- Thursday, June 3, 2021

release-team@conference.openafs.org

Thursday, June 3, 2021< ^ >

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

Room Configuration

Room Occupants

GMT+0
[00:14:44] mvita2 joins the room
[10:54:32] mbarbosa joins the room
[11:43:20] meffie joins the room
[13:55:05] Cheyenne joins the room
[15:59:20] wiesand joins the room
[16:00:03] <kaduk@jabber.openafs.org/barnowl> greetings
[16:00:08] <wiesand> Hi
[16:00:23] <mvita2> <afk>
[16:00:46] <meffie> good day
[16:01:36] <wiesand> Hello Mike
[16:02:29] <wiesand> Made some progress today on 1.8.8pre1… merged away all changes I planned to get in and updated NEWS
[16:02:41] <Cheyenne> Hello all
[16:04:16] <meffie> excellent news, thanks wiesand
[16:04:18] <wiesand> Re NEWS: I'm a bit confused about the "14291 should be 14292" comments, since 14292 is just the master version of 14298 which was already on the list…
[16:04:31] <wiesand> So I just removed 14291
[16:04:42] <meffie> ah, thanks.
[16:05:57] <wiesand> Also there was a comment by Mark about 14441, but I seem to remember that in a meeting it was said that it was a problem encountered in the field after all - I just can't find that in the logs.
[16:07:10] <meffie> i think the wording you have for "NEWS" is good enough
[16:07:32] <wiesand> ok, thanks.
[16:08:02] <wiesand> Anyway, please have a look at PS4 and comment on what you don't like.
[16:09:29] <wiesand> Unless there's a late showstopper, after merging that and 14264 we should have a pre1.
[16:09:44] <meffie> Nice!
[16:11:19] <kaduk@jabber.openafs.org/barnowl> Huzzah!
[16:12:18] <wiesand> It's really about time. Shame on me.
[16:13:37] <Cheyenne> master and 1.8.x builds cleanly on linux-5.13-rc4+
[16:14:09] <wiesand> I was just about to ask :-) Thanks Cheyenne!
[16:14:29] <meffie> andrew mentions 14595 rx: Remove delays in multi_End_Ignore was merged to master, so should be ready to be pulled up to 1.8.x (i assume after the pre1 release)
[16:15:02] <meffie> he wanted to confirm he should not push directly to the 1.8.x gerrit branch.
[16:16:34] <wiesand> I do prefer pulling the changes up myself like we did for the 1.8.7/1.8.8 cycle. It makes things easier for me, and it's not what delayed 1.8.8pre1.
[16:17:02] <meffie> cool, just double checking
[16:18:37] <kaduk@jabber.openafs.org/barnowl> Your care in ordering changes that touch the same files seems quite
helpful -- I take a much more "free for all" approach on master and it
results in having to push the 'rebase' button a lot
[16:19:01] <kaduk@jabber.openafs.org/barnowl> (to be fair, it's probably harder to predict in what order things will
be mergeable onto master, so it's not clear I can do better)
[16:21:15] <wiesand> I don't envy you. It's indeed much simpler on 1.8.x, and since it avoids surprises and buildbot clogs I like it.
[16:21:39] <wiesand> 14595 is on the wish list. Looks like 1.8.9 fodder to me though.
[16:23:13] <meffie> thanks!
[16:24:53] <wiesand> NB it's ok to ask for the inclusion even before a change is merged on master (unless that queue would get all too long)
[16:25:24] <kaduk@jabber.openafs.org/barnowl> and asking can help me prioritize what to look at on master :)
[16:26:20] <wiesand> that's another benefit :)
[16:28:44] <wiesand> I feel there's general agreement that the current HEAD + NEWS + 14264 shall constitute 1.8.8pre1?
[16:28:53] <kaduk@jabber.openafs.org/barnowl> +1
[16:28:59] <meffie> +1
[16:29:24] <wiesand> (once NEWS is reviewed/agreed)
[16:29:27] <Cheyenne> +1
[16:30:14] <wiesand> That's all I have on 1.8.x for today.
[16:30:27] <meffie> thanks wiesand
[16:30:37] <wiesand> Buildbot behaviour is a bit funny currently.
[16:31:01] <kaduk@jabber.openafs.org/barnowl> Marks verified+1 on first builder return and then comments later with
the rest of the results?
[16:31:17] <wiesand> right ;-)
[16:31:39] <kaduk@jabber.openafs.org/barnowl> That does seem funny, yes.  I have no idea why...
[16:31:48] <meffie> yes, because we have two masters running, the new one has only one builder :)
[16:32:07] <kaduk@jabber.openafs.org/barnowl> only one builder, which is fast :)
[16:32:12] <meffie> yes
[16:32:37] <wiesand> but why doesn't the old master add a "verified +1"?
[16:32:45] <kaduk@jabber.openafs.org/barnowl> I will try to look this weekend at cutting over to the new machine.
[16:32:54] <meffie> and we have buildbot configured to do an "all or nothing" verification, so it has to wait for all of them to finish on a given master.
[16:33:01] <kaduk@jabber.openafs.org/barnowl> It runs as the same gerrit account and it can't set +1 again if it's
already set
[16:33:22] <wiesand> It's not an urgent problem. Just funny.
[16:33:37] <wiesand> ah
[16:33:48] <kaduk@jabber.openafs.org/barnowl> I didn't get much done this past week due to a combination of spending
time with family over the holiday and some work deadlines
[16:34:57] <kaduk@jabber.openafs.org/barnowl> So I still need to, e.g., get back to Andrew on the "remove
rpctestlib" and "remove unused opr/util components" changes that are
early in his tests stack
[16:35:02] <wiesand> Will you be available to give me a tag tomorrow?
[16:35:26] <kaduk@jabber.openafs.org/barnowl> I think so
[16:35:26] <meffie> no worries. i was on break last week, but did manage to push 14626 libadmin: xdr allocate rpc output strings
[16:36:17] <kaduk@jabber.openafs.org/barnowl> thanks for that one!
[16:36:36] <meffie> thanks.
[16:37:06] <Cheyenne> I've been looking at the static anlysis commits -- trying to split some of them up so they aren't as huge
[16:37:48] <meffie> (maybe the commit message could be more clear.  OUT string x<n>, needs a buffer  char x[n+1]  (if not allocated by xdr)
[16:37:51] <kaduk@jabber.openafs.org/barnowl> thanks for that as well.  The static-analysis stuff can be tiresome at
times but does catch some good fixes
[16:38:22] <kaduk@jabber.openafs.org/barnowl> In 202x we probably should not be trying to "optimize" by using
stack-allocated output arrays.
[16:38:28] <Cheyenne> also.. doing a indentation/whitespace cleanup of rx.h (andrews comment on the kdump removal)
[16:38:46] <kaduk@jabber.openafs.org/barnowl> +1
[16:38:59] <kaduk@jabber.openafs.org/barnowl> (even though I -1'd the kdump removal as a placeholder for that)
[16:39:41] <kaduk@jabber.openafs.org/barnowl> Anyway, sorry to try to rush along, but I had a small breakfast today
and apparently that means lunchtime is imminent.  Is there much else
to cover for master/1.9?
[16:39:51] <meffie> agreed, Cheyenne and I have patches to push to gerrit that get rid of all the char x[BSSIZE] in bozo. the patches are in github, just cleaning up and adding some ctap unit tests before sharing in gerrit.
[16:40:02] <kaduk@jabber.openafs.org/barnowl> excellent
[16:40:22] <meffie> and removes all strcpy, strcats too.
[16:41:05] <kaduk@jabber.openafs.org/barnowl> bonus!
[16:41:07] <meffie> that's all i have.
[16:42:06] <wiesand> Let's adjourn then, and allow Ben to go for lunch.
[16:42:32] <wiesand> Please review NEWS (14540) - I hope to get 1.8.8pre1out the door tomorrow.
[16:42:35] <Cheyenne> sounds good.
[16:42:37] <kaduk@jabber.openafs.org/barnowl> thanks everyone :)
[16:43:00] <wiesand> Thanks everybody!
[16:43:02] <meffie> thanks
[17:24:31] wiesand leaves the room
[17:36:43] mbarbosa leaves the room
[17:36:45] mbarbosa joins the room
[22:18:30] mbarbosa leaves the room
[22:23:10] meffie leaves the room
[22:37:36] Cheyenne leaves the room