ReleaseEngineering/Meetings/2007-dec-10

Release Engineering Meeting :: Monday 2007-12-10

IRC Transcript

-!- f13 changed the topic of #fedora-meeting to: Fedora Release Engineering meeting		<a href="#t13:00" class="time">13:00</a>
f13	ping notting jeremy rdieter wwoods spot lmacken poelcat jwb	<a href="#t13:01" class="time">13:01</a>
* notting is here		<a href="#t13:01" class="time">13:01</a>
jwb	here	<a href="#t13:01" class="time">13:01</a>
rdieter	here	<a href="#t13:01" class="time">13:01</a>
jeremy	f13: I'll be back in just a minute	<a href="#t13:01" class="time">13:01</a>
* poelcat here		<a href="#t13:02" class="time">13:02</a>
* wwoods here		<a href="#t13:02" class="time">13:02</a>
jeremy	ok	<a href="#t13:03" class="time">13:03</a>
f13	alright.	<a href="#t13:03" class="time">13:03</a>
f13	I suspect spot is sleeping off the jet lag	<a href="#t13:03" class="time">13:03</a>
f13	oh warren ping	<a href="#t13:03" class="time">13:03</a>
jeremy	I was talking to him before lunch, but perhaps	<a href="#t13:03" class="time">13:03</a>
f13	well, once again not much on the agenda.	<a href="#t13:04" class="time">13:04</a>
f13	FC6 is dead.	<a href="#t13:04" class="time">13:04</a>
wwoods	so long, zod	<a href="#t13:04" class="time">13:04</a>
f13	We're going to try reversing the netapp streams next week	<a href="#t13:04" class="time">13:04</a>
f13	(sync from PHX to other places rather than from RDU)	<a href="#t13:04" class="time">13:04</a>
f13	and we have a buildsystem refresh coming	<a href="#t13:04" class="time">13:04</a>
f13	and a glibc/gcc update coming that could use a mass rebuild	<a href="#t13:05" class="time">13:05</a>
* poelcat put out a 2nd request for feature pages		<a href="#t13:05" class="time">13:05</a>
wwoods	I'm working on a rawhide dashboard (<a href="http://wwoods.fedorapeople.org/rawhide.html">http://wwoods.fedorapeople.org/rawhide.html</a>)	<a href="#t13:05" class="time">13:05</a>
jeremy	wwoods: coolio	<a href="#t13:06" class="time">13:06</a>
jwb	f13, update for what?	<a href="#t13:06" class="time">13:06</a>
wwoods	need to find a way to get mash and/or the rawhide sync to send a signal to the rawhide monitor thingy	<a href="#t13:06" class="time">13:06</a>
jeremy	jwb: gcc 4.3	<a href="#t13:06" class="time">13:06</a>
wwoods	so the top section(s) of that page get updated after rawhide finishes building/syncing	<a href="#t13:06" class="time">13:06</a>
jwb	jeremy, and glibc?	<a href="#t13:06" class="time">13:06</a>
poelcat	wwoods: will it signal whether it is installable?	<a href="#t13:06" class="time">13:06</a>
jeremy	jwb: no big glibc change that I know of	<a href="#t13:06" class="time">13:06</a>
wwoods	poelcat: yes	<a href="#t13:07" class="time">13:07</a>
jwb	yet	<a href="#t13:07" class="time">13:07</a>
f13	maybe I was mistaken on the glibc	<a href="#t13:07" class="time">13:07</a>
wwoods	poelcat: it checks to see if the tree is there, checks the metadata, checks for boot images	<a href="#t13:07" class="time">13:07</a>
jwb	wwoods, so does it have a way to query why something went bad?	<a href="#t13:07" class="time">13:07</a>
f13	re-reading the mail it looks like just gcc	<a href="#t13:07" class="time">13:07</a>
wwoods	if there's boot images it will (eventually) launch automatic tests and report those results	<a href="#t13:07" class="time">13:07</a>
wwoods	well, it links to the logs	<a href="#t13:07" class="time">13:07</a>
wwoods	so you can figure it out	<a href="#t13:07" class="time">13:07</a>
jwb	it does?	<a href="#t13:07" class="time">13:07</a>
poelcat	f13: glibc goes to 2.8	<a href="#t13:08" class="time">13:08</a>
jwb	oh, i see	<a href="#t13:08" class="time">13:08</a>
jwb	poelcat, not until the end	<a href="#t13:08" class="time">13:08</a>
jwb	wwoods, odd.  a 0 log for ppc	<a href="#t13:08" class="time">13:08</a>
poelcat	wwoods: why "make the user figure out it"... why not a clear status that says "WORKY" or "NOWORKY"	<a href="#t13:08" class="time">13:08</a>
jwb	anyway, this is cool	<a href="#t13:08" class="time">13:08</a>
* poelcat has a daily internal rhts job doing the same thing right now		<a href="#t13:09" class="time">13:09</a>
wwoods	poelcat: there *is* clear status	<a href="#t13:09" class="time">13:09</a>
* jwb wonders why ppc has a log size of 0		<a href="#t13:09" class="time">13:09</a>
wwoods	either "no tree", "bad/missing metadata", "no boot images", or "tree OK"	<a href="#t13:09" class="time">13:09</a>
wwoods	followed by results of the attempted install test	<a href="#t13:10" class="time">13:10</a>
f13	jwb: because the pungify-ppc didn't actually comlete	<a href="#t13:10" class="time">13:10</a>
f13	complete	<a href="#t13:10" class="time">13:10</a>
jwb	f13, i see that	<a href="#t13:10" class="time">13:10</a>
wwoods	will will probably be either "OK" or "FAIL" with a link to anaconda logs	<a href="#t13:10" class="time">13:10</a>
f13	due to an nfs mounting error which I have reported to mmcgrath	<a href="#t13:10" class="time">13:10</a>
wwoods	err which will ..	<a href="#t13:10" class="time">13:10</a>
jwb	f13, i see that too :)	<a href="#t13:10" class="time">13:10</a>
f13	jwb: I'm going to try it again manually in a few minutes.	<a href="#t13:10" class="time">13:10</a>
wwoods	I can't magically determine why anaconda failed when it fails	<a href="#t13:10" class="time">13:10</a>
jwb	f13, ok cool	<a href="#t13:10" class="time">13:10</a>
wwoods	but there should be obvious pass/fail status	<a href="#t13:10" class="time">13:10</a>
wwoods	so yeah.	<a href="#t13:12" class="time">13:12</a>
jwb	wwoods, we need the doom-o-meter on there	<a href="#t13:12" class="time">13:12</a>
jwb	otherwise it's cool	<a href="#t13:13" class="time">13:13</a>
jwb	*crickets*	<a href="#t13:14" class="time">13:14</a>
wwoods	ah yes, the doom-o-meter!	<a href="#t13:14" class="time">13:14</a>
wwoods	yeah that'll come once I have the other bits working	<a href="#t13:14" class="time">13:14</a>
jwb	cool	<a href="#t13:14" class="time">13:14</a>
wwoods	but yeah, the main thing is working out a way for the rawhide build/sync process to signal that it's ready for smoke testing	<a href="#t13:15" class="time">13:15</a>
f13	and that has some entertaining details if we reverse the stream	<a href="#t13:17" class="time">13:17</a>
notting	f13: actually, it's a lot simpler	<a href="#t13:17" class="time">13:17</a>
f13	since we'd just be doing an rsync to $box in phx as part of compose, and relying upon snapmirror to move the bits around.	<a href="#t13:17" class="time">13:17</a>
f13	notting: we can't tell $process when the bits are ready in RDU	<a href="#t13:18" class="time">13:18</a>
notting	f13: but we can tell them when they're ready in phx. which is probably 'close enough'	<a href="#t13:18" class="time">13:18</a>
notting	as that's public	<a href="#t13:18" class="time">13:18</a>
f13	but not nfs mountable by wwoods' farm.	<a href="#t13:19" class="time">13:19</a>
f13	which is the problem he's going to face	<a href="#t13:19" class="time">13:19</a>
wwoods	I don't need nfs mountability	<a href="#t13:19" class="time">13:19</a>
f13	I thought you did for some tests?	<a href="#t13:19" class="time">13:19</a>
wwoods	not really.	<a href="#t13:19" class="time">13:19</a>
f13	(well, the obvious one being nfs install test)	<a href="#t13:19" class="time">13:19</a>
wwoods	the stuff is designed to use http/ftp for installs, since that seems to be the typical way people get trees and do network installs	<a href="#t13:20" class="time">13:20</a>
f13	sure, that just means we'll miss nfs install bugs until later (:	<a href="#t13:20" class="time">13:20</a>
jeremy	f13: the really scary ones are iso based methods... so meh.	<a href="#t13:21" class="time">13:21</a>
f13	heh	<a href="#t13:21" class="time">13:21</a>
wwoods	well, I can set up some stuff to wait for the bits to land in RDU to do local nfs-using tests on a local cluster	<a href="#t13:21" class="time">13:21</a>
wwoods	but for doing simple smoke tests	<a href="#t13:21" class="time">13:21</a>
wwoods	you kind of want that to happen as soon as the bits are built	<a href="#t13:21" class="time">13:21</a>
wwoods	NFS is tier2	<a href="#t13:22" class="time">13:22</a>
* jeremy would be quite happy with some ftp/http smoketesting and isn't going to look a gift horse in the mouth :)		<a href="#t13:22" class="time">13:22</a>
wwoods	a simple %packages --default http kickstart install would make a good basic smoketest	<a href="#t13:22" class="time">13:22</a>
wwoods	and/or %packages @base	<a href="#t13:22" class="time">13:22</a>
* f13 neither		<a href="#t13:23" class="time">13:23</a>
wwoods	%packages --default kickstart install covers a lot of the key codepaths - if that works we're not completely doomed	<a href="#t13:23" class="time">13:23</a>
f13	by all means, do whatever you can for the love of god (:	<a href="#t13:23" class="time">13:23</a>
wwoods	so, hm. if I set up something (e.g. an xmlrpc/cgi thing) could you have it be pinged as soon as the bits are baked?	<a href="#t13:24" class="time">13:24</a>
f13	couldn't you set a watch on a file ?	<a href="#t13:27" class="time">13:27</a>
f13	like if we touched a file that said "all done"?	<a href="#t13:28" class="time">13:28</a>
poelcat	wwoods: why can't you go off of internal gridtrees (or whatever it is called)	<a href="#t13:28" class="time">13:28</a>
f13	poelcat: that's not used by Fedora	<a href="#t13:28" class="time">13:28</a>
wwoods	because gridtrees is a horrible abomination	<a href="#t13:28" class="time">13:28</a>
wwoods	that must be stopped	<a href="#t13:28" class="time">13:28</a>
poelcat	but it tells you when a new rawhide tree is there	<a href="#t13:28" class="time">13:28</a>
wwoods	and also it's internal-only	<a href="#t13:28" class="time">13:28</a>
wwoods	just no.	<a href="#t13:29" class="time">13:29</a>
f13	poelcat: I don't recall any current code that touches gridtrees	<a href="#t13:29" class="time">13:29</a>
wwoods	if gridtrees is being updated with rawhide info that's a side-effect or a hack	<a href="#t13:29" class="time">13:29</a>
wwoods	horrible deprecated	<a href="#t13:29" class="time">13:29</a>
wwoods	ugh	<a href="#t13:29" class="time">13:29</a>
wwoods	the entire point of having .composeinfo/.treeinfo was to stop using gridtrees	<a href="#t13:29" class="time">13:29</a>
f13	yeah, no rawhide content in gridtrees	<a href="#t13:29" class="time">13:29</a>
wwoods	you really really really don't want multiple processes writing the same file over NFS	<a href="#t13:30" class="time">13:30</a>
f13	yes, much doomage	<a href="#t13:30" class="time">13:30</a>
f13	wwoods: so theoretically we can "ping" you in some form.	<a href="#t13:30" class="time">13:30</a>
f13	the most simple way possible would be preferred	<a href="#t13:30" class="time">13:30</a>
wwoods	f13: yeah, if you write a given file last	<a href="#t13:31" class="time">13:31</a>
wwoods	or atomically move the tree into place after syncing it to a temp location	<a href="#t13:31" class="time">13:31</a>
f13	actually, the repodata files are synced last	<a href="#t13:31" class="time">13:31</a>
wwoods	I'd prefer that the signal was a push rather than having to poll	<a href="#t13:31" class="time">13:31</a>
f13	but you're right	<a href="#t13:31" class="time">13:31</a>
jwb	thinking generally, it might be nice to have a compose framework	<a href="#t13:31" class="time">13:31</a>
f13	a very simplistic push would be best	<a href="#t13:31" class="time">13:31</a>
jwb	ala, koji, but for composes	<a href="#t13:31" class="time">13:31</a>
f13	jwb: and what would it do?	<a href="#t13:32" class="time">13:32</a>
jwb	something the secondary arches could plug into to get notification of mashing and send it when they've finished composing	<a href="#t13:32" class="time">13:32</a>
wwoods	f13: right. I can set up a .cgi and you can just do something like 'wget ${cgi_url}?newtree=${dirname}'	<a href="#t13:32" class="time">13:32</a>
wwoods	or similar	<a href="#t13:33" class="time">13:33</a>
f13	nod	<a href="#t13:33" class="time">13:33</a>
wwoods	or just ${cgi_url}?dopoll=true	<a href="#t13:33" class="time">13:33</a>
wwoods	whatevs	<a href="#t13:33" class="time">13:33</a>
f13	jwb: ok, yeah, I had at one point evisioned a big communication bus between things like bodhi, koji, composes, etc... that things could subscribe too just for notifications ( police scanner ) , drop messages on, or otherwise react to calls.	<a href="#t13:34" class="time">13:34</a>
jwb	yeah	<a href="#t13:34" class="time">13:34</a>
wwoods	yeah, we've talked about that in QA-land as well	<a href="#t13:34" class="time">13:34</a>
notting	f13: dbus++?	<a href="#t13:34" class="time">13:34</a>
wwoods	there's messaging stuff going on in RH (aqmp for example) but those are a bit fancy for us, I think	<a href="#t13:34" class="time">13:34</a>
notting	f13: steroiD-bus?	<a href="#t13:34" class="time">13:34</a>
jwb	notting, er fbus man	<a href="#t13:34" class="time">13:34</a>
wwoods	dbus supports multihost stuff, doesn't it?	<a href="#t13:35" class="time">13:35</a>
jwb	does it?	<a href="#t13:35" class="time">13:35</a>
wwoods	it has some vestigial support for doing that anyway	<a href="#t13:35" class="time">13:35</a>
jwb	where's J5?	<a href="#t13:35" class="time">13:35</a>
wwoods	there's a tcp transport for dbus	<a href="#t13:35" class="time">13:35</a>
J5	jwb: here	<a href="#t13:35" class="time">13:35</a>
jwb	J5, can dbus do multihost stuff?	<a href="#t13:35" class="time">13:35</a>
jwb	easily?	<a href="#t13:35" class="time">13:35</a>
jwb	so that a 5 year old could code something up?	<a href="#t13:35" class="time">13:35</a>
J5	multihost?	<a href="#t13:35" class="time">13:35</a>
jwb	J5, machine 1 sends an event on the bus and machines 2-5 get it	<a href="#t13:36" class="time">13:36</a>
wwoods	it's not very well documented IIRC, but you can do stuff like: <listen>tcp:host=localhost,port=12434</listen>	<a href="#t13:36" class="time">13:36</a>
wwoods	in dbus system.conf	<a href="#t13:36" class="time">13:36</a>
jwb	wwoods, is there an equivalent broadcast?	<a href="#t13:36" class="time">13:36</a>
wwoods	you'd think you could use multicast for that	<a href="#t13:36" class="time">13:36</a>
wwoods	but now I'm speculating wildly	<a href="#t13:37" class="time">13:37</a>
J5	jwb: ya, not sure what our tcp support is like	<a href="#t13:37" class="time">13:37</a>
notting	maybe amqp is better for this	<a href="#t13:37" class="time">13:37</a>
* jwb shrugs		<a href="#t13:37" class="time">13:37</a>
J5	we are primarily a local bus	<a href="#t13:37" class="time">13:37</a>
jwb	that's what i thought	<a href="#t13:37" class="time">13:37</a>
J5	tubes has multicast support I think	<a href="#t13:37" class="time">13:37</a>
J5	we do dbus over tubes on the OLPC	<a href="#t13:38" class="time">13:38</a>
wwoods	that sounds like it might be the Right Thing To Do	<a href="#t13:38" class="time">13:38</a>
f13	"tubes"	<a href="#t13:39" class="time">13:39</a>
wwoods	oh, so tubes is.. basically dbus over xmpp?	<a href="#t13:39" class="time">13:39</a>
J5	wwoods: tubes abstracts the transport	<a href="#t13:40" class="time">13:40</a>
wwoods	well, details aside because.. yeah	<a href="#t13:40" class="time">13:40</a>
wwoods	heh	<a href="#t13:40" class="time">13:40</a>
J5	xmpp is one of the transports	<a href="#t13:40" class="time">13:40</a>
wwoods	but yeah. sounds like Tubes is the answer for "dbus over network"	<a href="#t13:40" class="time">13:40</a>
f13	so...	<a href="#t13:41" class="time">13:41</a>
f13	we're pretty far off in the weeds	<a href="#t13:41" class="time">13:41</a>
wwoods	right. that's like a post-f9/f10 kind of target	<a href="#t13:41" class="time">13:41</a>
jwb	sure	<a href="#t13:41" class="time">13:41</a>
f13	wwoods: if you get something simple up that can be poked via urllib or whatnot, let us know and we will update the compose job	<a href="#t13:41" class="time">13:41</a>
wwoods	for testing f9 I'll set up some cutely stupid cgi that you can poke	<a href="#t13:41" class="time">13:41</a>
f13	but under a similar topic of 'weeds', who all is going to be at FUDCon?	<a href="#t13:41" class="time">13:41</a>
f13	and if there is a number of us, should we get together and voice some things out?	<a href="#t13:42" class="time">13:42</a>
notting	i suppose i can make the trip	<a href="#t13:42" class="time">13:42</a>
wwoods	since F9 is going to become RHEL6 when it gets growed up, the RHEL QA guys are going to be working on F9/rawhide testing	<a href="#t13:42" class="time">13:42</a>
f13	notting: now kind of you	<a href="#t13:42" class="time">13:42</a>
wwoods	yeah I'll probably be at FUDCon, seeing as it's, y'know, here	<a href="#t13:42" class="time">13:42</a>
notting	f13: heh	<a href="#t13:42" class="time">13:42</a>
notting	f13: i suspect if i try and expense travel someone will laugh at me	<a href="#t13:42" class="time">13:42</a>
f13	notting: sadly, you could probably get away with it	<a href="#t13:43" class="time">13:43</a>
notting	f13: are you coming?	<a href="#t13:45" class="time">13:45</a>
f13	yes	<a href="#t13:46" class="time">13:46</a>
f13	I'm assuming wwoods will be there	<a href="#t13:48" class="time">13:48</a>
* jeremy isn't going to be at fudcon		<a href="#t13:48" class="time">13:48</a>
f13	and j5 will be	<a href="#t13:48" class="time">13:48</a>
wwoods	I'll be there, yeah.	<a href="#t13:49" class="time">13:49</a>
f13	well maybe will figure out something to do then.	<a href="#t13:49" class="time">13:49</a>
f13	On the signing server front, I think Mike McGrath is ready for testing new hosted environment, and we're going to be fodder.	<a href="#t13:49" class="time">13:49</a>
jwb	cool	<a href="#t13:50" class="time">13:50</a>
f13	so we may ahve a public repo in the next day or 3.  Luke and I are going to get together and hash out some of the API stuff, and I think I'm going ot let him work on the server side, and I'll focus on a fedora specific client for it that deals with koji and figuring out /what/ to sign and where to import the signed copy from	<a href="#t13:50" class="time">13:50</a>
lmacken	I did a bunch of hacking on the signing server over the weekend	<a href="#t13:51" class="time">13:51</a>
f13	this may need a new koji call, to be able to import from 'relative' location, relative to the koji server.	<a href="#t13:51" class="time">13:51</a>
lmacken	it's almost ready for testing	<a href="#t13:51" class="time">13:51</a>
f13	cool	<a href="#t13:51" class="time">13:51</a>
f13	Are there any other topics people would like to discuss?	<a href="#t13:53" class="time">13:53</a>
* wwoods sits on hands		<a href="#t13:54" class="time">13:54</a>
jwb	oh, i will not be at FUDCon	<a href="#t13:55" class="time">13:55</a>
f13	darn	<a href="#t13:55" class="time">13:55</a>
jwb	no surprise there	<a href="#t13:55" class="time">13:55</a>
f13	jwb: what about the Boston one with RH Summit attached?	<a href="#t13:55" class="time">13:55</a>
jwb	when is that?	<a href="#t13:55" class="time">13:55</a>
f13	June 18-20	<a href="#t13:56" class="time">13:56</a>
f13	although I may not actually be around then, let me check.	<a href="#t13:56" class="time">13:56</a>
jwb	i'll try to get to that one	<a href="#t13:58" class="time">13:58</a>
f13	anywho, hours up, seems we're done.  Back on your heads!	<a href="#t13:58" class="time">13:58</a>
-!- f13 changed the topic of #fedora-meeting to: Channel is used by various Fedora groups and committees for their regular meetings | Note that meetings often get logged | For questions about using Fedora please ask in #fedora | See ttp://fedoraproject.org/wiki/Communicate/FedoraMeetingChannel for meeting schedule		<a href="#t13:58" class="time">13:58</a>
--- Log closed Mon Dec 10 20:53:28 2007

Generated by irclog2html.py 2.3 by Marius Gedminas - find it at mg.pov.lt!