
[13:43:25] <kaduk@jabber.openafs.org/barnowl> Such a good movie
[13:43:38] <wiesand> where's the cadillac?!
[13:43:45] <mvita> heh
[13:44:58] <wiesand> looks like we have a plan for the near future
[13:45:08] <wiesand> ben: more on master?
[13:45:43] <kaduk@jabber.openafs.org/barnowl> not much noteworthy that I remember, other than buildbot woes.  I
guess, and asking if I should rebase the xstat_pthread changes or
someone else wanted to
[13:45:54] <meffie> i did
[13:46:16] <meffie> i thought i did....
[13:46:29] <meffie> i pushed new ones. do they need rebased?
[13:46:32] <kaduk@jabber.openafs.org/barnowl> cool
but let me look at gerrit once more just in case I forget something
[13:46:45] <kaduk@jabber.openafs.org/barnowl> Something to cause buildbot to look at them, at least
[13:47:25] <kaduk@jabber.openafs.org/barnowl> Oh, yeah, I commented on 13085 this morning, one of Pat's changes.
[13:47:30] <meffie> ok. i'd be happy to rebase again if needed
[13:47:39] <kaduk@jabber.openafs.org/barnowl> It would be good to get another opinion on a couple things there
[13:48:13] <kaduk@jabber.openafs.org/barnowl> But the gerrit search results for master are not triggering anything
else for me
[13:48:18] <kaduk@jabber.openafs.org/barnowl> So maybe we should call it?
[13:49:08] <meffie> ok, thanks, have a good weekend.
[13:49:15] <kaduk@jabber.openafs.org/barnowl> You too!
[13:49:55] <wiesand> thanks everyone!
[13:50:00] <mvita> bye
[13:50:06] wiesand leaves the room
[13:50:08] meffie leaves the room
[13:50:21] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[14:22:40] mvita leaves the room
[15:03:03] mvita joins the room
[16:53:01] mvita leaves the room
[17:01:41] mvita joins the room
[17:40:32] mvita leaves the room
[18:10:36] mvita joins the room
[21:47:01] mvita leaves the room
[22:12:08] mbarbosa joins the room
[22:19:40] mbarbosa leaves the room
[22:21:26] mbarbosa joins the room
[23:44:58] mbarbosa leaves the room