- Friday, August 16, 2019

release-team@conference.openafs.org

Friday, August 16, 2019< ^ >

Room Configuration

Room Occupants

GMT+0
[12:04:40] mbarbosa joins the room
[14:45:12] meffie joins the room
[14:56:50] <meffie> good day.
[14:59:23] wiesand joins the room
[14:59:28] <kaduk@jabber.openafs.org/barnowl> greetings
[14:59:32] <wiesand> Hi
[15:00:17] <mvita> present
[15:01:12] <wiesand> Sigh - kept shifting the 1.8.4pre1 rollout day after day, then today's time slot for it was stolen again. Sorry. "Will do it a.s.a.p." is all I can say.
[15:01:39] <kaduk@jabber.openafs.org/barnowl> I think I am going to get around to debianizing it today.  "probably."
[15:01:55] <meffie> ok, thanks wiesand and kaduk
[15:02:37] <meffie> cheyenne has build rhel7 (and 6 i think) rpms based on the git tag, including kmods
[15:04:33] <wiesand> over the flurry of activity I lost track of the Linux 5.3 changes - but they seem stuck in some way?
[15:05:19] <meffie> i think all the changes needed to build 5.3 are in gerrit now.
[15:06:02] <wiesand> but not merged on master yet, so nightly builds still all fail
[15:06:14] <kaduk@jabber.openafs.org/barnowl> I think I didn't push the 'merge' button on the first one until I
looked at the second one, and then I had a question on the second one.
[15:06:46] yadayada joins the room
[15:06:58] <yadayada> Hello All, sorry for joining late
[15:07:07] <meffie> greetings yadav
[15:07:24] <wiesand> hello
[15:08:41] <wiesand> Looks like the answer to the question on the second one was insufficient?
[15:08:53] <kaduk@jabber.openafs.org/barnowl> greetings.
Did we tell you about the chat room archives already?
[15:09:15] <kaduk@jabber.openafs.org/barnowl> I'm not sure if it was insufficient or I just failed to parse it
properly.
[15:10:01] <wiesand> https://conference.openafs.org/release-team@conference.openafs.org/2019-08-16.html
[15:10:13] <kaduk@jabber.openafs.org/barnowl> But I think I get it now, and will proceed to fix the comment nit...
[15:10:45] <wiesand> "changing the date to match other meetings is left as an excercise to the reader"
[15:11:35] <kaduk@jabber.openafs.org/barnowl> There's even an index in the parent directory (takes a long time to
load, though).  For some reason the docroot 403's, and I never was
able to figure out why.
[15:14:28] <kaduk@jabber.openafs.org/barnowl> I can merge the first linux-5.3 commit now, at least.
[15:15:12] <kaduk@jabber.openafs.org/barnowl> maybe.  If gerrit cooperates
[15:15:46] <kaduk@jabber.openafs.org/barnowl> Ah, it was just a display issue, phew.
[15:16:16] <kaduk@jabber.openafs.org/barnowl> Anyway, sorry for the linux-5.3 distraction.  Was there more on stable
branches other than "1.8.4pre1 coming soon"?
[15:16:32] <wiesand> no  :-(
[15:18:22] <kaduk@jabber.openafs.org/barnowl> On the master front, I tried to clean out my open browser tabs for "I
should look at that gerrit change" and got most of the way through
them, but that means I didn't look at the recent submissions, like
Yadav's aklog fixes :(
[15:18:36] <yadayada> Last week, I was working on aklog mem leak patches  (13760, 13761) . which has been reviewed by Andrew. I have few more patches mainly for aklog impersonate functionality. I will push those patches in coming week.
[15:18:57] <yadayada> I think aklog impersonate functionality in 1.8 branch is not working
[15:19:36] <yadayada> I have few more fixes for it, will push in coming week.
[15:19:53] <kaduk@jabber.openafs.org/barnowl> I also think it is not working in 1.8, and am happy that you can work
on fixing it
[15:20:51] <mvita> yes, it would be nice to have aklog impersonate working again
[15:21:17] <yadayada> I was also testing Andtew;s native mount support. I think by next week will be able to test/review native mounts support
[15:21:26] <yadayada> *Andrew's
[15:22:01] <kaduk@jabber.openafs.org/barnowl> That's pretty exciting!
[15:22:05] <meffie> i have some gerrits for aklog impersonate, but the changes were not correct.
[15:22:25] <meffie> (or could be better)
[15:22:43] <yadayada> yes Mike I saw that and was referring to it. We need to take some changes from your patches for sure. If you remember some incorrect #ifdef.
[15:22:52] <wiesand> It seems neither "aklog —help" nor the manpage mention impersonate at all… what am I missing?
[15:22:59] <yadayada> Currently impersonate code even don;t get compiled
[15:23:05] <yadayada> we just return -1
[15:23:38] <meffie> i dont remember the details, i remember andrew had some ideas to fix it correctly.
[15:24:32] <yadayada> right, I think with current fixes which I had, my basic tests were working.
[15:24:52] <meffie> nice. thanks!
[15:24:54] <kaduk@jabber.openafs.org/barnowl> impersonate is the -keytab argument (and one other, from memory), and
semi-intentionally undocumented.  It's only useful if you have the
cell's secret key available, of course...
[15:25:08] <yadayada> that;s right
[15:25:21] <yadayada> and I see that tokens are for infinte time
[15:25:28] <yadayada> expires on 17 Jan 2038
[15:25:39] <yadayada> max of signed int
[15:25:39] <meffie> i feel a separate command name would be nice. it think hiemdal does that.
[15:26:36] <wiesand> hmm, no mention of -keytab either…
[15:26:57] <meffie> esp since they are now versions of akog that are broken wrt akimpersonate
[15:28:35] <kaduk@jabber.openafs.org/barnowl> (the hardcoded list of enctypes that only lists single-DES is at least
part of why akimpersonate is broken -- the keytab isn't likely to have
such a key anymore.)
[15:28:58] <kaduk@jabber.openafs.org/barnowl> Anyway, we probably don't need to hammer out the akimpersonate fixes
during the meeting :)
[15:29:27] <wiesand> print "dark side of the force stuff\n" if /aklog.*(keytab|impersonate)/;
[15:30:30] <kaduk@jabber.openafs.org/barnowl> in other news, I see there's a new "rx_dead" topic in gerrit; is there
much to say about it?
[15:31:48] <wiesand> "foo"
[15:33:52] <meffie> bar
[15:36:40] <kaduk@jabber.openafs.org/barnowl> Hmm, gerrit's git interface seems to be very slow at the moment (and
buildbot had some builds fail becaues they couldn't find the remote
ref)
[15:37:16] <meffie> a git push failed for me yesterday too. retry worked.
[15:38:10] <kaduk@jabber.openafs.org/barnowl> I wonder if I should go restart the process
[15:38:33] <yadayada> even for me git push failed on wednesday  ... I got error as "romote end hung"
[15:38:48] <yadayada> I was able to push after waiting for ~ 3 hrs
[15:41:20] <mvita> sorry I got a call, but I'm back now
[15:42:09] <kaduk@jabber.openafs.org/barnowl> Do we have much else for today?
[15:42:19] <kaduk@jabber.openafs.org/barnowl> (I'm still sitting here waiting on 'git fetch')
[15:42:32] <mvita> did you have any questions about the rx_dead stuff
[15:42:33] <mvita> ?
[15:42:41] <mvita> sorry for the dumb topic name
[15:44:20] <kaduk@jabber.openafs.org/barnowl> No specific questions yet, just wondering if there's a particular
reason it's needed, if it's urgent, etc.
[15:44:30] <mvita> no, not urgent
[15:44:49] <kaduk@jabber.openafs.org/barnowl> I mean, it's not like you used the topic name of rx_idledead; we don't
want to give Stephan a heart attack
[15:44:51] <mvita> the problem it addresses (hung volserver) seems relatively rare
[15:45:12] <meffie> how come that topic name has a an underscore rather than a dash? is that significant?
[15:45:22] <meffie> e.g. not rx-dead?
[15:45:25] <mvita> um no
[15:45:38] <kaduk@jabber.openafs.org/barnowl> I believe there is no significance to the choice of punctuation in
gerrit topic names.
[15:46:03] <meffie> hmm. ok, i though it was a symbol or something. normally topics have dashes.
[15:46:29] <mvita> I will eschew underscores in future topic names
[15:46:54] <meffie> ok, thanks.
[15:46:59] <kaduk@jabber.openafs.org/barnowl> Any objections to my restarting gerrit shortly?  These hangs are
getting annoying...
[15:47:07] <meffie> none yere
[15:47:09] <mvita> no objections
[15:47:10] <meffie> here
[15:47:13] <wiesand> none
[15:47:23] <yadayada> no objection
[15:48:33] <mvita> WAIT!
[15:48:37] <mvita> nope, just kidding
[15:48:55] <kaduk@jabber.openafs.org/barnowl> Starting Gerrit Code Review: OK
[15:51:54] <kaduk@jabber.openafs.org/barnowl> Anything else we want to talk about today?
[15:52:31] <yadayada> nothing from my side
[15:52:40] <mvita> no thank you
[15:52:50] <meffie> no thank you
[15:53:03] <yadayada> Thanks
[15:53:33] <kaduk@jabber.openafs.org/barnowl> motion to adjourn?
[15:53:41] <meffie> second
[15:53:52] <meffie> oh, one thing :)
[15:54:12] <meffie> i pushed a change to NEWS for 1.9.x
[15:54:26] <meffie> thats all
[15:54:35] <kaduk@jabber.openafs.org/barnowl> Oh yes, I saw that notification; thank you
[15:54:44] <kaduk@jabber.openafs.org/barnowl> Was it just to resync with 1.8.4's NEWS?
[15:54:49] <meffie> yes
[15:54:54] <kaduk@jabber.openafs.org/barnowl> ok, cool
[15:57:17] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[15:57:23] <wiesand> Thanks
[15:57:27] <mvita> have a great weekend
[15:57:35] <meffie> thanks.
[15:57:42] meffie leaves the room
[15:57:46] wiesand leaves the room
[16:47:20] yadayada joins the room
[16:54:08] yadayada leaves the room
[17:10:04] yadayada leaves the room
[17:37:18] yadayada joins the room
[17:51:35] yadayada joins the room
[17:55:20] yadayada leaves the room
[18:32:58] mbarbosa leaves the room
[18:33:11] mbarbosa joins the room
[18:34:45] yadayada joins the room
[18:48:47] yadayada leaves the room
[19:51:58] yadayada joins the room
[19:52:56] yadayada leaves the room
[20:23:25] mvita leaves the room
[20:39:05] yadayada joins the room
[20:39:21] yadayada leaves the room
[20:49:03] mvita joins the room
[20:55:34] yadayada leaves the room
[21:28:58] yadayada joins the room
[21:34:37] mbarbosa leaves the room
[21:49:31] yadayada leaves the room
[23:00:41] yadayada joins the room
[23:04:37] yadayada leaves the room
[23:38:06] meffie joins the room
[23:41:33] yadayada joins the room
[23:44:19] meffie leaves the room