Home
release-team@conference.openafs.org
Friday, August 7, 2020< ^ >
kaduk@jabber.openafs.org/barnowl has set the subject to: openafs release team
Room Configuration
Room Occupants

GMT+0
[11:40:20] meffie joins the room
[12:00:19] mbarbosa joins the room
[15:49:13] kaduk@jabber.openafs.org/barnowl joins the room
[16:00:00] <kaduk@jabber.openafs.org/barnowl> Greetings
[16:00:41] cwills joins the room
[16:01:01] <cwills> Good morning
[16:01:25] yadayada joins the room
[16:01:45] <yadayada> Hi All
[16:02:07] <kaduk@jabber.openafs.org/barnowl> Stephan sent his regrets over email
[16:02:32] <kaduk@jabber.openafs.org/barnowl> And I hardly got anything done this week due to a messed-up sleep
schedule
[16:02:39] <kaduk@jabber.openafs.org/barnowl> So I predict a pretty short meeting
[16:03:13] <meffie> greetings.
[16:03:31] <kaduk@jabber.openafs.org/barnowl> Any 1.8 news?
[16:04:07] <meffie> andrew asked if we could include 14255 in 1.8.
[16:04:35] <meffie> to avoid nul chars in volumes
[16:04:41] <kaduk@jabber.openafs.org/barnowl> Someone at Debian noticed that the 1.8.6 in the tree doesn't build
with moodern gcc that defaults to -fno-common, so I'll be pulling
those patches in
[16:05:04] <meffie> ok, thanks.
[16:05:35] <yadayada> Last week I did not made much progress on Nat Ping issue. Hope to do this in coming week. I did couple of review on salvager exec patches
[16:05:39] <meffie> looks like debian currently ships 1.8.2 (plus some patches?) is that correct?
[16:06:24] <meffie> thanks yadav.
[16:06:29] <kaduk@jabber.openafs.org/barnowl> 1.8.2 in the current stable release, yes, but that will not get
updated ever, basically.
https://packages.debian.org/search?keywords=openafs-client lists the
different releases and what version they have
[16:06:36] <kaduk@jabber.openafs.org/barnowl> Thanks Yadav.
[16:07:11] <kaduk@jabber.openafs.org/barnowl> w.r.t. 14255 in 1.8, that might require a little thought, since it is
perhaps a behavior change (but only in case of error).
[16:07:28] <kaduk@jabber.openafs.org/barnowl> But maybe I'm just not awake enough yet
[16:08:14] <meffie> ok thanks.
[16:09:12] <meffie> i'd say it fixes a silent data corruption, if i understand correctly
[16:09:27] <cwills> The -fno-common should already be in 1.8.x (14217)
[16:10:21] <kaduk@jabber.openafs.org/barnowl> Yeah, but we don't have a 1.8.7pre1 yet
[16:10:34] <cwills> Ah
[16:11:35] <cwills> Linux 5.9 brought a new problem.  Simply fix thankfully.
[16:12:19] <kaduk@jabber.openafs.org/barnowl> I saw that; thanks for the patch.  (I have basically the same reaction
as Andrew.)
[16:12:54] <cwills> but my change did kill support for older (2.9.10 and older) kernels.  If that's a problem, let me know
[16:13:48] <kaduk@jabber.openafs.org/barnowl> 2.9.10 or 2.6.10?
[16:13:56] <cwills> 2.6 -- sorry
[16:14:04] <meffie> whew
[16:14:48] <kaduk@jabber.openafs.org/barnowl> Off the top of my head, I think I'm okay desupporting 2.6.10
[16:15:34] <meffie> anyone running 2.6.10 is probably stuck with openafs 1.4.x ?
[16:15:45] <kaduk@jabber.openafs.org/barnowl> Kind of my thinking, yes
[16:15:55] <cwills> Is there a place out on the openafs.org site that lists what the minimum kernel level are?
[16:15:55] <meffie> sounds about right.
[16:16:27] <meffie> not that i know of.
[16:17:29] <cwills> Anyway.. I'll continue to keep a watch on the 5.9 rcs
[16:17:50] <kaduk@jabber.openafs.org/barnowl> Looks like we put a note in README when we dropped support for Linux
2.4
[16:18:24] <kaduk@jabber.openafs.org/barnowl> I think that implies that we try to support 2.6.0+, but of course in
practice we've been largely ignoring anything pre-git (or at least, I
have).
[16:19:38] <kaduk@jabber.openafs.org/barnowl> (But I don't have a clone of linux.git on this machine and don't
remember exactly which rc that corresponds to)
[16:20:14] <kaduk@jabber.openafs.org/barnowl> But if 2.6.10 is from 2004, I think we are okay not supporting it.
[16:20:29] <cwills> 2.6.12-rc2
[16:20:49] <cwills> April 2005
[16:21:09] <kaduk@jabber.openafs.org/barnowl> Thanks
[16:21:30] <kaduk@jabber.openafs.org/barnowl> Anything else for today?
[16:21:43] <yadayada> nothing from my side
[16:21:55] <meffie> just to mention i changed the test steps on the buildbot to show test details.
[16:22:03] yadayada leaves the room
[16:22:17] <meffie> just on the ubuntu18 builder.
[16:22:31] <kaduk@jabber.openafs.org/barnowl> Should be interesting to see how that works.
[16:22:55] <meffie> that's the only builder currently set up to run `make check`,  it just runs runtest -o for each test.
[16:22:56] yadayada joins the room
[16:23:02] <kaduk@jabber.openafs.org/barnowl> I think my next step for buildbot is to set up the current
buildbot.openafs.org to listen on port 444 (or whatever) and proxy to
the new machine, so we can have a real cert and do a smoother cutover.
[16:23:56] <meffie> i think gerrits 1429{4,5} would be nice, then the buildbot could just run `make check` again.
[16:24:33] <meffie> not sure if we'd ever have those on 1.8 though.
[16:24:53] <kaduk@jabber.openafs.org/barnowl> Since it's just the test code I think it would be okay for 1.8
[16:25:12] <meffie> excellent! and thanks for the buildbot.openafs.org work. let me know if there is anything i can do.
[16:25:28] yadayada leaves the room
[16:25:43] <kaduk@jabber.openafs.org/barnowl> I think making sure that I know whatever is blocking a cutover is the
most important thing.  So far I know that the cert is an issue.
[16:26:42] <meffie> once the cert is in place i'll verify we can do gerrit verifications
[16:26:56] <meffie> it's easier to do with a web ui to see what is going on.
[16:27:06] yadayada joins the room
[16:27:15] <kaduk@jabber.openafs.org/barnowl> Sure
[16:27:58] <kaduk@jabber.openafs.org/barnowl> Motion to adjourn?
[16:28:12] <meffie> second, have a good week end
[16:28:31] <yadayada> Thanks all
[16:28:42] <meffie> yadayada: let us know if you'd like some help with the nat bug
[16:28:54] <meffie> maybe andrew could help if needed?
[16:29:01] <yadayada> sure Mike, next week will get in touch
[16:29:07] <meffie> thanks!
[16:29:16] <kaduk@jabber.openafs.org/barnowl> Thanks everyone; have a good weekend!
[16:29:20] <cwills> have a good week everyone
[16:29:25] meffie leaves the room
[17:15:14] yadayada joins the room
[17:19:59] yadayada leaves the room
[17:20:40] yadayada leaves the room
[17:33:18] mbarbosa leaves the room
[17:34:37] mbarbosa joins the room
[17:39:44] mbarbosa leaves the room
[17:39:51] mbarbosa joins the room
[19:07:01] kaduk@jabber.openafs.org/barnowl leaves the room
[21:56:52] mbarbosa leaves the room
[22:34:25] cwills leaves the room