17:04:02 <jzb> #startmeeting
17:04:02 <cs-meeting> Meeting started Fri Aug 17 17:04:02 2012 UTC. The chair is jzb. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:04:02 <cs-meeting> Useful Commands: #action #agreed #help #info #idea #link #topic.
17:04:41 * ke4qqq shows up - a bit late
17:04:47 <jzb> OK, hi all - so I was hoping we could meet briefly in real time to discuss anything relating to the 4.0 release that needs discussing. A reminder, this is discussion only.
17:04:57 <jzb> Any decisions to be taken should be on the mailing list.
17:05:27 <jzb> We didn't receive any agenda items on the list, so maybe we can go through each person here and see what they have to discuss/report.
17:05:33 <jzb> chipc: can we start with you?
17:05:47 <chipc> sure - I can report on the LICENSE file generation
17:05:54 <chipc> do that now? or collect agenda first?
17:06:20 <jzb> I think now is good - we have a small number of people here, so we can just go through alphabetically, I think.
17:06:24 <jzb> reasonable/
17:06:25 <jzb> ?
17:06:58 <chipc> ok, I've been using the target/jar folder to get through the bin dependencies
17:07:16 <chipc> I've had some success with Whisker, but it's clearly an early release version
17:07:27 <chipc> I have been patching that app locally (and will submit stuff back to that project)
17:07:40 <chipc> but for now, we won't be able to easily use it as a project team
17:08:09 <chipc> that being said, I'll be done with all of the required license and notice work by the end of the day… with a couple of exceptions
17:08:23 <chipc> ex: we use cloud-mysql-connector-java-5.1.7-bin.jar
17:08:26 <chipc> that's GPL
17:08:36 <ke4qqq> #info whisker not usable for 4.0, chipc upstreaming patches
17:08:45 <chipc> we also have cloud-trilead-ssh2-build213.jar in the tree, but we don't use it
17:08:57 <chipc> so that's about it… moving along
17:09:06 <chipc> 1200 lines in the LICENSE file as of right now
17:09:16 <ewanmellor> Anyone know why we have trilead but don't use it?
17:09:20 <ke4qqq> jzb: can you chair the folks here so we can add stuff to the summary and minutes?
17:09:30 <jzb> ke4qqq: yeah
17:09:50 <chipc> jzb: I'm done - any questions?
17:10:16 <jzb> #chair chipc ke4qqq cvittal edison_ ewanmellor jlkinsel topcloud
17:10:16 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor jlkinsel jzb ke4qqq topcloud
17:10:28 <ke4qqq> #info whisker not usable for 4.0, chipc upstreaming patches
17:10:34 <ke4qqq> tnx
17:10:41 <jzb> going once, going twice?
17:11:05 <jzb> OK, cvittal - do you have anything to report/discuss?
17:11:05 <topcloud> trilead is used.
17:11:37 <cvittal> yes
17:11:47 <topcloud> sorry...but can we ask questions to chip first?
17:11:47 <ewanmellor> Can we get question about trilead answered first?
17:11:58 <jzb> topcloud: yes, please go ahead.
17:12:03 <topcloud> or should we wait until everyone has gone through?
17:12:34 <topcloud> chipc: ewan's question is my question. where did you see trilead not being used?
17:12:35 <chipc> topcloud: thanks
17:12:57 <edison_> I think there are java code do use thrilead jar
17:13:00 <chipc> I thought that was the response on the list
17:13:00 <kelveny> trlead is used in VMware component
17:13:01 <jzb> #chair kelveny
17:13:01 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor jlkinsel jzb ke4qqq kelveny topcloud
17:13:18 <chipc> ok, I'll include it in the LICENSE then
17:13:29 <chipc> any other points?
17:13:35 <topcloud> kelveny: I thought we removed all mysql dependencies?
17:13:42 <kelveny> yes
17:13:49 <ke4qqq> we removed the deps in code
17:13:51 <kelveny> mysql static linkage is removed
17:13:52 <ke4qqq> but not the jar itself
17:14:13 <chipc> kelveny: so the cloud-mysql-connector-java-5.1.7-bin.jar can be removed then?
17:14:16 <kelveny> correct, of course not the jar itself
17:14:31 <ke4qqq> chipc: sorta
17:14:42 <ke4qqq> chipc: it's still a runtime dep
17:14:55 <ke4qqq> and needs to be noted as a system dependency
17:14:55 <chipc> #info I'll start an email thread about cloud-mysql-connector-java-5.1.7-bin.jar
17:14:58 <kelveny> jar file can' be removed, it is required at runtime for cloudstack to connect to MySql server
17:15:13 <ke4qqq> kelveny: we could make it a dep for the rpms or debs
17:15:19 <ke4qqq> which would then only screw up developers (smile)
17:15:22 <chipc> kelveny: we can't have it in the source or binary release
17:15:33 <chipc> kelveny: AFAIK
17:15:47 <ke4qqq> right - deps-ctrl lists that as a runtime dep iirc.
17:15:51 <chipc> but let's move that discussion onto the dev@ list
17:16:00 <ke4qqq> but yes, good to move that discussion.
17:16:06 <jzb> #chair techpubs
17:16:06 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor jlkinsel jzb ke4qqq kelveny techpubs topcloud
17:16:23 <kelveny> yeah, other developers not in IRC should be aware of
17:16:31 <topcloud> chipc: i'm done with my questions or comments.
17:16:34 <jzb> Any other q's for chipc?
17:16:46 <topcloud> chipc: thx!
17:17:22 <topcloud> so cvittal next?
17:17:24 <jzb> (Note - if you have questions but it's long one to type, you can just respond with "y" and we'll wait.)
17:17:31 <jzb> topcloud: indeed.
17:17:42 <cvittal> would like to move th discussion about systemvm forward
17:17:43 <jzb> #topic cvittal reporting
17:18:00 <cvittal> does anybody think that config files are licensable
17:18:17 <cvittal> there is only one way to configure the software
17:18:24 * ke4qqq doesn't know if what we think matters - also did we find a solution to iptables?
17:18:48 <ke4qqq> but you are right, config files probably are not copyrightable
17:18:49 <chipc> unsure, but perhaps we can take that question to legal-discuss@a.o
17:19:02 <cvittal> iptables?
17:19:03 <ewanmellor> config files are like any other textual work.
17:19:11 <jzb> chipc: +1
17:19:14 <ewanmellor> If they were written by someone else, then they are copyrightable.
17:19:29 <jzb> if you ask a specific question they're usually pretty quick to respond.
17:19:31 <cvittal> unless there is only one way to write them
17:19:33 <ewanmellor> Why don't you just write a config file from scratch.
17:19:35 <ewanmellor> ?
17:19:54 <cvittal> there is only one way to write configs for apache, haproxy, etc
17:19:59 <cvittal> dnsmasq
17:19:59 <ke4qqq> derivative work?
17:20:08 <ke4qqq> I mean if we think they are truly copyrightable
17:20:32 <cvittal> even if I wrote it from scratch it would look identical
17:21:01 <cvittal> ok, question for legal then
17:21:08 <ewanmellor> Where did they come from?
17:21:13 <ewanmellor> Originally, I mean.
17:21:23 <cvittal> some of them were hand written based on examples
17:21:37 <cvittal> others were modifications of defaults (e.g., /etc/ssh/
17:22:03 <ewanmellor> If you've written them by hand based on docs, then they are fine.
17:22:51 <jzb> #chair inteq
17:22:51 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor inteq jlkinsel jzb ke4qqq kelveny techpubs topcloud
17:23:03 <cvittal> ke4qqq: you had a q about iptables
17:23:14 <ke4qqq> yes - did we ever come up with a solution
17:23:30 <cvittal> remind me about the problem
17:23:48 <ke4qqq> iptables is gpl
17:23:54 <ke4qqq> we may not have gpl software
17:24:06 <ke4qqq> and we ship the deb (which is binary, which is also a no no
17:24:06 <ke4qqq> )
17:24:59 <cvittal> ah yes
17:25:25 <cvittal> we may have to remove this one
17:25:33 <edison_> need to remove iptables_1.4.8-3local1checksum1_i386.deb
17:25:49 <cvittal> and document a way for folks to get this deb
17:25:49 <ke4qqq> iirc the discussion said that it broke dhcp for Ubuntu or something
17:26:08 <cvittal> yes, if the vm and the router are on the same XS host
17:26:18 <edison_> people can build his own systemvm
17:26:55 <cvittal> so: systemvm can be
17:26:56 <edison_> for apache release, we may not include this dhcp fix
17:27:03 <topcloud> is this required for 4.0 release?
17:27:07 <topcloud> sounds like a lot of work.
17:27:09 <cvittal> a. built from scratch using source in apache
17:27:25 <ke4qqq> so when should we expect docs on how $CS_Consumer can build a systemVM?
17:27:27 <cvittal> b. downloaded as a convenience binary that may not be identical to the one built in apache master
17:27:54 <jzb> it sounds like we need to come to a conclusion on this on the mailing list, yes?
17:28:07 <cvittal> difference being the iptables fix
17:28:27 <cvittal> i'll put the document together
17:28:28 <jzb> cvittal: could you send out a proposal on that?
17:28:35 <cvittal> yes
17:28:46 <ke4qqq> #action cvittal to generate a proposal for sysvms on list.
17:28:56 <jzb> ke4qqq: you beat me to it.
17:29:02 * jzb typing slowly today, apparently.
17:29:14 <ke4qqq> jzb: prolly been using meetbot a bit longer than you (smile)
17:29:19 <jzb> ke4qqq: there is that
17:29:26 <jzb> for folks who haven't been using meetbot much
17:29:39 <jzb> the #action command will generate an action item
17:29:49 <jzb> the #info command will generate an info item that's called out
17:30:11 <topcloud> jzb: thanks for the info.
17:30:16 <jzb> cvittal: any other items, or any questions for cvittal?
17:30:33 <cvittal> no
17:30:45 <jzb> #link http://wiki.debian.org/MeetBot/
17:30:50 <jzb> more on using Meetbot there.
17:31:04 <jzb> OK, thanks cvittal!
17:31:20 <jzb> moving on - edison_ do you have anything to report or discuss?
17:31:26 <edison_> no #idea
17:31:38 <edison_> (smile)
17:31:39 <jzb> OK
17:31:47 <jzb> succinct. I like it.
17:31:58 <jzb> ewanmellor: have anything to report and/or discuss?
17:33:13 <topcloud> move on?
17:33:17 <jzb> yep
17:33:26 <jzb> OK, we'll circle back to ewanmellor later.
17:33:34 <jzb> inteq: do you have any items to report or discuss?
17:33:37 <ewanmellor> I've put a proposal out on the mailing list for a new release schedule.
17:33:47 <inteq> i have one item, hopefully it is relevant.
17:34:05 <inteq> i would like to start helping out with code reviews. i have registered at reviewboard: now what?
17:34:37 <jzb> that's a good question, are you a committer?
17:34:40 <inteq> i have also joined the right mailing lists , i believe.
17:34:43 <inteq> not yet , no
17:34:52 <inteq> i want to be.
17:34:57 <inteq> i will be an independent.
17:35:05 <jzb> ke4qqq: suggestions on how a non-contributor would help reviewing code, or is that possible?
17:35:19 <ke4qqq> inteq: start reviewing patches that come in. We'd welcome the help - also review the stuff hitting -commits
17:35:31 <ke4qqq> just because someone is a committer doesn't mean they write perfect code.
17:35:38 <ke4qqq> esp if the name attached is ke4qqq (smile)
17:35:40 * jzb is sohocked
17:35:45 <jzb> er, shocked
17:35:48 <inteq> ok. let me make sure i am on the right lists.
17:35:49 <inteq> hehe
17:35:54 <inteq> that is not nice ;')
17:36:17 <inteq> i am on cloudstack-dev and cloudstack-patches
17:36:21 <ke4qqq> cloudstack-dev@incubator.apache.org is the dev list
17:36:29 <ke4qqq> there's a cloudstack-patches?
17:36:36 <jlkinsel> no
17:36:41 <edison_> inteq: http://cloudstack.org/blog/151-contributing-to-apache-cloudstack-as-a-non-committer.html
17:36:43 <jzb> I'm guessing that's cloudstack-commits
17:36:59 <inteq> yes. sorry
17:37:03 <jzb> edison_: that covers sending patches, but not so much reviews.
17:37:05 <inteq> my filter is called cloudstack-patches
17:37:08 <inteq> sorry about that
17:37:12 <ke4qqq> ahhh - yeah cloudstack-commits is where all commits hit - but reply to -dev - 3x the number of eyeballs
17:37:18 <inteq> edison: thanks for the link
17:37:31 <jzb> #action I'll draft an addition to the non-committer guide for reviewing.
17:37:52 <ke4qqq> #undo
17:37:52 <cs-meeting> Removing item from minutes: <MeetBot.items.Action object at 0x97da14c>
17:38:05 <jzb> ke4qqq: no?
17:38:06 <ke4qqq> #action JZB will draft an addition to the non-committer guide for reviewing
17:38:23 <ke4qqq> meetbot doesn't preserve antecedent to that pronoun
17:38:30 <jzb> ah
17:38:47 <chipc> ^ - then my info at the beginning won't make sense in the minutes… or well...
17:38:54 <jzb> good to know.
17:39:06 <jzb> OK.
17:39:13 <jzb> inteq: any other items, or questions for inteq?
17:40:06 <jzb> OK - let's move on.
17:40:25 <jzb> #topic ewanmellor reporting on release schedule proposal
17:40:39 <jzb> ewanmellor: you wanted to discuss the updated release schedule proposal?
17:40:49 <ewanmellor> Yes,
17:40:56 <ewanmellor> It went out on the mailing list last night.
17:41:04 <jzb> #link http://markmail.org/thread/xk5d6aafiioadgb5
17:41:27 <ewanmellor> Basically, we said that we wanted all the legal / policy issues resolved before cutting our first release candidate, and we're not there.
17:41:58 <ewanmellor> I also think that we need to allow extra time for upgrade from 3.0.2 -> 4.0, because I presume that we would consider that a release blocker if it doesn't work.
17:42:09 <ewanmellor> And we need to add a week at the end for voting.
17:42:23 <ewanmellor> So my new proposal adds 3 weeks in total.
17:42:30 <ewanmellor> Two at the front, and one at the end.
17:42:35 <jzb> that gives us a release ~September 26th now.
17:42:48 <ewanmellor> Yes.
17:43:07 <chipc> ewanmellor: have you seen the a.o page that lays out a good release calendar for an incubating project?
17:43:13 <ewanmellor> Yes.
17:43:18 <ke4qqq> chipc: link?
17:43:24 <chipc> looking now
17:43:36 <inteq> jbz: no other topics. thanks
17:43:37 <ewanmellor> I don't think that we're very normal though. We are already 4.0, so we're mature in one sense.
17:43:41 <jlkinsel> that's still sorta aggressive - you're presuming only one release-candidate?
17:43:50 <jlkinsel> ah wait, 2 weeks worth
17:43:53 <ewanmellor> No, I'm presuming as many release candidates as we need.
17:43:59 <ewanmellor> Yes, two weeks of RCs.
17:44:18 <ewanmellor> It's 5.5 weeks from today.
17:44:37 <jzb> ewanmellor: do we have any plan for alpha or beta builds?
17:44:40 <ke4qqq> do we know who is going to be generating/signing the releases - thus far only 3 of us have keys in KEYS
17:44:41 <jzb> or is it straight to RC?
17:45:09 <ewanmellor> jzb: We're doing alpha builds every day.
17:45:40 <jlkinsel> I left alpha/beta out of the version docs purposefully - think we can add them later if we need them
17:45:46 <ewanmellor> There's no plan for a beta program, because people said that they wanted the policy issues to be the long pole, and as soon as those are resolved, we should release.
17:46:00 <jlkinsel> ke4qqq: the release plan says just chip and I
17:46:19 <ke4qqq> jlkinsel: awesome, ok, just as long as someone owns it
17:46:22 <ewanmellor> Can we do one issue at a time?
17:46:41 <ewanmellor> Are people OK with the release schedule as a whole?
17:46:46 * jlkinsel is
17:47:05 <topcloud> I think my problem with the release schedule is the number of patches coming in is showing the release is not yet stable
17:47:18 <ewanmellor> topcloud: It's definitely not stable.
17:47:31 <ewanmellor> topcloud: This gives us two more weeks to sort that out.
17:47:53 <topcloud> and also we just identify something for cvittal to do on iptables which is new.
17:48:03 <ewanmellor> topcloud: Do you think that two weeks is not enough?
17:48:12 <topcloud> ewanmellor: i'm agreeing with the extension but i'm worried it's not enough.
17:48:21 <topcloud> ewanmellor: yes.
17:49:10 <chipc> ewanmellor: suggestion - can you pull together a list of work, and then check in with the owners to get an estimate of the timeline for each?
17:49:12 <ewanmellor> Anyone else? I don't want us to add more weeks to the schedule if we can avoid it.
17:49:38 <chipc> seems like that would be the best way to set a schedule
17:49:41 <topcloud> i agree with chipc. i think we need a list of things to do.
17:49:51 <inteq> i also agree
17:49:55 <ewanmellor> https://cwiki.apache.org/confluence/display/CLOUDSTACK/CloudStack+4.0+Release+Checklist
17:50:07 <inteq> how is testing being handled to bake in each of the changes?
17:50:24 <inteq> i guess i am asking more about qa
17:50:28 <inteq> and automated testing
17:50:42 <ewanmellor> We have automated testing being set up on http://jenkins.cloudstack.org.
17:50:49 <ewanmellor> Unit tests are passing, FWIW.
17:50:52 <jzb> ewanmellor: what does that test, exactly?
17:50:59 <inteq> how many unit tests are there now?
17:51:01 <jzb> or generally - I suppose exactly might be a bit long
17:51:12 <ewanmellor> System testing is being set up next week.
17:51:13 <inteq> and how much coverage is there in what they test
17:51:31 <ewanmellor> We have 24 unit tests. Basically nothing.
17:51:54 <jzb> ewanmellor: so it can pass all tests and still not work on real systems.
17:51:57 <inteq> do you have enough confidence in that being enough to confirm stability
17:52:07 <jzb> sorry that was a question, not a statement.
17:52:12 <inteq> i certainly do not
17:52:12 <ke4qqq> inteq: A place that needs a lot of love, and for which we'd be eternally grateful for help (smile)
17:52:14 <ewanmellor> Citrix has ~500 functional tests that it's just preparing for submission. Those are much richer.
17:52:45 <ewanmellor> We still don't know the coverage precisely, but it's a significant chunk.
17:52:45 <chipc> ewanmellor - that rocks
17:52:46 <inteq> functional does not necessarily pair up with feature, does it?
17:52:52 <inteq> that is a lot of tests though
17:52:53 <inteq> for sure
17:52:55 <ke4qqq> ewanmellor: for marvin?
17:53:03 <ewanmellor> ke4qqq: Yes, for Marvin.
17:54:21 <inteq> i know nothing about marvin, but i am willing to help where i can for sure
17:54:29 <ewanmellor> #link http://wiki.cloudstack.org/display/QA/Testing+with+python
17:54:39 <inteq> we want OS to be as stable as possible and have confidence in the features that it provides
17:55:08 <chipc> ewanmellor: question for you - is there a date yet to be able to replace cloud-xenserver-5.6.100-1.jar with a download URL for the same with the new license?
17:55:40 <ewanmellor> chipc: No date. I need to beat those guys with a big stick.
17:56:05 <ewanmellor> We did find a bug in their latest SDK for them though, and that got resolved this week.
17:56:18 <chipc> ewanmellor: understood… I think that might be a release blocker though… should be factored into any timelines
17:56:31 <inteq> @ke4qqq: how can i get started assisting you?
17:56:40 <ke4qqq> chipc: indeed
17:56:46 <ke4qqq> inteq: lets talk after the meeting
17:56:50 <inteq> k
17:56:57 <jzb> OK, it sounds like we need to finish this on cloudstack-dev in the current thread that ewanmellor has started.
17:57:02 <ewanmellor> chipc: Yes. That's one of the unresolved legal issues. I've asked for 5.6.100 and 6.1 SDKs to be relicensed so that we can use either with CloudStack.
17:57:06 <ke4qqq> parts of CS that aren't HV related don't build without xsj
17:58:06 <chipc> jzb / ewanmellor - next topic or next person?
17:58:42 <jzb> do we have an action item on chipc's suggestion?
17:58:49 <topcloud> i looked through the list ewan sent out. I don't think it's fine grain enough.
17:59:01 <ewanmellor> topcloud: Contributions welcome!
17:59:05 <ke4qqq> topcloud: it's a wiki -patches welcome
17:59:34 <topcloud> but should we keep this on a wiki or should we enter them as bugs on jira?
17:59:35 <chipc> #action ewanmellor to ask Xen folks for an update on relicensing xsj
17:59:55 <ewanmellor> topcloud: We're getting a new Jira v. soon from Apache.
18:00:08 <ewanmellor> topcloud: I was going to put them all on there as soon as that is ready.
18:00:13 <topcloud> ic
18:00:23 <ewanmellor> topcloud: Next week, hopefully.
18:00:27 <topcloud> weanmeller: cool...thanks
18:00:32 <topcloud> sorry.
18:00:53 <jzb> ewanmellor: any other items?
18:00:56 * ke4qqq notes we are at an hour already
18:00:57 <ewanmellor> On the topic of automated testing – we're setting up a few machines inside a Citrix lab for us to use as a target system.
18:01:16 <ewanmellor> You will see new jobs arrive on Jenkins as soon as that is working.
18:01:26 <ke4qqq> ewanmellor: as jenkins slaves? or as direct targets for things like marvin?
18:01:27 <ewanmellor> There's a test-smoke-4.0 there already, but it doesn't yet work.
18:02:02 <ewanmellor> ke4qqq: I was thinking to have a Jenkins slave inside the lab, plus additional target machines.
18:02:25 <ke4qqq> ewanmellor: cool, that's a definite boon to testing things that aren't the simulator
18:02:30 <ewanmellor> The regression suite needs 4-8 physical machines (or nested hypervisor environments on really beefy machines).
18:02:52 <ewanmellor> Yeah, it should be a big step up.
18:03:23 <ewanmellor> That's all just FYI – there's not much help anyone else can give until the physical environment is working.
18:04:17 <ewanmellor> That's it from me.
18:04:25 <jzb> ewanmellor: thanks!
18:04:38 <jzb> OK, jlkinsel do you have any items?
18:04:52 <jlkinsel> I have nothing at this time
18:04:57 <jzb> jlkinsel: OK, thanks
18:05:23 <jzb> kelveny: any items?
18:05:36 <kelveny> not at this time
18:05:42 <jzb> OK, thanks
18:05:51 <jzb> Lu_: any items?
18:05:52 <cvittal> I have to leave sorry. will check back on action items
18:06:04 <jzb> cvittal: understood - thanks!
18:06:44 <jzb> since we're +1 hour already, I'm going to go quickly
18:06:48 <jzb> muddyMud: any items?
18:07:03 <muddyMud> no items
18:07:08 <jzb> thanks!
18:07:18 <jzb> techpubs: anything to report or discuss?
18:08:43 <jzb> topcloud: anything from you?
18:08:46 <topcloud> been merging the checkins, at the current rate, the release is not stable as discussed before. That's what I have to report.
18:09:12 <topcloud> As for discuss, my problem is the checkins come in with no reviewers and no unit testing.
18:09:13 <jzb> OK - any questions on that?
18:09:15 <chipc> topcloud: thank you for doing that painful job
18:09:21 <jzb> chipc: +1
18:09:41 <topcloud> it's a known problem with cloudstack today so i can't do much about it today. We need to change that for the next release.
18:09:47 <topcloud> that's all i have for now.
18:09:55 <jzb> topcloud: do we have any docs on creating unit tests?
18:09:57 <topcloud> chipc+jzb: thx
18:10:02 <jzb> anything to point people at ?
18:10:03 <topcloud> we do.
18:10:27 <topcloud> i'll send out something. There's really two ways. one is unit testing the components.
18:10:41 <topcloud> and one is using marvin to unit test and integrated cloudstack.
18:10:48 <topcloud> the second one i have nothing on.
18:10:59 <topcloud> but i can send the links to how to write unit tests for components.
18:11:04 <techpubs> jzb: sorry I glanced away for a second when you asked if I had any topics
18:11:06 <ke4qqq> #action topcloud to sent out a how to unit-test for components
18:11:29 <inteq> ok, so it seems like we will need to build out more robust testing around features as well?
18:11:31 <jzb> topcloud: anything further?
18:11:33 <inteq> ot am i wrong about that
18:11:35 <inteq> or*
18:11:42 <techpubs> I do have topics but I'm not sure we need to cover them in irc esp. when we're already at 1 hour
18:11:48 <topcloud> inteq: absolutely correct
18:12:29 <inteq> topcloud:ok
18:13:08 <jzb> topcloud: all good?
18:13:25 <topcloud> i'm good.
18:13:34 <jzb> techpubs: if you have topics we might as well cover them now. If anybody needs to leave, they can of course.
18:13:46 <jzb> we'll have the logs and such for them + to send to the mailing list.
18:14:52 <inteq> hate to ask yet one more question: where is the coding style guide for cloudstack?
18:15:06 <ke4qqq> inteq: on docs.cloudstack.org - search for coding conventions
18:15:15 <inteq> @ke4qqq:thanks
18:15:35 <ke4qqq> inteq: http://docs.cloudstack.org/CloudStack_Documentation/Design_Documents/Coding_Conventions
18:16:00 <inteq> are the expiration warnings normal?
18:16:02 <jzb> techpubs: going once, twice...
18:16:08 <jzb> inteq: they are not.
18:16:16 <ke4qqq> inteq: we've been discussing it on list......and no they aren't
18:16:22 <techpubs> which expiration warnings?
18:16:33 <jzb> techpubs: look at docs.cloudstack.org
18:16:42 <inteq> 'This site has expired and will be disabled-'
18:16:42 <techpubs> The site is not about to expire.
18:16:43 <jzb> the MT license is nearly expired.
18:16:48 <techpubs> The warning is in error.
18:16:52 <techpubs> I have been on this.
18:17:11 <jzb> OK. Well, the warning says that it's about to expire.
18:17:15 <techpubs> Received assurances yesterday that the license has been properly renewed.
18:17:46 <jzb> I just noticed it this morning myself.
18:17:49 <jzb> OK.
18:17:52 <inteq> OK.
18:18:06 <inteq> well, i know where the docs are now at any rate. so , thanks ;')
18:18:10 <jzb> techpubs: aside from mindtouch licensing goodness - anything you wanted to report?
18:18:40 <techpubs> well, the topic du jour is of course whether to take down this site in favor of something else proposed by ke4qqq
18:18:47 <techpubs> do we want to discuss that?
18:18:58 <ke4qqq> we can, but we can't make any decisions here
18:19:09 <jzb> we can - I've already responded on list, but I'm in favor of the Publican-generated docs, myself.
18:19:53 <ke4qqq> /win 28
18:19:55 <techpubs> ke4qqq: why are you proposing to do away with the site we've worked on?
18:20:30 <techpubs> The docs will still be generated by publican.
18:21:08 <ke4qqq> we have a preponderance of things to maintain, and don't even have our project site in decent shape. Plus MT is not free - it is quite expensive - which means it potentially endangers the independence of the project (e.g. if/when citrix decides to stop footing the bill who will pick it up?)
18:21:30 <topcloud> i have to run as well. good to meet everyone on irc!
18:21:40 <ke4qqq> nice to see you here topcloud, please come often
18:21:41 <jzb> topcloud: thanks!
18:21:44 <inteq> topcloud:definitely. s
18:21:51 <inteq> see you later*
18:22:04 <techpubs> If/when? We don't need to make decisions now based on hypotheticals.
18:22:15 <techpubs> We could take down the site then if need be.
18:22:42 <techpubs> I'm sorry but I am very firm about looking at this from the point of view of the users.
18:22:49 <topcloud> sorry...b4 i run off...please consider making this a weekly meeting until release is out...very helpful meeting...thanks jzb for running it.
18:23:09 <jzb> topcloud: good to know - I was going to ask whether people found it helpful.
18:23:16 <inteq> topcloud:+1
18:23:17 <ke4qqq> we could - but fwiu - publishing to mindtouch requires code that isn't yet written, and you have a ui designer on standby - all things that seem to be unneeded if we just publish using straight publican.
18:23:21 <jlkinsel> topcloud: +1
18:24:04 <techpubs> It is worth some effort to create an experience for the users that is much better than, excuse me, the electronic paper you're proposing
18:24:54 <ke4qqq> well - we can come to a decision on list - we can't do that here, so lets move it back there.
18:25:13 <techpubs> Thanks.
18:25:47 <jzb> techpubs: any other things you wanted to discuss?
18:25:56 <jzb> or report
18:26:25 <inteq> oh - the 'list' is the cloudstack-dev mailing list, correct?
18:26:30 <jzb> inteq: yes
18:26:47 <inteq> jbz:thanks
18:26:53 <jzb> welcome mrhinkle
18:27:08 <ke4qqq> hi mrhinkle
18:27:08 <mrhinkle> jzb: thanx
18:27:16 <jzb> #chair mrhinkle
18:27:16 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor inteq jlkinsel jzb ke4qqq kelveny mrhinkle techpubs topcloud
18:27:28 <chipc> jzb ke4qqq - be good, your boss is here now
18:27:33 <jlkinsel> ha
18:27:35 <ke4qqq> chipc: (smile)
18:27:38 <ke4qqq> widodh: welcome
18:27:42 <ke4qqq> #chair widodh
18:27:42 <cs-meeting> Current chairs: chipc cvittal edison_ ewanmellor inteq jlkinsel jzb ke4qqq kelveny mrhinkle techpubs topcloud widodh
18:27:55 <widodh> It's not time yet, is it?
18:28:03 <mrhinkle> chipc you don't know how it works, ke4qqq calls all the shots and I do what he tells me
18:28:07 <jlkinsel> it was time 90 mins ago (tongue)
18:28:14 <widodh> Seriously?
18:28:15 <jzb> widodh: what jlkinsel said (smile)
18:28:43 <jzb> techpubs: how can we help in porting the existing docs to Publican?
18:28:47 <widodh> I must be confused, I used some time calculator and it said 19:00 UTC was 21:00 my time
18:29:04 <ke4qqq> widodh: in CET?
18:29:16 <jzb> widodh: that may be, but the meeting was set for 17:00 UTC
18:29:17 <techpubs> jzb: First step is to look at the Install Guide and say which sections to NOT port. See my note on list
18:29:20 <widodh> Yes, but it must have been the summertime
18:29:34 <inteq> mrhinkle: did you give the OS talk at the atl cloudstack meetup this week?
18:30:08 <jzb> techpubs: what about the other guides while we're waiting on that?
18:30:15 <jzb> or do the other guides depend on the install guide?
18:30:35 <mrhinkle> inteq: that was ke4qqq
18:30:39 <inteq> ah, ok
18:30:46 <inteq> and i meant CS , not OS
18:30:46 <techpubs> jzb: The other guides are all done afaik. I mean developer's and admin guides.
18:30:48 <inteq> my bad
18:31:27 <inteq> ke4qqq:nice talk. very informative and peaked my interest in helping out.
18:31:29 <jzb> for reals? I didn't realize that. I thought there was still stuff to do.
18:31:31 <widodh> But, sorry, it was my intention to be here on time
18:31:46 <jzb> that's pretty awesome if so.
18:32:07 <jzb> OK.
18:32:14 <jzb> mrhinkle: anything you wanted to discuss?
18:32:24 <ke4qqq> inteq: I'll pay you the promised money later for saying that in front of my boss (smile)
18:32:52 <inteq> ke4qqq: indeed (twists mustache)
18:33:34 <techpubs_> jzb: please repeat, I got kicked out temporarily
18:33:49 <jzb> techpubs_: sure.
18:34:12 <jzb> I just said that I thought there were sections of the admin / developer guides that needed to be ported.
18:34:13 <techpubs_> jzb: sorry, typo in my nick. It's still me. (smile)
18:34:29 <jzb> techpubs_: but if it's down to the install guide only, that's awesome.
18:34:35 <ke4qqq> i don't see much that needs to be removed from install - but plenty that needs to be added
18:34:46 <techpubs_> jzb: if you find unported sections, file a doc bug! Those should have all been done by now.
18:34:48 <widodh> Regarding the install, I'm working on that
18:35:04 <widodh> in /docs. Since cloud-setup-agent is highly unreliable
18:35:27 <mrhinkle> jzb: I have no agenda, just waiting for ke4qqq and you to tell me what I am doing (wink)
18:35:33 <jzb> widodh: no components of CloudStack are unreliable. We prefer "interesting."
18:35:45 <inteq> feature-rich
18:35:57 <widodh> good choice of words (smile)
18:36:04 <widodh> See commit b6a610e2db5798ae69cfd2cc8a9d7f85617f3926
18:36:05 <jzb> (The goal, of course, is for everything to be "boring")
18:36:14 <ke4qqq> jzb: or challenging (smile)
18:36:19 <jzb> indeed.
18:36:33 <widodh> Something I'll throw on the mailinglist, but maybe is there is some time to talk about
18:36:56 <widodh> how do you feel about cloud-setup-agent modifying firewalls, changing libvirt configuration, touching network config
18:37:36 <ke4qqq> scary, but perhaps better than having people do it by hand.....my concern is that it would be different from distro to distro
18:37:53 <ke4qqq> widodh: I'd almost rather we break that out into puppet or chef
18:38:03 <jzb> ke4qqq: I was just about to say
18:38:12 <jzb> isn't that a bit of what you've been trying to do with Puppet?
18:38:12 <widodh> ke4qqq: That's my point. From Ubuntu 10.04 to 12.04 some config things changed regarding libvirt
18:38:15 <ke4qqq> document how to do it, but provide them an out with a puppet module or chef recipe
18:38:30 <widodh> ke4qqq: I started documenting it, steps to configure a hypervisor
18:38:31 <jlkinsel> it'd be great if there were advanced instructions on how to do it without having fw automatically modified. I found that obnoxious.
18:38:50 <ke4qqq> jlkinsel: me too - same thing with the installer.
18:39:06 <widodh> I personally (again, will go on ml), think cloud-setup-agent should only generate a agent.properties for you
18:39:06 <jzb> I kind of wish we just had an appliance to spin onto machines.
18:39:10 <ke4qqq> it's nice that it makes it simple, but I promise I know how to use yum/apt.
18:39:13 <inteq> so does dev-cloud not do a good job of this?
18:39:32 <widodh> And complain if stuff is missing
18:39:49 <ke4qqq> inteq: devcloud does a good job of giving you a dev test bed, but not for production use.
18:40:04 <inteq> ke4qqq:ah, i see
18:40:05 <jlkinsel> jzb: sounds like somebody could do a cloudstack "distro"
18:40:10 <chipc> windoh: to me, step 1 is to document what it's doing
18:40:21 <ke4qqq> jzb: ohhh - suse studio (smile)
18:40:24 <techpubs_> I'll leave now if no one has any questions on docs that need urgent discussion here
18:40:32 <chipc> s/windoh/widoh
18:40:35 <jlkinsel> thx techpubs_
18:40:40 <jzb> thanks techpubs_
18:40:45 <inteq> techpubs:great to meet you.
18:41:05 <techpubs_> inteq: does this count as meeting? (smile)
18:41:13 <inteq> heh
18:41:38 <chipc> widodh: after it'd done being doc'ed, let'd think about puppet install stuff… I, for one, want to tie it into my PXE boot installation process for XenServer
18:41:50 <ke4qqq> for here it does.
18:42:02 <ke4qqq> chipc: does your PXE stuff for XS handle hotfixes/patches?
18:42:09 <jlkinsel> I'd love to see puppet install stuff, happy to work on that
18:42:13 * ke4qqq wonders if i have asked you this before
18:42:24 <chipc> ke4qqq, not yet, since I use advanced networking - and you have asked
18:42:26 <ke4qqq> jlkinsel: http://github.com/ke4qqq/puppet-cloudstack
18:42:31 <chipc> but it's VERY easy to have a post install script hook
18:42:43 <jlkinsel> yeah I've looked at that before briefly
18:43:03 <jlkinsel> actually somebody was looking for similar on #cloudstack this week
18:43:11 <ke4qqq> yeah, kekec
18:43:25 * ke4qqq wonders if we can get this over before we go over 2 hours?
18:43:43 <jlkinsel> yeah. focus.
18:44:18 <jzb> widodh: did you have anything else to report?
18:45:26 <widodh> jzb: No, I'm working on the documentation mainly and still hunting a small RBD bug
18:45:50 <jzb> widodh: OK, thanks - and happy hunting!
18:46:08 <jzb> unless anyone else has something urgent, I'll close out the meeting with one final note.
18:46:31 <jzb> That is - it's been suggested we make this weekly. I'm in favor, but I'll be on a plane at this time next week.
18:46:46 <jzb> so we'd need someone else to chair next week. (I'm happy to do it regularly, though.)
18:46:56 <jlkinsel> I can cover for you
18:47:05 <jlkinsel> or if ke4qqq wants to
18:47:05 <ke4qqq> awesome
18:47:18 <ke4qqq> no - I'd far rather you run it.
18:47:29 <jzb> jlkinsel: as sad as ke4qqq will be at not chairing, I think you're on the hook now. :-0
18:47:33 <jzb> er, (smile)
18:47:39 <widodh> Every friday at this time?
18:47:42 <jlkinsel> yes
18:47:43 <ke4qqq> widodh: no
18:47:45 <jlkinsel> no?
18:47:46 <ke4qqq> at the right time (smile)
18:47:52 <jlkinsel> yes (wink)
18:47:52 <ke4qqq> not 90 mins late (smile)
18:48:03 <jlkinsel> next time you're late you have to bring donuts.
18:48:08 <widodh> ke4qqq: get it (wink) That is however going to be tight for me.
18:48:17 <widodh> Friday evening 19:00
18:48:29 <ke4qqq> well if this any inkling, it should run long
18:48:37 <jzb> widodh brings up a good point, we should probably rotate times
18:48:51 <jlkinsel> i'm open to that
18:49:01 <jzb> I'll bring that up for discussion on the list, OK?
18:49:05 <ke4qqq> proposer of time has to run the meeting (smile)
18:49:08 <widodh> jzb: Yes, please
18:49:09 <jlkinsel> jzb: yes
18:49:13 <jzb> groovy
18:49:22 <jzb> OK, I think that's a good first meeting. Thanks everybody!
18:49:28 <jlkinsel> thx guys
18:49:31 <ke4qqq> thanks for running it jzb
18:49:35 <jzb> #endmeeting

  • No labels