21:02:17 <mattgriffin> #startmeeting HA-Guide
21:02:18 <openstack> Meeting started Thu Apr  9 21:02:17 2015 UTC and is due to finish in 60 minutes.  The chair is mattgriffin. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:02:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:02:21 <openstack> The meeting name has been set to 'ha_guide'
21:02:37 <megm> So is there anything worth discussing?  I was looking forward to Nick's report about his discussion with Bogdan...
21:02:44 <mattgriffin> megm, i have a few agenda items at https://wiki.openstack.org/wiki/Documentation/HA_Guide_Update#Next_Meeting
21:02:55 <mattgriffin> we can skip the first two
21:03:14 <mattgriffin> and jump to 3. ok?
21:03:33 <megm> Ok by me
21:03:35 <mattgriffin> #topic add priorities
21:04:05 <mattgriffin> i reviewed Sam-I-Am's priorities and I'm +1. they look good enough to get started
21:04:15 <mattgriffin> #link https://wiki.openstack.org/wiki/HAGuideImprovements/TOC
21:04:17 <megm> I sent email with a couple of concerns but got no response
21:05:00 <mattgriffin> megm, what problems did you see?
21:05:36 <nickchase> hey, all, sorry I'm late.
21:05:58 <mattgriffin> hey nickchase
21:06:01 <megm> I was concerned that MySQL, RabbitMQ, and Keystone are not priority 1 -- if we agree that priority 1-3 must be done to proceed, that's not an issue
21:06:13 <megm> Hi, nickchase!
21:06:22 <mattgriffin> we were discussing the priorities that Sam-I-Am added to the TOC - https://wiki.openstack.org/wiki/Documentation/HA_Guide_Update#Next_Meeting
21:06:28 <mattgriffin> nickchase, ^
21:06:44 <mattgriffin> megm, good point
21:07:42 <mattgriffin> need to get that clarified
21:07:48 * nickchase is having trouble finding the TOC....
21:08:07 <megm> https://wiki.openstack.org/wiki/HAGuideImprovements/TOC
21:08:23 <nickchase> thx
21:09:00 <megm> Also not sure why Swift, Glance, and Cinder are prios 4,5,6 respectively and not the same priority, but that may just be because mentally I group them together under "storage"
21:09:32 <megm> And Nova compute (which includes ephemeral storage, I suppose) is Priority 6
21:10:18 <mattgriffin> keep in mind that you need to do more than P1 in order to get to HA... maybe "Priority" is the wrong word. more like step
21:10:55 <mattgriffin> maybe that was Sam-I-Am's thinking
21:11:05 <mattgriffin> about not making P2s actually P1
21:11:13 <megm> Yes, it looks like sam-i-am was heavily thinking of the order in which things should be written, and some of the priorities are not 1 because they have dependencies on other pieces
21:11:31 <mattgriffin> ah
21:11:49 <megm> As I say, I just have questions -- wanted to make sure these made sense to everyone else
21:11:55 <mattgriffin> ack
21:12:13 <megm> I'm hoping we'll have some parallel development here
21:12:14 <mattgriffin> megm, do you think your concerns would prevent us from moving forward?
21:12:35 <mattgriffin> 3 steps forward, 1 step back
21:12:38 <mattgriffin> more like
21:12:48 <nickchase> long as it's not 1 step forward, 3 steps back....
21:12:51 <megm> No, definitely nothing to stop us from moving forward
21:12:53 <mattgriffin> heh
21:13:00 <mattgriffin> cool
21:13:13 <mattgriffin> let's hit the first two items on the agenda and get updates from nickchase
21:13:19 <megm> It clarifies that Heat, Ceilometer, Trove, and Sahara are a little bit deferred
21:13:25 <mattgriffin> :)
21:13:28 <mattgriffin> #agenda items
21:13:29 <megm> Yes, I want to hear from nick!
21:13:32 <mattgriffin> #topic agenda items
21:13:47 <mattgriffin> #link https://wiki.openstack.org/wiki/Documentation/HA_Guide_Update#Next_Meeting
21:13:55 <mattgriffin> nickchase, ^
21:14:29 <nickchase> Well, I think that we're probably good; I'm trying really hard to meet with Bogdan to finalize but we've been playing meeting tag for 2 weeks. :(
21:14:37 <nickchase> So here's my feeling:
21:15:23 <nickchase> I think that we are probably good to start while we get the spec approved -- I'm trying to go ahead and get it set up so taht Bogdan can simply approve or leave comments.  The big question is...
21:15:45 <nickchase> how to get content while we're waiting for the spec approval, since Andreas is insisting on one before we start.
21:16:01 <nickchase> (I did get the initial framework in, but Andreas -2'd it because we don't have a spec.)
21:17:12 <mattgriffin> nickchase, where's the spec process documented?
21:17:18 <megm> So what exactly is required to get a spec?  I thought the doc that the blueprint points to made a good spec
21:17:23 <nickchase> I was actually about to ask that. :)
21:17:34 <nickchase> I'm looking for it right now.
21:17:40 <mattgriffin> heh... i think i saw an email from Anne about it recently
21:18:00 <nickchase> let me just ask her
21:18:05 <mattgriffin> ok
21:18:33 <nickchase> ok, she's not onlne
21:18:34 <megm> April 6 mail
21:18:38 <nickchase> ah, great.
21:19:15 <megm> No, those links are about PTL
21:19:46 <mattgriffin> http://lists.openstack.org/pipermail/openstack-docs/2015-April/006218.html ?
21:19:52 <mattgriffin> doh
21:19:55 <megm> Just this sentence:  Please write a spec and link it to a blueprint, get that spec and blueprint approved before proposing the patch.
21:20:05 <megm> We have a blue print that points to the spec...
21:20:33 <megm> We may need to put the spec under git -- does that mean code it in RST?
21:20:53 <mattgriffin> here's andreas' $0.02 - http://lists.openstack.org/pipermail/openstack-docs/2015-April/006220.html
21:22:27 <nickchase> Do we have a blueprint?
21:23:07 <nickchase> (nm, I see that (doh))
21:23:22 <mattgriffin> please link
21:23:44 <mattgriffin> nickchase, ^
21:23:59 <nickchase> can't find it
21:24:02 <nickchase> (the spec)
21:24:38 <mattgriffin> nickchase, is it in launchpad or git.openstack.org?
21:24:44 <mattgriffin> the spec, that is
21:25:27 <nickchase> I'm looking for it in luanchpad and not finding it
21:26:32 <megm_> nick, what is "it" that you seek in lp?
21:28:04 <megm_> blueprint at https://blueprints.launchpad.net/openstack-manuals/+spec/improve-ha-guide
21:29:21 <nickchase> that is, indeed, the "it".
21:29:23 <nickchase> thank you.
21:29:24 <mattgriffin> :
21:29:25 <mattgriffin> :)
21:29:42 <mattgriffin> cool. let's go from there...
21:30:04 <megm_> I'm adding the bp link to the "spec"
21:30:04 <mattgriffin> so based on https://wiki.openstack.org/wiki/Blueprints#Blueprints_and_Specs, is Docs using "Spec + Blueprints" or "Blueprints only?"
21:30:56 <nickchase> We've been "Blueprints only" but I think they're pushing towards "Spec + Blueprints".
21:31:18 <mattgriffin> megm_, thnaks
21:31:22 <megm_> Should we revise/update the BP text
21:31:48 <megm_> ?
21:31:49 <nickchase> I think that they're looking for a spec in git that's been approved.
21:32:23 <megm_> More than networking has changed, we want to leverage the new work being done for install-guide and network-guide
21:33:05 <mattgriffin> megm_, yes. we should update the description of the LP BP - https://blueprints.launchpad.net/openstack-manuals/+spec/improve-ha-guide
21:33:41 <mattgriffin> megm_, the "Strategy and assumptions" content should work
21:33:47 <mattgriffin> from the TOC doc
21:34:42 <megm_> Add that to the BP?  Sounds like a plan -- compress the exchange about A/A and A/P into a statement of what we are doing...
21:35:04 <mattgriffin> megm_, please
21:35:07 <megm_> editing now
21:35:11 <nickchase> I don't think they're going to be satisfied with that.
21:35:21 <nickchase> I think we are going to have to add the whole TOC in there.
21:35:22 <mattgriffin> nickchase, megm_ i can write up a spec according to https://wiki.openstack.org/wiki/Blueprints#Blueprints_and_Specs
21:35:27 <mattgriffin> in "Spec + Blueprints lifecycle"
21:35:29 <nickchase> Great.
21:36:03 <mattgriffin> #action megm_ update the Launchpad BP https://blueprints.launchpad.net/openstack-manuals/+spec/improve-ha-guide
21:36:51 <megm_> action: done, I think.
21:36:56 <mattgriffin> #action mattgriffin write a spec according to https://wiki.openstack.org/wiki/Blueprints#Spec_.2B_Blueprints_lifecycle
21:37:36 <nickchase> great
21:37:43 <megm_> Terrific!
21:37:52 <mattgriffin> ok. so nickchase : do we still need the two action items for you at https://wiki.openstack.org/wiki/Documentation/HA_Guide_Update#Next_Meeting?
21:37:56 <nickchase> as soon as you have it, fi yuc ould let me know so I can push Bogdan to approve it that would be great.
21:38:06 <mattgriffin> nickchase, ok
21:38:58 <mattgriffin> i had another item on the agenda... just to get you thinking about what's next...
21:39:35 <mattgriffin> please take a look at https://etherpad.openstack.org/p/openstack-haguide-update-next-steps and add your thoughts on how we should proceed after everything is approved and we are ready to write
21:39:38 <megm_> Should the spec mention repos?  And, if is going public, perhaps a brief word about what the priority markers mean?
21:40:04 <mattgriffin> megm_, not sure
21:40:13 <mattgriffin> re: repos
21:40:35 <mattgriffin> i'll include a bit about the priorities (and that they might change or move in parallel as we get into things)
21:40:43 <megm_> How about "emphasize P1-3"?  That gets us around the MySQL/Keystone/RabbitMQ issues
21:41:03 <mattgriffin> fair enough
21:41:38 <megm_> This gives no statement about existing material that needs to be updated...  Earlier discussions said that would also be P1
21:42:14 <mattgriffin> megm_, this = the pad?
21:42:26 <megm_> Is there a way to start the push for Ceilometer/MongoDB, Sahara, Trove, etc to get HA info into their docs?
21:42:48 <megm_> Yes, the epad spec with priorities...
21:43:36 <megm_> One approach is to just convert that doc, put it in the right place, and mark it as not updated...  It preserves text without distracting us from the main plan
21:43:42 <megm_> But would it fly?
21:45:40 <megm_> Do we have a plan for testing the content after it is written?
21:46:03 <mattgriffin> megm_, i think that would work. the etherpad was just a scratch pad for discussion on how we see the needed activities to be completed ... if we agree that after the spec is approved that we start writing according to the Priorities then we're good
21:46:37 <mattgriffin> megm_, re: testing, not sure how that's done... nickchase?
21:47:04 <mattgriffin> megm_, testing that it's formatted correctly or assessing the content itself?
21:47:24 <megm_> Testing this doc is a non-trivial enterprise...  To ensure that, at each step, the preliminary information has been covered, et cetera
21:47:55 <megm_> Maybe put it as a blocker for now?  Without testing, we have a good theoretical doc that may still be valuable...
21:48:35 <mattgriffin> megm_, yes. should look into what other Docs teams do and use reviews
21:49:02 <mattgriffin> ok. i think that's all for now. anything else?
21:50:13 <mattgriffin> action items will be in the meeting minutes soon to be available on the wiki
21:50:26 <mattgriffin> if there's nothing else i'm going to end the meeting...
21:50:42 <megm_> I just mucked with your CRUDE PROECT PLAN
21:50:49 <mattgriffin> megm_, cool :)
21:50:55 <megm_> Delete/fix as you like
21:51:03 <mattgriffin> ah excellent
21:51:06 <mattgriffin> thanks
21:51:18 <mattgriffin> ok. bye megm_ , nickchase
21:51:20 <nickchase> sorry, got called away for a moment
21:51:25 <mattgriffin> nickchase, hi
21:51:26 <nickchase> bye guys
21:51:27 <nickchase> thanks
21:51:31 <megm_> I may muck some more...  But I'm good with ending the meeting
21:51:31 <mattgriffin> #endmeeting