- Thursday, July 29, 2021

release-team@conference.openafs.org

Thursday, July 29, 2021< ^ >

kaduk@jabber.openafs.org/barnowl has set the subject to: openafs release team

Room Configuration

Room Occupants

GMT+0
[00:02:47] Cheyenne leaves the room
[03:06:08] mvita2 leaves the room
[03:51:18] mvita2 joins the room
[13:08:20] meffie joins the room
[13:48:07] Cheyenne joins the room
[14:03:53] mbarbosa joins the room
[15:43:49] wiesand joins the room
[15:46:06] kaduk@jabber.openafs.org/barnowl joins the room
[15:51:57] <wiesand> To put this up for thought/discussion: The 1.8.8 release has been prepared. Tarballs were uploaded, and the proposed web change is in gerrit 14736. Hence we could release 1.8.8 during the meeting. ALAS, the pre2 announcement I sent last thursday hasn't made it into the public :-(
[15:53:41] <wiesand> Having a rather short fuse today, I'm inclined to release today all the same (and declare the -announce list retired - I think Ben has given up on it a while ago already)
[16:00:24] <meffie> happy thursday
[16:01:06] <wiesand> hello mike
[16:01:13] <Cheyenne> Hello all
[16:01:26] <kaduk@jabber.openafs.org/barnowl> I don't think I've given up on the announce list; I just have some
things that I cross-post to other places as well as -announce.
[16:01:42] <wiesand> ah, ok
[16:06:21] <wiesand> hi cheyenne
[16:06:42] <wiesand> so, should we try again and waste another week?
[16:07:23] <kaduk@jabber.openafs.org/barnowl> I could go either way.
[16:07:38] <kaduk@jabber.openafs.org/barnowl> I'm happy to try to pester the list moderator out of band if we think
that's useful
[16:08:41] <wiesand> I do that every time I send anything to -announce. This time it just didn't help.
[16:08:49] <kaduk@jabber.openafs.org/barnowl> Ah.
[16:08:56] <wiesand> No response whatsoever.
[16:10:48] <Cheyenne> I'm fine either way as well.
[16:11:02] <wiesand> Let's vote. Please respond with either "now" or "later".
[16:11:09] <meffie> now
[16:11:33] <kaduk@jabber.openafs.org/barnowl> [checking on one thing first]
[16:12:13] <Cheyenne> now is fine
[16:12:33] <kaduk@jabber.openafs.org/barnowl> now is fine
[16:13:32] <wiesand> Let's go then. Ben, please tag commit 8c307b42f0b0b80bcf69de09031991d2ab776ed9 as openafs-stable-1_8_8
[16:14:00] <Cheyenne> (yay)
[16:15:01] <kaduk@jabber.openafs.org/barnowl> * [new tag]             openafs-stable-1_8_8 -> openafs-stable-1_8_8
[16:17:44] <kaduk@jabber.openafs.org/barnowl> Does that take care of the 1.8 part of the agenda, or are we planning
1.8.9 already?
[16:20:25] <Cheyenne> clean intel builds for linux-5.14-rc3+ for 1.8.x and master
[16:22:37] <wiesand> 1.8.9 already has quite a wish list - let me work on that for the week
[16:23:57] <kaduk@jabber.openafs.org/barnowl> ok
[16:24:33] <meffie> thanks wiesand.
[16:24:51] <kaduk@jabber.openafs.org/barnowl> As feared/promised, I was preoccupied with IETF week and don't have
any real progress to report.
I did get to merge Yadav's afs_remulink() change that helps with the
Crowdstrike Falcon issue reported on the list.
[16:24:55] <wiesand> 1.8.8 release should be complete - i'll send the announcement while you discuss 1.9.1/master stuff
[16:25:46] <kaduk@jabber.openafs.org/barnowl> So I don't think I have much to add other than the usual "help me
prioritize what to look at" topic
[16:26:41] <meffie> there are only 500 open gerrits for the master branch :)
[16:26:57] <kaduk@jabber.openafs.org/barnowl> Could be worse
[16:27:07] <meffie> could be stabbed ?
[16:27:27] <meffie> (life of brian movie reference.)
[16:27:49] <kaduk@jabber.openafs.org/barnowl> OpenSSL has 1.2k open issues in github.
And yes, also that.
[16:28:11] <meffie> oh. that is worse.
[16:29:21] <kaduk@jabber.openafs.org/barnowl> Only 194 open pull requests for openssl, though, which is not the
breakdown I was remembering.
[16:29:44] <meffie> it might be good to get cheyenne's static analyzer fixes in. since, that touches a lot of different places.
[16:31:20] <kaduk@jabber.openafs.org/barnowl> That's a good point.  And those should be fairly localized in terms of
what needs to be reviewed, so I wouldn't need a large contiguous chunk
of time to understand a large corpus of work.
[16:32:00] <meffie> yes, each one is independent, so they dont have to be merged in a stack or any certain order.
[16:32:14] <meffie> or all at once.
[16:32:49] <meffie> i was planning to review more of them today in fact.
[16:32:51] <Cheyenne> that's correct.  The splitup that I did makes each commit fairly small.  They are by directory
[16:33:44] <Cheyenne> there is just a lot of them :(
[16:34:16] <kaduk@jabber.openafs.org/barnowl> It turns out that coding practices 30 years ago were ... not good.
And automated tooling has gotten pretty good at helping find common
issues.
[16:36:59] <meffie> and it looks like stuff like remove kdump is ready
[16:39:09] <kaduk@jabber.openafs.org/barnowl> Oh yeah, thanks for the reminder about that; long overdue
[16:39:13] <Cheyenne> there is the enhancements to the audit facility
[16:41:17] <meffie> i pushed 'vos-restore-check-dump-before-delete', but andrew has not had a chance to look at it yet.
[16:43:33] <kaduk@jabber.openafs.org/barnowl> Interesting.  This sounds like the sort of thing developed after an
unfortunate incident in production...
[16:44:20] <meffie> yes
[16:45:49] <meffie> i imagine vos restore evolved over time, and the code to check the dump stream predates the volume deletion in UV_RestoreVolume*
[16:46:16] <wiesand> Ben: RT #135254 probably means I should hold my horses merging anything on 1.8.x?
[16:46:34] <kaduk@jabber.openafs.org/barnowl> At least for this week, while we get a sense of scope
[16:46:44] <wiesand> ok
[16:48:48] <kaduk@jabber.openafs.org/barnowl> I think we may be out of topics for today.  Anything else to cover?
[16:49:32] <meffie> oh, questions about the gerrit web service
[16:49:41] <kaduk@jabber.openafs.org/barnowl> ok
[16:50:12] <meffie> last friday a lot of the buildbot builds failed randomly during the git stage.
[16:50:20] <wiesand> I could ask one more time that issues not all that secret be handled like any other change, but I guess it wouldn't help...
[16:51:13] <kaduk@jabber.openafs.org/barnowl> I think we will be able to do that for some things
[16:51:27] <meffie> i restarted the buildbot coordinator, but it seems it was an intermittent gerrit issue?
[16:51:29] <kaduk@jabber.openafs.org/barnowl> Mike: yes, it seems that's been happening with some regularity
[16:51:39] <kaduk@jabber.openafs.org/barnowl> I'm pretty sure it's on the gerrit side, yes
[16:52:28] <meffie> ok, i noticed it after making a small change on the buildbot, and then backed it out, fearing i broke something.
[16:52:59] <meffie> and i was in a car at the time, (my wife was driving)
[16:53:04] <kaduk@jabber.openafs.org/barnowl> restarting gerrit usually helps for a few days, so there may be a
resource leak or under-resourcing or something
[16:53:39] <meffie> ah, ok, sorry.
[16:53:39] <kaduk@jabber.openafs.org/barnowl> moving the gerrit service over to the new machine would improve the
resourcing situation and allow us to take a newer version of gerrit as
well.
[16:54:03] <meffie> is that something cheyenne or i can help with?
[16:54:07] <kaduk@jabber.openafs.org/barnowl> But it has proven challenging for me to schedule some time to actually
work on that part
[16:54:30] <kaduk@jabber.openafs.org/barnowl> Hmm, possibly.
[16:54:53] <meffie> cheyenne has been setting up local gerrits just to let us do buildbot testing.
[16:55:00] <Cheyenne> I've been working with setting up a gerrit for some testing, so it's something I'm getting familiar with
[16:55:03] <kaduk@jabber.openafs.org/barnowl> Right now the new system is a somewhat base setup of RHEL7, and I
haven't even tried to look for the list of dependencies.
[16:55:39] <kaduk@jabber.openafs.org/barnowl> Hopefully everything is available in EL, or at least EPEL.  I remember
for setting it up on RHEL6 there were a couple things that we got Ken
to give us patched srpms for and I rebuilt them locally, which was
kind of exciting.
[16:56:07] <meffie> heh
[16:56:25] <kaduk@jabber.openafs.org/barnowl> But IIRC, once the dependencies are installed, actually cutting over
is fairly straightforward, just stopping the old one, taking a db
dump, copying that over, and `java gerrit.war -init` or so
[16:57:38] <Cheyenne> it's something I can try .. I have a gerrit that I think is close to the same level.  I'm just not sure what some of the behind the scenes configurations are
[16:57:41] <kaduk@jabber.openafs.org/barnowl> So I guess the action here would be to get a list of steps (installing
dependencies?) to go from a stock RHEL7 system to running gerrit
[16:57:59] <kaduk@jabber.openafs.org/barnowl> I could probably pull some of the config files out and send them over
[16:58:09] <Cheyenne> that would be very helpful
[16:58:18] <meffie> thanks.
[16:58:30] <kaduk@jabber.openafs.org/barnowl> thank you for offering to help!
[16:58:43] <Cheyenne> (I've been using debian, but I will switch over to a centos7 system)
[16:59:11] <meffie> ok, that's all i have. thanks.
[16:59:20] <meffie> (off to the next meeting)
[16:59:21] <wiesand> beats me that nobody is considering EL8
[16:59:42] <kaduk@jabber.openafs.org/barnowl> Yeah, I tried to push for EL8 but didn't get much traction
[17:00:48] <kaduk@jabber.openafs.org/barnowl> Thanks Mike/Cheyenne
[17:00:57] <kaduk@jabber.openafs.org/barnowl> And I guess it's time for over to the next meeting, yes.
[17:00:58] <Cheyenne> Thank you
[17:01:07] <kaduk@jabber.openafs.org/barnowl> So thanks everyone :)
and have a good week
[17:01:13] <wiesand> Thanks a lot everyone!
[17:01:18] <Cheyenne> Have a safe week everyone
[17:01:24] <wiesand> Off to watering the garden…
[17:01:30] <Cheyenne> +1
[17:18:45] meffie leaves the room
[20:50:05] wiesand leaves the room
[21:30:04] mbarbosa leaves the room