- Friday, April 24, 2020

release-team@conference.openafs.org

Friday, April 24, 2020< ^ >

meffie has set the subject to: openafs release team

Room Configuration

Room Occupants

GMT+0
[04:45:44] meffie joins the room
[04:50:43] meffie leaves the room
[04:51:17] meffie joins the room
[08:31:18] meffie leaves the room
[08:31:28] meffie joins the room
[09:30:58] meffie leaves the room
[09:31:50] meffie joins the room
[12:20:43] mbarbosa joins the room
[15:23:25] meffie leaves the room
[15:38:08] cwills joins the room
[15:38:22] cwills leaves the room
[15:38:38] cwills joins the room
[15:47:21] yadayada joins the room
[16:00:39] <kaduk@jabber.openafs.org/barnowl> Oops, have to go check the laundry
[16:00:55] <yadayada> Hi All
[16:01:08] <cwills> Hello all.
[16:07:08] wiesand joins the room
[16:07:29] <wiesand> Hello
[16:08:25] <cwills> Hello Stephan, hope all is well with you
[16:08:25] <kaduk@jabber.openafs.org/barnowl> [back now -- sorry!]
[16:08:52] <wiesand> So what's new? I was about to roll tarballs and upload them when Christof's message arrived…
[16:10:10] <kaduk@jabber.openafs.org/barnowl> I guess I'm curious if Cheyenne can test on new or bleeding-edge
32-bit kernels easily
[16:10:11] <wiesand> Do we have a solution? Should we upload and announce pre2 anyway, and just add a disclaimer re some 32-bit Linux distros to the announcement?
[16:11:09] <kaduk@jabber.openafs.org/barnowl> Note that opensuse tumbleweed is the "bleeding edge" distro and is one
of the earliest places that new linux kernels show up.
[16:11:21] <cwills> I think Mark is looking at the issue(s) Christof brought up.  I haven't looked at what it would take to set up a build/test for 32bit ..
[16:11:30] <kaduk@jabber.openafs.org/barnowl> So it may well just be a typical "bleeding-edge linux has some rough
edges" thing, but I'm not 100% sure
[16:11:57] <kaduk@jabber.openafs.org/barnowl> Cheyenne: okay, thanks for the update.  I didn't know how tied to
amd64 your current setup was.
[16:12:43] <kaduk@jabber.openafs.org/barnowl> My inclination is to announce pre2 "as-is", but would like to hear
from others
[16:12:46] <cwills> Just a FWIW.. I use a gentoo system and it provides a "sys-kernel/git-sources" which is kept up to date with the linux repo (with no distro specific updates)
[16:13:03] <wiesand> Well who's using 32-bit OSs these days - except embedded stuff?
[16:13:11] <kaduk@jabber.openafs.org/barnowl> "not me!"
[16:13:12] <cwills> So I just need to figure out what I need to do in order to do a 32bit build
[16:13:44] <kaduk@jabber.openafs.org/barnowl> Ah
[16:13:54] <wiesand> Speaks volumes, doesn't it?
[16:14:02] <yadayada> it looks we are not completely broken for 32 bit. So I agree to add disclaimer that some 32 bit distros can have issues
[16:14:28] <cwills> And on the linux 5.7 front.. I might have a simple fix
[16:14:47] <wiesand> Great news.
[16:15:16] <wiesand> Re pre2, I too am inclined to upload & announce it as is.
[16:16:01] <wiesand> I simply was hesitant to do it with a reported problem ans someone working on it.
[16:16:26] <wiesand> I tend to hesitate a lot these days. Should work on that…
[16:18:09] <kaduk@jabber.openafs.org/barnowl> Life is a work in progress, as they say...
[16:18:20] <kaduk@jabber.openafs.org/barnowl> It sounds like that is about it for 1.8.6?
[16:18:43] <wiesand> In the absence of any comments during this meeting speaking against it, I'll get 1.8.6pre2 done and out over the weekend.
[16:19:35] <wiesand> And yes, I think that's it. Lx5.7 fix would be good to have, but should probably wait for pre3 or 1.8.7.
[16:19:41] <kaduk@jabber.openafs.org/barnowl> +1
[16:20:59] <wiesand> On to master then?
[16:21:14] <cwills> I think the fix is fairly simple.  Its slightly retroactive and I've ran a quick test using a Centos7 system.  I haven't fired up a 5.7 system yet with the fix to actually test yet.
[16:21:39] <wiesand> 5.7?
[16:21:49] <cwills> linux 5.7
[16:22:18] <wiesand> ah, thought it might be about RHEL 5.7 ;-)
[16:22:28] <cwills> Sorry..
[16:22:45] <kaduk@jabber.openafs.org/barnowl> It feels like I didn't get a whole lot done on master this week,
though I guess there's a decent number of things that got merged
(Yadav's -setpag stuff plus some of the FreeBSD bits that have been
lingering).
[16:23:23] <kaduk@jabber.openafs.org/barnowl> I am happy to see the autoconf-archive import bits, and just saw a
couple 'i's left to dot before it goes in.
[16:23:46] <kaduk@jabber.openafs.org/barnowl> No progress on ubik-reads-during-recovery yet, but that is hopefully
queued up for the weekend.
[16:24:01] <yadayada> Thanks for review Ben for setpag. I think we need to make chnages to documentation of klog setpag, as it does not work for non-root process. Should I submit a gerrit with that documentation change ?
[16:24:37] <wiesand> that's about "aklog -setpag" ?
[16:24:50] <kaduk@jabber.openafs.org/barnowl> A gerrit for documentation changes sounds good, thanks!
[16:25:04] <yadayada> this is for klog -setpag
[16:25:23] <kaduk@jabber.openafs.org/barnowl> klog, the unloved stepchild of aklog, as it were.
[16:25:35] <kaduk@jabber.openafs.org/barnowl> (never mind that klog came first...)
[16:25:42] <wiesand> really "klog", w/o the "a"?
[16:26:34] <kaduk@jabber.openafs.org/barnowl> Stephan: yes, really.  Some people are quite attached to it, even if
it is secretly klog.krb5 nowadays.
[16:26:52] <wiesand> ah :-)
[16:27:05] <yadayada> that's right :) klog.krb5
[16:28:11] <wiesand> NB I still believe klog.krb5 should rather be named "k5log".
[16:28:52] <kaduk@jabber.openafs.org/barnowl> I would have supported that ("k5log") were it done from the start ...
it feels a bit late to try to make a change now, though.
[16:28:53] <wiesand> Since it much more corresponds to "klog" than "klog.krb".
[16:29:32] <wiesand> I will admit though that "klog.krb5" probably was my proposal. Way back then…
[16:30:38] <wiesand> Maybe it's not too late?
[16:31:26] <kaduk@jabber.openafs.org/barnowl> We can always look at a gerrit that proposes changes :)
[16:32:08] <wiesand> "No matter how for you've walked down the wrong path, it's never too early for turning back"
[16:32:36] <wiesand> s/for/far/
[16:33:14] <kaduk@jabber.openafs.org/barnowl> Anyway, at risk of forking the thread, the main other 1.9.0 topic I
have is that I think I should just merge the NEWS now to plant a stake
in the sand in terms of which commits the current version covers
(i.e., to make future review/updates easier), even if the 1.9.0 tag
isn't quite ready yet.
[16:33:45] <wiesand> Sounds good
[16:34:26] <kaduk@jabber.openafs.org/barnowl> Relatedly, I can't remember if Mike was going to try to follow the
draft instructions I mailed out last week for getting the rxgk bits
tested.
[16:35:13] <kaduk@jabber.openafs.org/barnowl> Mike, who seems to not be in the room today, I should say...
[16:35:31] <cwills> He's stuck in a meeting I believe
[16:35:46] <kaduk@jabber.openafs.org/barnowl> I know the feeling :)
[16:36:06] <wiesand> "Me too " ;-)
[16:36:43] <kaduk@jabber.openafs.org/barnowl> Anyway, I feel like I talked a lot.  Were there other things that
happened during the week or should get updates?
[16:36:58] <cwills> It's starting to rain ...
[16:37:13] <kaduk@jabber.openafs.org/barnowl> I heard Boston got snow last week
[16:37:16] <wiesand> sheer envy!
[16:37:36] <cwills> (and my wife just left a few minutes ago to walk the dog)
[16:37:45] <kaduk@jabber.openafs.org/barnowl> Seattle seems to have remembered that it's supposed to have a
reputation for being rainy -- we had a good few sunny weeks, though.
[16:38:03] <wiesand> this place is becoming a desert
[16:38:31] <cwills> Where are you?
[16:38:39] <wiesand> Berlin area
[16:38:46] <cwills> Ah..
[16:40:29] <wiesand> I'm spending 2h and ~5EUR per day to keep my garden somewhat alive. Audible hearbooks help a lot.
[16:40:32] <cwills> correction.. not rain.. it's gruppel (a form of snow.. think ball bearing sized pellets)
[16:40:52] meffie joins the room
[16:41:06] <wiesand> cwills: where are *you* ?
[16:41:15] <kaduk@jabber.openafs.org/barnowl> gruppel, everyone's favorite new word? :)
[16:41:20] <cwills> Colorado.. rocky mountains (outside of Boulder)
[16:41:22] <meffie> (sorry, being late)
[16:41:59] <wiesand> Interestingly, the german word for it is "Graupel" (a minor form fo hail).
[16:42:53] <cwills> your right.. spelling error on my part
[16:44:24] <wiesand> you mean it's "graupel" in English as well?
[16:44:33] <cwills> yes
[16:45:17] <wiesand> Nice, I can add that to my collection of "Kindergarten" and "Gedankenexperiment" then :-)
[16:45:41] <kaduk@jabber.openafs.org/barnowl> Nice :)
[16:45:50] <kaduk@jabber.openafs.org/barnowl> Mike, thanks for joining.  Did you have any topics you wanted to
cover?
[16:49:32] <meffie> just read the scrollback. i dont have anything else. i agree we can do a 1.8.6pre2 as is.
[16:50:04] <meffie> mark was going to try the rxgk setup
[16:50:32] <meffie> can i put that draft rxgk notes on the wiki.openafs.org as well?
[16:50:46] <kaduk@jabber.openafs.org/barnowl> Putting it on the wiki sounds like a good idea, thanks!
[16:50:51] <meffie> ok, thanks.
[16:51:21] <kaduk@jabber.openafs.org/barnowl> Okay.  I guess that's a motion to adjourn, then.
[16:51:33] <yadayada> thanks
[16:51:47] <meffie> thank you everyone, have a safe weekend.
[16:52:13] <cwills> Thanks
[16:53:28] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[16:54:22] cwills leaves the room
[16:54:51] <wiesand> Thanks everybody!
[16:54:57] wiesand leaves the room
[16:57:38] meffie leaves the room
[17:07:12] meffie joins the room
[17:25:46] mbarbosa leaves the room
[17:26:12] mbarbosa joins the room
[17:31:12] meffie leaves the room
[17:32:05] meffie joins the room
[17:35:38] meffie leaves the room
[17:41:37] yadayada leaves the room
[20:40:08] yadayada joins the room
[21:27:16] yadayada leaves the room
[21:56:27] mbarbosa leaves the room