Tuesday, 2016-01-12

openstackgerritPetr Malik proposed openstack/trove: Fix leaked mocks in 'test_dbaas'  https://review.openstack.org/26500800:06
openstackgerritPetr Malik proposed openstack/trove: Fix leaked mocks in 'test_dbaas'  https://review.openstack.org/26500800:06
*** amrith is now known as _amrith_00:08
*** jengeb1 has quit IRC00:33
*** shakamunyi has quit IRC00:35
*** shakamunyi has joined #openstack-trove00:35
*** imandhan has quit IRC00:47
*** barra204 has joined #openstack-trove00:56
*** shakamunyi has quit IRC00:57
*** barra204 has quit IRC01:14
*** shakamunyi has joined #openstack-trove01:15
*** imandhan has joined #openstack-trove01:17
*** sonali5 has quit IRC01:28
*** barra204 has joined #openstack-trove01:31
*** shakamunyi has quit IRC01:32
*** imandhan has quit IRC01:35
*** shakamunyi has joined #openstack-trove01:36
*** imandhan has joined #openstack-trove01:36
*** barra204 has quit IRC01:38
*** barra204 has joined #openstack-trove01:38
*** shakamunyi has quit IRC01:39
*** shakamunyi has joined #openstack-trove01:44
*** barra204 has quit IRC01:45
*** imandhan has quit IRC01:47
*** haomaiwang has quit IRC01:48
*** shakamunyi has quit IRC01:48
*** 14WAAN277 has joined #openstack-trove01:48
*** shakamunyi has joined #openstack-trove01:48
*** imandhan has joined #openstack-trove01:49
*** 14WAAN277 has quit IRC01:50
*** tellesnobrega is now known as tellesnobrega_af01:50
*** csoukup has joined #openstack-trove01:52
*** shakamunyi has quit IRC01:54
*** _amrith_ is now known as amrith01:56
*** csoukup has quit IRC01:58
*** imandhan has quit IRC02:00
*** imandhan has joined #openstack-trove02:00
*** shakamunyi has joined #openstack-trove02:08
*** shakamunyi has quit IRC02:10
*** qianli has joined #openstack-trove02:11
*** shakamunyi has joined #openstack-trove02:12
*** imandhan has quit IRC02:21
*** shakamunyi has quit IRC02:22
*** shakamunyi has joined #openstack-trove02:23
*** imandhan has joined #openstack-trove02:23
*** mingdang1 has joined #openstack-trove02:28
*** shakamunyi has quit IRC02:32
*** imandhan has quit IRC02:34
*** imandhan has joined #openstack-trove02:36
*** pmackinn has quit IRC02:42
*** hub_cap has joined #openstack-trove02:55
*** hub_cap has left #openstack-trove02:55
*** cheneydc has joined #openstack-trove03:25
openstackgerritPetr Malik proposed openstack/trove: Fix leaked mocks in 'test_dbaas'  https://review.openstack.org/26500804:18
*** cheneydc has quit IRC04:25
openstackgerritPetr Malik proposed openstack/trove-specs: Implement cassandra cluster provisioning  https://review.openstack.org/26613904:25
openstackgerritPetr Malik proposed openstack/trove-specs: Add support for 'root' actions to Cassandra  https://review.openstack.org/26614004:25
*** links has joined #openstack-trove04:26
*** coolsvap|away is now known as coolsvap04:29
openstackgerritPetr Malik proposed openstack/trove-specs: Implement cassandra cluster provisioning  https://review.openstack.org/26613904:30
*** imandhan has quit IRC04:56
*** imandhan has joined #openstack-trove04:57
*** itlinux has joined #openstack-trove05:04
*** itlinux has quit IRC05:04
*** imandhan has quit IRC05:08
*** imandhan has joined #openstack-trove05:08
*** imandhan has quit IRC05:19
*** imandhan has joined #openstack-trove05:22
*** imandhan has quit IRC05:43
*** imandhan has joined #openstack-trove05:51
*** cheneydc has joined #openstack-trove05:57
openstackgerritting wang proposed openstack/python-troveclient: Add debug testenv in tox  https://review.openstack.org/26620406:06
*** imandhan has quit IRC06:10
*** qianli has quit IRC06:18
*** qianli has joined #openstack-trove06:19
*** 6A4ABMVDU has joined #openstack-trove06:26
openstackgerritvenkatamahesh proposed openstack/trove: Improvements in python condition code  https://review.openstack.org/26622006:33
*** sgotliv_ has joined #openstack-trove06:58
*** 6A4ABMVDU has quit IRC07:03
*** haomaiwang has joined #openstack-trove07:05
*** haomaiwang has quit IRC07:25
*** haomaiwang has joined #openstack-trove07:26
*** sgotliv_ has quit IRC07:49
*** haomaiwang has quit IRC07:56
*** haomaiwa_ has joined #openstack-trove07:58
*** haomaiwa_ has quit IRC07:59
*** haomaiwa_ has joined #openstack-trove07:59
*** haomaiwa_ has quit IRC08:01
*** haomaiwa_ has joined #openstack-trove08:01
*** haomaiwa_ has quit IRC08:26
*** haomaiwa_ has joined #openstack-trove08:33
openstackgerritvenkatamahesh proposed openstack/trove: Improvements in python condition code  https://review.openstack.org/26622008:38
*** haomaiwa_ has quit IRC08:42
*** haomaiwa_ has joined #openstack-trove08:44
*** haomaiwa_ has quit IRC09:01
*** 77CAAI2UM has joined #openstack-trove09:01
*** sgotliv_ has joined #openstack-trove09:12
*** openstackgerrit has quit IRC09:17
*** openstackgerrit has joined #openstack-trove09:17
*** sgotliv_ has quit IRC09:23
*** sgotliv_ has joined #openstack-trove09:38
*** sgotliv_ has quit IRC09:50
*** davideag_ has joined #openstack-trove09:53
*** davideagnello has quit IRC09:56
*** 77CAAI2UM has quit IRC10:01
*** jengeb has joined #openstack-trove10:01
openstackgerritWang Bo proposed openstack/trove: Replaces __builtin__ with six.moves.builtins  https://review.openstack.org/26627110:01
*** sgotliv_ has joined #openstack-trove10:03
*** bhaskarduvvuri has joined #openstack-trove10:06
*** haomaiwa_ has joined #openstack-trove10:07
*** bhaskarduvvuri has quit IRC10:10
*** bhaskarduvvuri has joined #openstack-trove10:11
*** bhaskarduvvuri has quit IRC10:13
*** haomaiwa_ has quit IRC10:20
*** bhaskarduvvuri has joined #openstack-trove10:34
*** bhaskarduvvuri has quit IRC10:35
*** coolsvap is now known as coolsvap|away10:56
*** tosky has joined #openstack-trove11:06
*** openstackgerrit has quit IRC11:17
*** openstackgerrit has joined #openstack-trove11:17
*** dalees has quit IRC11:37
*** mingdang1 has quit IRC11:46
*** mingdang1 has joined #openstack-trove11:47
*** akanksha_ has joined #openstack-trove11:51
*** mingdang1 has quit IRC11:51
*** dalees has joined #openstack-trove12:11
*** mingdang1 has joined #openstack-trove12:39
*** cheneydc has quit IRC12:43
*** qianli has quit IRC12:53
*** qianli has joined #openstack-trove12:53
*** haomaiwang has joined #openstack-trove12:56
*** sonali5 has joined #openstack-trove12:57
*** haomaiwang has quit IRC13:01
*** shakamunyi has joined #openstack-trove13:04
*** haomaiwang has joined #openstack-trove13:06
*** haomaiwang has quit IRC13:11
*** cheneydc has joined #openstack-trove13:21
*** haomaiwang has joined #openstack-trove13:23
openstackgerritPeter Stachowski proposed openstack/trove: Add support for configuration groups in int-tests  https://review.openstack.org/25385513:23
*** mingdang_ has joined #openstack-trove13:23
*** mingdang1 has quit IRC13:24
*** haomaiwang has quit IRC13:28
*** cheneydc has quit IRC13:29
*** sgotliv_ has quit IRC13:30
*** sgotliv has joined #openstack-trove13:39
openstackgerritPeter Stachowski proposed openstack/trove-specs: Add license mangement to Trove  https://review.openstack.org/26637413:40
sonali5need help getting error on git review13:51
sonali5^^pmalik13:52
sonali5peterstac13:52
sonali5http://paste.openstack.org/show/483579/13:52
*** bhaskarduvvuri has joined #openstack-trove13:52
*** miqui has joined #openstack-trove13:56
*** sonali5 has quit IRC13:58
*** haomaiwa_ has joined #openstack-trove13:59
*** haomaiwa_ has quit IRC14:01
*** haomaiwang has joined #openstack-trove14:01
*** pmackinn has joined #openstack-trove14:04
*** pmackinn_ has joined #openstack-trove14:08
*** pmackinn has quit IRC14:11
*** mingdang_ has quit IRC14:20
*** mingdang1 has joined #openstack-trove14:24
*** cheneydc has joined #openstack-trove14:28
*** tellesnobrega_af is now known as tellesnobrega14:29
*** bhaskarduvvuri has quit IRC14:33
*** kbyrne has quit IRC14:46
*** cheneydc has quit IRC14:47
*** tellesnobrega is now known as tellesnobrega_af14:47
*** links has quit IRC14:50
*** amrith is now known as _amrith_14:52
*** HimanshuGarg has joined #openstack-trove15:06
*** HimanshuGarg has left #openstack-trove15:07
*** mingdang1 has quit IRC15:20
*** haomaiwang has quit IRC15:24
*** kbyrne has joined #openstack-trove15:29
*** kbyrne has quit IRC15:29
*** kbyrne has joined #openstack-trove15:30
*** _amrith_ is now known as amrith15:39
amrithsonali5, do you have network issues?15:41
*** kbyrne has quit IRC15:50
*** surabhi has joined #openstack-trove15:52
surabhivkmc: Hello15:54
*** surabhi has quit IRC15:58
*** amrith is now known as _amrith_15:59
*** _amrith_ is now known as amrith16:00
*** kbyrne has joined #openstack-trove16:09
*** cweid has joined #openstack-trove16:44
*** bhaskarduvvuri has joined #openstack-trove16:44
openstackgerritPeter Stachowski proposed openstack/trove-dashboard: Add missing .gitreview file  https://review.openstack.org/26378016:44
*** akanksha_ has quit IRC16:57
openstackgerritDoug Shelley proposed openstack/trove-integration: Fix Percona XtraDB Cluster guest to work with v5.6  https://review.openstack.org/26648417:05
openstackgerritDoug Shelley proposed openstack/trove: Fix Percona XtraDB Cluster guest to work with v5.6  https://review.openstack.org/26648517:05
*** tellesnobrega_af is now known as tellesnobrega17:07
*** bhaskarduvvuri has quit IRC17:30
*** imandhan has joined #openstack-trove17:31
openstackgerritmariam john proposed openstack/trove-specs: Enable CouchDB replication in Trove  https://review.openstack.org/25543718:02
*** cweid has quit IRC18:13
*** pmackinn_ has quit IRC18:21
openstackgerritamrith proposed openstack/trove: Sometimes prepare messages are delayed  https://review.openstack.org/26652818:51
*** sabeen has joined #openstack-trove18:58
cp16netpmalik: this is strange http://paste.openstack.org/show/483646/19:01
cp16neti used an existing cluster and somehow it decided to use the host ip instead of the instances ip for data19:02
cp16netnotice the 10.0.0.1 instead of the list of ips from the cluster call "ip": ["10.0.0.2", "10.0.0.3", "10.0.0.4"]19:03
amrithSlickNik, an old enemy ... https://review.openstack.org/26652819:04
*** sabeen3 has joined #openstack-trove19:04
amrithcp16net, peterstac, vkmc, johnma ^^19:04
*** sabeen has quit IRC19:05
peterstacamrith, seems like such a small fix for such a big issue :)19:07
amrithto be precise peterstac, it is such a small band-aid on such a big wound.19:08
SlickNikamrith: will take a look19:08
amrithSlickNik, I've updated the bug with details.19:08
*** pmackinn has joined #openstack-trove19:09
cp16neti'm not sure i completely understand why waiting helps19:10
cp16neteven after reading the bug report19:10
openstackgerritamrith proposed openstack/trove: Update requirements.txt  https://review.openstack.org/26653619:17
amrithcp16net, stop() isn't synchronous19:21
*** itlinux has joined #openstack-trove19:21
amrithso after stop() returns, rabbit still is working on getting TM off the list of servers for the queue19:22
amrithand a prepare gets cast19:22
amrithso rabbit treats it as a valid server.19:22
amrithhowever since tm doesn't process the message and stop completes, the message sits in the queue waiting for a janitor to come and deal with it19:22
amrithwhich happens 300s later19:22
amrithby calling wait() we ensure that prepare never gets dispatched right back to the TM19:23
amrithcp16net ^^19:23
amrithtry the repro19:23
cp16netis there something set in rabbit to rehandle the message with a timer?19:23
amrithyou can enable debugging (which is what the two conf files do) and you can see this19:23
amrithcp16net, not that I can find19:23
amrithbut if you enable rabbit logging you may be able to see this; I didn19:23
amrithdidn't19:23
cp16netyeah i'm wondering if rabbit has a handler on the queue and waits for a while to register a new handler19:24
cp16netif there the message is already acknowledged19:24
amrithcould be; I'm looking further to see if there are more gremlins here.19:27
cp16netamrith: is there any other code in openstack that we could see if they do things differently for a scenario like this?19:31
cp16netmaybe sahara?19:31
amriththis code (using wait) comes from the oslo.messaging documentation19:32
cp16netnot sure that others are pushing a message before the server exists19:32
amrithhttp://docs.openstack.org/developer/oslo.messaging/server.html#oslo_messaging.MessageHandlingServer.wait19:32
amritheven their sample code does this ;)19:33
cp16nethmm ok i do recall this pattern from rabbit examples i think as well19:37
*** sgotliv has quit IRC19:44
openstackgerritMerged openstack/trove: Code refactoring for couchdb  https://review.openstack.org/21017020:39
*** amrith is now known as _amrith_21:00
pmackinncp16net, ping pxc21:16
cp16netpmackinn: howdy21:17
pmackinncp16net, was there an original bp for pxc as its own datastore? was there ever discussion to somehow fold it under percona?21:18
pmackinnor too distinct?21:18
cp16netyeah there is21:18
cp16netso it make it a separate datastore from percona?21:19
pmackinnwell it looks like that is how it is today21:19
cp16netright you are asking why?21:20
pmackinnpercona,pxc dirs21:20
cp16nethttp://docs-draft.openstack.org/74/266374/1/check/gate-trove-specs-docs/20b651a//doc/build/html/specs/liberty/percona-cluster.html21:21
pmackinnjust trying to understand galera clustering as an aspect of a datastore, or really its own animal21:21
pmackinnah cool thx21:21
cp16neti talked with some of the percona guys and they said pxc releases lag behind the percona releases21:21
cp16netso some features expected in percona might not exist yet in pxc21:22
pmackinnright, so for mariadb-galera, you would recommend a similar model?21:23
cp16neti'd say if a separate image is required then it should be a different datastore21:24
cp16netbut if its not required then it should use the same21:24
cp16neteven if a new image is required because the galera package needs to be installed shouldnt require a new datastore21:25
pmackinnlike in fedora, they would keep those in lock-step21:25
pmackinnin terms of compat21:25
cp16netif the galera package or new packages needed doesnt affect the single instance21:26
pmackinnhttps://mariadb.com/kb/en/mariadb/what-is-mariadb-galera-cluster/#galera-wsrep-provider-versions21:27
cp16netso if there is something like mariadb, mariadb-galera, mariadb-other-cluster-solution21:27
cp16netmaybe its just a version of the datastore21:28
pmackinnyeah...cool thx21:29
cp16netnp21:29
*** pmackinn has quit IRC21:29
cp16netthose ideas may need to be discussed with others in the community but thats my take21:29
vgnbkrpxc is a different datastore because it runs different code in mysqld.  It wouldn't be good for a bug in pxc to affect a non-clustered Percona database.21:47
vgnbkrAlso, there could be more than one clustering implementation for each datastore.  You could want Percona/pxc or Percona/galera, MySQL/pxc, MySQL/galera, MySQL/mysql-old, MySQL/mysql-new, etc.21:48
*** jengeb has quit IRC22:11
dougshelley66cp16net i have a Q about the stable/kilo test here - https://review.openstack.org/#/c/266484/22:11
dougshelley66the change i made depends on a change in trove, that i don't think we should backport22:12
dougshelley66but how will that stable/kilo test ever pass then?22:12
*** sgotliv has joined #openstack-trove22:12
cp16netdougshelley66: yeah seems like it wont ever pass22:18
cp16netwithout the changes needed on the trove side22:18
dougshelley66well it passed on mitaka22:18
cp16neti see that22:19
dougshelley66it just won't pass the stable/kilo test because i don't think we need to backport the trove fix22:19
dougshelley66i believe the breaking change for stable/kilo is in the ubuntu_mysql elements22:19
dougshelley66the 30-mysql file22:19
cp16netyeah i think it is22:20
cp16netthats the only thing thats used from the changes you made22:20
openstackgerritPeter Stachowski proposed openstack/trove-integration: Added enum to elements and admin_alt_demo to tests  https://review.openstack.org/25058822:21
cp16netinside the elements do we know what branch is being used?22:22
dougshelley66?22:22
cp16netis there an environment variable that we can key off of?22:22
dougshelley66ah i understand - you mean put an if statement in to use the old version of that changed line22:24
cp16netBRANCH_OVERRIDE=stable/kilo22:24
cp16neti think thats available22:24
dougshelley66ok - i can look at that22:24
*** _amrith_ is now known as amrith22:25
cp16neti see it used to installed the requirements in the guest22:25
dougshelley66so waht is the reason we don't version this repo?22:25
cp16neti dunno how else to get around this without removing the tests22:25
cp16netfor kilo22:25
cp16netwhich we want to keep working with older versions of trove22:25
cp16neti think the reason is that its not released code22:26
*** jengeb has joined #openstack-trove22:26
cp16neti think this was part of the reason that we were looking at moving the elements from t-i22:27
cp16netdont think much has happened around that tho22:27
dougshelley66ah i see22:27
dougshelley66ok i will take a look at using that variable22:27
cp16netcool i think that would be good22:29
dougshelley66i assume at some point someone will add a stable-liberty test to this job?22:29
dougshelley66which would break in this case too22:29
cp16netright i was thinking about that22:29
cp16netnot sure why there isnt a liberty now22:29
cp16netis kilo still supported?22:30
dougshelley66i think so22:30
openstackgerritPeter Stachowski proposed openstack/python-troveclient: Client Changes for Guest Log File Retrieval  https://review.openstack.org/25058922:30
dougshelley66so it looks like $ADD_BRANCH has the "stable" stripped off22:30
*** davideag_ has quit IRC22:31
*** mingdang1 has joined #openstack-trove22:32
*** davideagnello has joined #openstack-trove22:32
cp16netbut i'm not sure thats a global variable22:32
cp16netmight only be used in that element22:32
dougshelley66right22:33
cp16nethavnt looked further than the log i see for the kilo that failed in your patch22:33
*** davideagnello has quit IRC22:35
dougshelley66so it looks like i have to put something like this in the element:22:35
openstackgerritVictoria Martinez de la Cruz proposed openstack/trove-specs: Add support for MariaDB GTID Replication  https://review.openstack.org/26069022:36
dougshelley66 if [[ $BRANCH_OVERRIDE == "stable/liberty" || $BRANCH_OVERRIDE == "stable/kilo" ]]; then <use old code> else <new code>22:36
*** davideagnello has joined #openstack-trove22:36
*** pmackinn has joined #openstack-trove22:36
cp16netyeah something like that would work22:36
cp16netadd a comment in there why22:36
dougshelley66so you believe that $BRANCH_OVERRIDE is available in 30-mysql element?22:36
cp16neti'd assume it is22:36
dougshelley66ok thx22:36
cp16netif its avaiable in /tmp/image.gHDIGfvG/hooks/extra-data.d/15-reddwarf-dep22:37
cp16netlooks like it *should* be22:37
cp16neti hope it is :-P22:37
cp16netotherwise this might be a bit more work22:37
openstackgerritPeter Stachowski proposed openstack/trove: Implement Guest Log File Retrieval  https://review.openstack.org/25059022:37
*** imandhan has quit IRC22:46
*** imandhan has joined #openstack-trove22:47
*** davideagnello has quit IRC22:54
*** davideagnello has joined #openstack-trove22:56
openstackgerritMerged openstack/trove: Sometimes prepare messages are delayed  https://review.openstack.org/26652823:03
*** arist has joined #openstack-trove23:06
*** amrith is now known as _amrith_23:07
*** arist has left #openstack-trove23:08
*** davideagnello has quit IRC23:14
*** davideagnello has joined #openstack-trove23:16
*** pmackinn has quit IRC23:16
*** mingdang1 has quit IRC23:22
*** pmackinn has joined #openstack-trove23:24
*** _amrith_ is now known as amrith23:36
*** pmackinn has quit IRC23:39
*** sabeen3 has quit IRC23:40
*** logan2 has joined #openstack-trove23:42
*** logan2 is now known as logan-23:42

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