Wednesday, 2014-01-15

« Tuesday, 2014-01-14 Index Thursday, 2014-01-16 »
*** CaptTofu has quit IRC00:08
*** CaptTofu has joined #openstack-dns00:08
*** CaptTofu has quit IRC00:08
*** CaptTofu has joined #openstack-dns00:26
*** CaptTofu has quit IRC01:36
*** CaptTofu has joined #openstack-dns01:37
*** CaptTofu has quit IRC01:37
*** nosnos has joined #openstack-dns01:38
*** jmcbride has joined #openstack-dns02:51
*** HenryG has quit IRC03:09
*** CaptTofu has joined #openstack-dns03:40
*** jmcbride has quit IRC03:52
*** CaptTofu has quit IRC04:36
*** CaptTofu has joined #openstack-dns04:37
*** CaptTofu has quit IRC04:37
*** ctracey|away is now known as ctracey04:41
*** vinod has joined #openstack-dns04:43
*** vinod has quit IRC05:38
*** ctracey is now known as ctracey|away08:09
openstackgerritEndre Karlson proposed a change to stackforge/designate: FloatingIP PTR record functionality  https://review.openstack.org/5831609:17
ekarlsokiall: / mugsie there ya go09:17
openstackgerritEndre Karlson proposed a change to stackforge/designate: FloatingIP PTR record functionality  https://review.openstack.org/5831609:51
ekarlsoforgotten to expand 1 of the tests in the previous09:53
*** cflmarques has joined #openstack-dns11:43
*** nosnos has quit IRC12:59
*** pasquier-s has quit IRC13:11
*** pasquier-s has joined #openstack-dns13:11
*** cflmarques has quit IRC13:16
mugsieartom: i think what puck said is pretty ok... as long as you aren't mean about it13:20
*** vinod has joined #openstack-dns13:33
kiallartom / mugsie / puck: It depends, If the spelling mistake is something end users would see, I would -1.13:37
kiallIf it's internal, or a code comment etc, I wouldn't -1, I would leave a review saying "If you do another patchset, please correct <mistake>"13:38
kiallAnd  -1 for anything thats an "API" - internal or not - e.g. method names13:38
*** vinod has quit IRC13:40
*** cflmarques has joined #openstack-dns14:12
*** artom has quit IRC14:31
*** artom has joined #openstack-dns14:38
*** vinod has joined #openstack-dns14:38
*** jmcbride has joined #openstack-dns15:02
*** jmcbride has quit IRC15:06
*** jmcbride has joined #openstack-dns15:16
openstackgerritA change was merged to stackforge/designate: Moved Limits API endpoint to a RESTController  https://review.openstack.org/6659915:16
*** betsy has joined #openstack-dns15:17
openstackgerritEndre Karlson proposed a change to stackforge/designate: FloatingIP PTR record functionality  https://review.openstack.org/5831615:21
openstackgerritKiall Mac Innes proposed a change to stackforge/designate: Ensure APIv2 list respose formats are correct  https://review.openstack.org/6687015:30
*** vinod has quit IRC15:38
*** jmcbride has quit IRC15:38
*** betsy has quit IRC15:38
*** vinod has joined #openstack-dns15:38
*** betsy has joined #openstack-dns15:39
*** vinod has quit IRC15:42
*** betsy has quit IRC15:43
*** CaptTofu has joined #openstack-dns16:16
mugsiehey, reminder that we have weeking meeting in 10 mins16:50
*** sballe has joined #openstack-dns16:56
*** jmcbride has joined #openstack-dns16:57
*** sballe has quit IRC16:57
*** sballe has joined #openstack-dns16:57
*** tsimmons has joined #openstack-dns16:58
*** vinod has joined #openstack-dns16:59
kiallmugsie: still on this call, can you start the meet for me17:00
*** vinod has quit IRC17:21
kialltsimmons: heya17:23
tsimmonskiall: hey17:24
tsimmonsSo17:24
tsimmonsI remember you said it would be pretty straightforward to convert the interactions with storage to proper transactions. But I looked into a bit and I wanted to see what your thoughts were on what I kind of thought up.17:24
kiallWell - I thought I said "straight forward in theory" or something along those lines :D There's always an edge case or 1217:25
tsimmonshah, right.17:26
tsimmonsI was thinking of implementing the begin, commit, and rollback functions in storage/impl_sqlalchemy/__init.py__ using the session that is retrieved there.17:26
tsimmonsUsing something like the logic in here http://docs.sqlalchemy.org/en/latest/orm/session.html#id117:26
tsimmons(Scroll down a bit)17:26
kiallYea - that looks like what I'd have expected ..17:28
tsimmonsI've never worked with SQLA so this stuff is a bit foreign to me.17:28
kiallSQLA is .. a learning curve alright :)17:29
tsimmonsSo I'm unsure on how to inform the "session" of the changes made by a call to self.storage.get_server() so that it could roll them back properly in the case that something goes wrong when it sends it back to central to process in the backend etc17:29
*** tsimmons1 has joined #openstack-dns17:29
*** tsimmons has quit IRC17:29
tsimmons1ugh sorry, internet dropped.17:30
kiallSo, we don't actually need to tell the session what changed, the session tracks that all for us17:30
kiallIt's a little indirect, but all DB queries run via that session object.. So calling begin / commit / rollback should "just work"17:30
kiallWell - All DB interactions, for a single DB17:31
kialle.g. the designate and powerdns (if you use it) DB's each have their own session17:31
tsimmons1Hm. So theoretically the functions in __init could just be self.session.begin() etc ?17:32
kialla small example would be the create_quota metod17:32
kiallmethod*17:32
kiallhttps://github.com/stackforge/designate/blob/master/designate/storage/impl_sqlalchemy/__init__.py#L13217:32
kiallself.session get's passed to the models save method.. which is:17:32
kiallhttps://github.com/stackforge/designate/blob/master/designate/sqlalchemy/models.py#L2817:33
kiallWhere we apply the necessary changes to the session17:33
tsimmons1Ok. That's what I figured was happening there.17:33
kiallin storage/impl_sqlalchemy/__init__.py, we could end up with 3 new methods for begin / commit / rollback where they literally just call self.session.begin() etc17:34
tsimmons1So by calling commit() and rollback() even if it's not in that explicit place in the code, should recognize the things that have happened and roll them back should we ask.17:34
kiallThen , then storage/api.py class with make use of those, rather than attempting to manually restore the previous state17:34
kialltsimmons1: exactly :)17:34
tsimmons1Ok. That's pretty much what I was thinking, but I wanted to run it by you so that I didn't come out with something totally stupid.17:35
kialllol :)17:36
kiallIt's sounds like you're planning to do it exactly how I've got it in my head.. So, either we're both stupid or not ;)17:36
tsimmons1Hah, nah I just wanted to make sure that I wasn't doing something totally unnecessary or arduous. I think if you're stupid we're all screwed :P17:37
tsimmons1I'll get working on that here in the next few days and get something out there for folks to look at.17:38
kiallCool :)17:40
tsimmons1Thanks for your help, as always. Go enjoy your evening!17:40
*** jmcbride has quit IRC17:41
*** tsimmons1 has quit IRC17:43
*** jmcbride has joined #openstack-dns17:52
*** jmcbride has quit IRC17:57
*** tsimmons has joined #openstack-dns18:01
*** ctracey|away is now known as ctracey18:16
openstackgerritSascha Peilicke proposed a change to stackforge/designate: Sync with global requirements  https://review.openstack.org/6690218:19
openstackgerritA change was merged to stackforge/designate: FloatingIP PTR record functionality  https://review.openstack.org/5831618:21
*** openstackgerrit has quit IRC18:27
*** openstackgerrit has joined #openstack-dns18:27
*** ChanServ sets mode: +v openstackgerrit18:27
openstackgerritKiall Mac Innes proposed a change to stackforge/designate: Ensure APIv2 list respose formats are correct  https://review.openstack.org/6687018:29
*** jmcbride has joined #openstack-dns18:37
*** jmcbride has quit IRC18:42
*** jmcbride has joined #openstack-dns18:43
*** jmcbride has quit IRC18:43
*** cflmarques has quit IRC18:46
*** tsimmons has quit IRC18:59
openstackgerritKiall Mac Innes proposed a change to stackforge/designate: Ensure APIv2 list respose formats are correct  https://review.openstack.org/6687019:06
*** jmcbride has joined #openstack-dns19:26
*** jmcbride has quit IRC19:30
*** jmcbride has joined #openstack-dns19:33
*** vinod has joined #openstack-dns19:34
*** jmcbride has quit IRC19:40
*** sballe_ has joined #openstack-dns19:47
*** sballe has quit IRC19:48
*** jorgem has joined #openstack-dns19:55
openstackgerritA change was merged to stackforge/designate: Ensure APIv2 list respose formats are correct  https://review.openstack.org/6687019:56
*** vipul is now known as vipul-away19:57
*** vipul-away is now known as vipul19:57
*** vinod has quit IRC20:06
*** vipul is now known as vipul-away20:13
*** jmcbride has joined #openstack-dns20:15
*** jmcbride has quit IRC20:35
*** vinod has joined #openstack-dns20:38
*** jmcbride has joined #openstack-dns20:43
*** vipul-away is now known as vipul20:49
*** HenryG has joined #openstack-dns20:52
*** vinod has quit IRC21:00
*** CaptTofu has quit IRC21:10
*** vinod has joined #openstack-dns21:14
*** tsimmons has joined #openstack-dns21:16
*** tsimmons has left #openstack-dns21:19
*** richm has joined #openstack-dns21:31
*** vinod has quit IRC21:36
*** jmcbride has quit IRC21:40
*** jmcbride has joined #openstack-dns21:54
*** vinod has joined #openstack-dns21:57
*** vinod has quit IRC22:01
*** vinod has joined #openstack-dns22:07
*** EmilienM has quit IRC22:10
*** EmilienM has joined #openstack-dns22:12
*** vinod has quit IRC22:17
*** jmcbride has quit IRC22:19
*** CaptTofu has joined #openstack-dns22:30
*** HenryG has quit IRC22:42
*** jmcbride has joined #openstack-dns22:43
*** jmcbride has quit IRC22:45
*** jmcbride has joined #openstack-dns22:47
*** jmcbride has quit IRC22:48
*** CaptTofu has quit IRC23:06
*** nkinder has joined #openstack-dns23:11
*** pasquier-s_ has joined #openstack-dns23:16
*** pasquier-s has quit IRC23:17
*** bauruine has quit IRC23:17
*** bauruine has joined #openstack-dns23:18
*** sballe_ has quit IRC23:24
« Tuesday, 2014-01-14 Index Thursday, 2014-01-16 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!