| vipul | I hear this YAML thing is pretty hip | 00:00 |
|---|---|---|
| *** michael-yu has quit IRC | 00:01 | |
| esp | xml hurts me | 00:01 |
| juice | until it is formatted properly, they all hurt me | 00:02 |
| juice | don't tell me some of the big blobs of json coming back from openstack don't make you feel a bit nauseous | 00:02 |
| amcrn | xml is like violence, if it doesn't solve your problem, you're not using enough of it. | 00:02 |
| juice | amcrn: I'm getting the tattooed on my arm | 00:02 |
| amcrn | haha | 00:03 |
| esp | I feel like if you are going to make me use xml you might as well throw ninja stars in my face while you are at it. | 00:05 |
| esp | k, sorry for the drama. back to work. | 00:06 |
| *** tattabbum has quit IRC | 00:07 | |
| *** matsuhashi has joined #openstack-trove | 00:26 | |
| *** esp has left #openstack-trove | 00:40 | |
| *** isviridov has quit IRC | 00:50 | |
| *** juantwo has joined #openstack-trove | 00:52 | |
| *** dmakogon_ has quit IRC | 00:58 | |
| *** zigo has quit IRC | 00:58 | |
| *** isviridov has joined #openstack-trove | 00:58 | |
| *** erkules has quit IRC | 00:58 | |
| *** dmakogon__ has joined #openstack-trove | 00:58 | |
| *** openstack has joined #openstack-trove | 01:04 | |
| *** erkules has joined #openstack-trove | 01:05 | |
| *** ramashri has joined #openstack-trove | 01:10 | |
| *** eghobo has quit IRC | 01:13 | |
| *** Barker has joined #openstack-trove | 01:18 | |
| *** achampion has joined #openstack-trove | 01:19 | |
| *** zigo has quit IRC | 01:27 | |
| *** zigo has joined #openstack-trove | 01:33 | |
| *** michael-yu has joined #openstack-trove | 01:33 | |
| *** michael-yu has quit IRC | 01:33 | |
| *** nosnos has joined #openstack-trove | 01:39 | |
| *** khyati_ has quit IRC | 01:56 | |
| openstackgerrit | A change was merged to openstack/trove: Root_on_create per datastore https://review.openstack.org/77275 | 01:56 |
| *** matsuhashi has quit IRC | 01:58 | |
| *** erkules has quit IRC | 02:07 | |
| *** matsuhashi has joined #openstack-trove | 02:17 | |
| *** zigo has quit IRC | 02:19 | |
| *** erkules has joined #openstack-trove | 02:20 | |
| *** zigo has joined #openstack-trove | 02:23 | |
| *** michael-yu has joined #openstack-trove | 02:24 | |
| *** khyati has joined #openstack-trove | 02:27 | |
| *** michael-yu has quit IRC | 02:37 | |
| *** harlowja is now known as harlowja_away | 02:48 | |
| *** haomaiwang has quit IRC | 02:51 | |
| *** haomaiwa_ has joined #openstack-trove | 02:51 | |
| *** SushilKM has joined #openstack-trove | 03:02 | |
| *** amcrn has quit IRC | 03:05 | |
| *** haomaiwa_ has quit IRC | 03:05 | |
| *** mattgriffin has joined #openstack-trove | 03:19 | |
| *** rwsu has quit IRC | 03:19 | |
| *** haomaiwang has joined #openstack-trove | 03:20 | |
| *** Barker has quit IRC | 03:23 | |
| *** matsuhashi has quit IRC | 03:23 | |
| *** tanisdl has quit IRC | 03:26 | |
| cp16net | oh phely | 03:31 |
| cp16net | foowey | 03:32 |
| cp16net | found another bug fixing the other one... :-P | 03:32 |
| *** haomaiwang has quit IRC | 03:36 | |
| *** haomaiwang has joined #openstack-trove | 03:37 | |
| *** haomaiw__ has joined #openstack-trove | 03:40 | |
| *** SushilKM has quit IRC | 03:42 | |
| *** haomaiwang has quit IRC | 03:43 | |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove: Parses default configuration of an instance https://review.openstack.org/79838 | 03:44 |
| *** eghobo has joined #openstack-trove | 03:45 | |
| *** rwsu has joined #openstack-trove | 03:50 | |
| *** achampio1 has joined #openstack-trove | 03:51 | |
| *** achampion has quit IRC | 03:54 | |
| *** achampio1 has quit IRC | 03:55 | |
| *** achampion has joined #openstack-trove | 03:56 | |
| *** esp has joined #openstack-trove | 03:58 | |
| *** mattgriffin has quit IRC | 04:05 | |
| *** matsuhashi has joined #openstack-trove | 04:25 | |
| *** achampion has quit IRC | 04:32 | |
| *** juantwo has quit IRC | 04:40 | |
| *** juantwo has joined #openstack-trove | 04:44 | |
| *** timfreund has quit IRC | 04:52 | |
| *** ramashri has quit IRC | 05:04 | |
| *** juantwo has quit IRC | 05:05 | |
| openstackgerrit | Sam Morrison proposed a change to openstack/trove: Make hostnames in designate driver all lower case https://review.openstack.org/78439 | 05:19 |
| *** timfreund has joined #openstack-trove | 05:21 | |
| *** ramashri has joined #openstack-trove | 05:42 | |
| *** yidclare has joined #openstack-trove | 05:46 | |
| *** ramashri has quit IRC | 05:48 | |
| *** ViswaV_ has quit IRC | 05:55 | |
| *** SushilKM has joined #openstack-trove | 05:57 | |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove: Datastore Configuration Parameters stored in db https://review.openstack.org/79850 | 06:03 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/python-troveclient: adding configuration paramters mgmt api https://review.openstack.org/79851 | 06:03 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: Configurations stored in the database https://review.openstack.org/79852 | 06:04 |
| cp16net | there we go... hub_cap a little late i guess | 06:04 |
| *** yidclare has quit IRC | 06:11 | |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove: Datastore Configuration Parameters stored in db https://review.openstack.org/79850 | 06:21 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove: Parses default configuration of an instance https://review.openstack.org/79838 | 06:23 |
| *** khyati has quit IRC | 06:29 | |
| openstackgerrit | shivam shukla proposed a change to openstack/trove: Tests for heat based instance workflow https://review.openstack.org/66499 | 06:29 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: Configurations stored in the database https://review.openstack.org/79852 | 06:34 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/python-troveclient: adding configuration paramters mgmt api https://review.openstack.org/79851 | 06:34 |
| openstackgerrit | Craig Vyvial proposed a change to openstack/trove: Datastore Configuration Parameters stored in db https://review.openstack.org/79850 | 06:34 |
| *** denis_makogon has joined #openstack-trove | 06:50 | |
| *** PradeepChandani has joined #openstack-trove | 06:51 | |
| *** saju_m has joined #openstack-trove | 06:52 | |
| *** flaper87|afk is now known as flaper87 | 07:24 | |
| *** eghobo has quit IRC | 07:28 | |
| *** matsuhashi has quit IRC | 08:21 | |
| *** amcrn has joined #openstack-trove | 08:35 | |
| *** saju_m has quit IRC | 08:41 | |
| *** tattabbum has joined #openstack-trove | 08:47 | |
| *** iartarisi has joined #openstack-trove | 08:50 | |
| *** amcrn_ has joined #openstack-trove | 08:50 | |
| *** amcrn has quit IRC | 08:53 | |
| *** amcrn_ is now known as amcrn | 08:53 | |
| *** saju_m has joined #openstack-trove | 09:09 | |
| *** ashestakov_ has joined #openstack-trove | 09:16 | |
| *** dukhlov has quit IRC | 09:24 | |
| *** dukhlov has joined #openstack-trove | 09:30 | |
| *** amcrn has quit IRC | 09:38 | |
| *** SushilKM has quit IRC | 09:39 | |
| *** SushilKM has joined #openstack-trove | 09:44 | |
| *** denis_makogon has quit IRC | 09:51 | |
| *** ashestakov_ has quit IRC | 10:04 | |
| *** iartarisi has quit IRC | 10:33 | |
| *** saju_m has quit IRC | 10:49 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add timeout on prepare call per datastore https://review.openstack.org/77160 | 10:58 |
| openstackgerrit | Sascha Peilicke proposed a change to openstack/python-troveclient: Add py33 trove classifier https://review.openstack.org/71895 | 11:05 |
| openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Add point in time recovery https://review.openstack.org/77223 | 11:05 |
| *** IvanZ has joined #openstack-trove | 11:13 | |
| *** ChanServ changes topic to "test/gate jobs are queuing now in preparation for gerrit maintenance at 12:00 utc (eta to resume is 12:30 utc)" | 11:28 | |
| *** SushilKM has quit IRC | 11:34 | |
| openstackgerrit | shivam shukla proposed a change to openstack/trove: Tests for heat based instance workflow https://review.openstack.org/66499 | 11:35 |
| *** SushilKM has joined #openstack-trove | 11:37 | |
| *** IvanZ has quit IRC | 11:48 | |
| *** ChanServ changes topic to "gerrit on review.openstack.org is down for maintenance (eta to resume is 12:30 utc)" | 12:07 | |
| *** ChanServ changes topic to "gerrit on review.openstack.org is down for maintenance (revised eta to resume is 13:00 utc)" | 12:24 | |
| *** IvanZ has joined #openstack-trove | 12:36 | |
| *** saju_m has joined #openstack-trove | 12:37 | |
| *** yidclare has joined #openstack-trove | 12:47 | |
| *** openstackgerrit has quit IRC | 12:54 | |
| *** openstackgerrit has joined #openstack-trove | 12:54 | |
| *** Ranjitha has joined #openstack-trove | 13:04 | |
| *** achampion has joined #openstack-trove | 13:08 | |
| *** openstackgerrit has quit IRC | 13:08 | |
| *** openstackgerrit has joined #openstack-trove | 13:08 | |
| *** Ranjitha has quit IRC | 13:08 | |
| *** Ranjitha has joined #openstack-trove | 13:09 | |
| *** iartarisi has joined #openstack-trove | 13:17 | |
| *** robertmyers has joined #openstack-trove | 13:20 | |
| *** robertmyers has quit IRC | 13:21 | |
| *** robertmyers has joined #openstack-trove | 13:22 | |
| *** jcru has joined #openstack-trove | 13:26 | |
| *** radez_g0n3 is now known as radez | 13:38 | |
| *** nosnos has quit IRC | 13:43 | |
| *** mattgriffin has joined #openstack-trove | 13:49 | |
| *** matsuhashi has joined #openstack-trove | 13:50 | |
| *** matsuhashi has quit IRC | 13:56 | |
| *** demorris has joined #openstack-trove | 13:56 | |
| *** SushilKM has quit IRC | 13:58 | |
| *** IvanZ has quit IRC | 13:59 | |
| *** mattgriffin has quit IRC | 13:59 | |
| *** mattgriffin has joined #openstack-trove | 14:00 | |
| *** matsuhashi has joined #openstack-trove | 14:00 | |
| *** openstack has joined #openstack-trove | 14:04 | |
| *** IvanZ has joined #openstack-trove | 14:08 | |
| *** iartarisi has quit IRC | 14:09 | |
| *** grapex has joined #openstack-trove | 14:17 | |
| *** demorris has quit IRC | 14:18 | |
| *** demorris has joined #openstack-trove | 14:20 | |
| *** matsuhashi has quit IRC | 14:23 | |
| *** matsuhashi has joined #openstack-trove | 14:30 | |
| *** Barker has joined #openstack-trove | 14:34 | |
| *** mattgriffin has quit IRC | 14:35 | |
| *** datsun180b has joined #openstack-trove | 14:37 | |
| *** grapex has quit IRC | 14:37 | |
| *** grapex has joined #openstack-trove | 14:38 | |
| *** yidclare has quit IRC | 14:39 | |
| *** ChanServ changes topic to "Trove Openstack Database as a Service | Docs https://github.com/openstack/trove | Channel Logs http://eavesdrop.openstack.org/irclogs/ | IRC must read http://sackheads.org/~bnaylor/spew/away_msgs.html" | 14:42 | |
| *** Ranjitha has quit IRC | 14:43 | |
| *** shivamshukla has joined #openstack-trove | 14:44 | |
| *** mattgriffin has joined #openstack-trove | 14:49 | |
| *** thedodd has joined #openstack-trove | 14:50 | |
| *** kevinconway has joined #openstack-trove | 14:53 | |
| *** jmontemayor has joined #openstack-trove | 14:55 | |
| *** jmontemayor has quit IRC | 14:55 | |
| *** jmontemayor has joined #openstack-trove | 14:57 | |
| *** saju_m has quit IRC | 15:03 | |
| *** demorris has quit IRC | 15:04 | |
| *** saju_m has joined #openstack-trove | 15:04 | |
| *** demorris has joined #openstack-trove | 15:06 | |
| *** demorris has quit IRC | 15:10 | |
| *** rramirez has joined #openstack-trove | 15:14 | |
| *** IvanZ has quit IRC | 15:15 | |
| rramirez | Hello everyone! I'm one of the newest additions to the Rackspace Trove team. | 15:16 |
| *** IvanZ has joined #openstack-trove | 15:17 | |
| datsun180b | hiya rueben | 15:17 |
| datsun180b | nice silence, rest of the channel | 15:18 |
| *** rramirez has left #openstack-trove | 15:18 | |
| *** matsuhashi has quit IRC | 15:18 | |
| *** dmakogon__ is now known as denis_makogon | 15:20 | |
| denis_makogon | rramirez, welcome =) | 15:20 |
| denis_makogon | datsun180b, we are out at least its too early for showing up | 15:21 |
| *** rramirez has joined #openstack-trove | 15:26 | |
| *** ViswaV has joined #openstack-trove | 15:35 | |
| *** ViswaV_ has joined #openstack-trove | 15:36 | |
| *** PradeepChandani has quit IRC | 15:39 | |
| *** ViswaV has quit IRC | 15:39 | |
| *** kpom is now known as k-pom | 15:39 | |
| *** saju_m has quit IRC | 15:40 | |
| *** shivamshukla has quit IRC | 15:40 | |
| *** PradeepChandani has joined #openstack-trove | 15:40 | |
| *** shivamshukla has joined #openstack-trove | 15:41 | |
| *** SushilKM has joined #openstack-trove | 15:44 | |
| *** tanisdl has joined #openstack-trove | 15:46 | |
| *** Barker has quit IRC | 15:48 | |
| *** Barker has joined #openstack-trove | 15:52 | |
| *** tanisdl has quit IRC | 15:53 | |
| *** tanisdl has joined #openstack-trove | 15:54 | |
| *** eghobo has joined #openstack-trove | 15:56 | |
| *** eghobo has quit IRC | 15:58 | |
| SlickNik | Hey guys | 15:58 |
| SushilKM | Hello @SlickNik | 15:58 |
| SlickNik | Reminder that we have our trove meeting in about an hours time, so update the wiki page at https://wiki.openstack.org/wiki/Meetings/TroveMeeting if you want to add items to the agenda. | 15:59 |
| grapex | SlickNik: Don't you mean two hours? | 15:59 |
| SlickNik | grapex: good point. We're tied to UTC, so we don't observe DST. | 16:00 |
| SlickNik | So yes, two hours :) | 16:00 |
| grapex | SlickNik: I know because all of the Texans are looking forward to lunch | 16:01 |
| grapex | because everything is bigger in Texas, including our waistlines | 16:01 |
| * grapex begins firing two revolvers into the air while yelling "yehaw" | 16:01 | |
| *** rramirez has quit IRC | 16:02 | |
| SlickNik | lol | 16:03 |
| *** eghobo has joined #openstack-trove | 16:04 | |
| *** IvanZ has quit IRC | 16:06 | |
| *** demorris has joined #openstack-trove | 16:14 | |
| *** digambar has joined #openstack-trove | 16:16 | |
| *** SushilKM has quit IRC | 16:20 | |
| *** khyati has joined #openstack-trove | 16:41 | |
| *** amcrn has joined #openstack-trove | 16:42 | |
| *** Barker has quit IRC | 16:44 | |
| *** Barker has joined #openstack-trove | 16:45 | |
| *** ViswaV_ has quit IRC | 16:45 | |
| *** ViswaV has joined #openstack-trove | 16:45 | |
| *** eghobo has quit IRC | 16:47 | |
| *** eghobo has joined #openstack-trove | 16:48 | |
| *** rramirez has joined #openstack-trove | 17:02 | |
| *** sballe has joined #openstack-trove | 17:03 | |
| *** harlowja_away is now known as harlowja | 17:04 | |
| *** jmontemayor has quit IRC | 17:04 | |
| *** jmontemayor has joined #openstack-trove | 17:05 | |
| *** atomic77 has joined #openstack-trove | 17:08 | |
| *** shivamshukla has left #openstack-trove | 17:08 | |
| *** shivamshukla has joined #openstack-trove | 17:08 | |
| *** sballe has quit IRC | 17:10 | |
| *** demorris has quit IRC | 17:16 | |
| *** saju_m has joined #openstack-trove | 17:16 | |
| *** rramirez has quit IRC | 17:18 | |
| *** saurabhs has joined #openstack-trove | 17:19 | |
| denis_makogon | https://github.com/openstack/trove/blob/master/trove/tests/unittests/guestagent/test_backups.py#L106-L114 ||| http://logs.openstack.org/60/77160/5/check/gate-trove-python27/199b1e7/console.html | 17:21 |
| denis_makogon | guys merged patch doesn't fix all issues | 17:21 |
| *** digambar has quit IRC | 17:21 | |
| denis_makogon | it still tries to execute real shell command on the slave environment | 17:22 |
| *** digambar has joined #openstack-trove | 17:22 | |
| *** demorris has joined #openstack-trove | 17:24 | |
| openstackgerrit | Ramashri Umale proposed a change to openstack/trove: Added checks for backup-create API call to ensure datastore supports backup strategy Also for trove-create API call (when restoring from backup) added checks to ensure that datastore and datastore version of backup being restored match with datastore and https://review.openstack.org/80010 | 17:24 |
| *** yidclare has joined #openstack-trove | 17:25 | |
| *** michael-yu has joined #openstack-trove | 17:28 | |
| *** rramirez has joined #openstack-trove | 17:30 | |
| *** michael-yu has quit IRC | 17:30 | |
| *** michael-yu has joined #openstack-trove | 17:31 | |
| *** openstackstatus has quit IRC | 17:38 | |
| *** amytron has joined #openstack-trove | 17:38 | |
| *** openstackstatus has joined #openstack-trove | 17:39 | |
| *** yogesh has joined #openstack-trove | 17:42 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Refactor mysqla and redis apps https://review.openstack.org/80020 | 17:42 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Refactor mysql and redis apps https://review.openstack.org/80020 | 17:43 |
| abramley | denis_makogon - is this refactor basically the same thing I was looking at as part of https://bugs.launchpad.net/trove/+bug/1276860 ? | 17:45 |
| denis_makogon | abramley, yes, you was right, different scopes | 17:45 |
| denis_makogon | abramley, i guess, now we can me this patches dependent (your and mine) | 17:46 |
| denis_makogon | abramley, since they are following the almost the same goal | 17:46 |
| denis_makogon | abramley, is it ok ? | 17:46 |
| abramley | denis_makogon - just surprised since only yesterday you suggested I refactor the code along these lines. Wasn't aware you were already working on a bp for it | 17:46 |
| denis_makogon | abramley, i should've notify you, that i took this | 17:47 |
| denis_makogon | abramley, now we can co-work | 17:48 |
| digambar | Guys looking into this bug https://bugs.launchpad.net/trove/+bug/1130929 | 17:49 |
| digambar | trove-mgmt-cli | 17:49 |
| denis_makogon | abramley, my patch addresses to memory usage and refactoring, your to the hiding stack traces | 17:49 |
| denis_makogon | digambar, rd* are rackspace specific mgmt apis. They need to push that functionality into nova before these calls will actually work. | 17:50 |
| denis_makogon | digambar, this is rax specific code, as hub_cap said | 17:51 |
| digambar | ok | 17:51 |
| abramley | denis_makogon - if I understand the new process, your bp will need to be approved before you can move forward? | 17:51 |
| digambar | then what is the workaround for this or should skip this bug ? | 17:51 |
| denis_makogon | abramley, i guess no, but the patch probably would go through several iteration | 17:52 |
| denis_makogon | digambar, lets ask hub_cap or grapex about that | 17:52 |
| denis_makogon | hub_cap, are you with us ? | 17:52 |
| abramley | denis_makogon - I will hopefully have mine pushed soon (transitioning to a new dev machine so taking longer than I wanted) - and then you can make yours depend on mine ? | 17:53 |
| denis_makogon | abramley, i guess, we should do opposite thing | 17:53 |
| denis_makogon | abramley, your patch should be dependent on mine | 17:54 |
| denis_makogon | abramley, or even they could stay separate | 17:54 |
| denis_makogon | abramley, as far as i can see, your patch looks like a "feature" and it could land after march 27 | 17:55 |
| abramley | denis_makogon - but then we would need yet another one to identify further methods that need to be moved down into the base class. I prefer to keep mine focused on adding in the improved exception handling - and part of that was to add a base class. Anything beyond that to refactor the datastores seems more related to your 'refactor the datqstores' blueprint - and part of that refactoring would include pushing | 17:55 |
| abramley | common code into the base class | 17:55 |
| *** SnowDust has joined #openstack-trove | 17:56 | |
| denis_makogon | abramley, methods are not changed in my patch | 17:56 |
| denis_makogon | abramley, your patch totally fine, mine, i guess, too, but they follow different approaches | 17:57 |
| *** shivamshukla has quit IRC | 17:58 | |
| denis_makogon | abramley, the difference is that i added constructor to each class that instantiate the main datastore app, status app | 17:59 |
| *** NehaV has joined #openstack-trove | 17:59 | |
| amcrn | denis_makogon abramley: i'll reply to this once the weekly meeting is over | 18:00 |
| digambar | @denis_makogon - I think hub_cap is not there | 18:01 |
| denis_makogon | amcrn, fine | 18:01 |
| *** Ranjitha has joined #openstack-trove | 18:05 | |
| digambar | denis_makogon - want to contribute on clientside as well as securitygroup side | 18:05 |
| digambar | should I assign bug from that category to me ? | 18:05 |
| denis_makogon | digambar, we have weekly meeting now | 18:05 |
| digambar | yes | 18:05 |
| denis_makogon | digambar, ping me in the 1h | 18:05 |
| digambar | ok | 18:05 |
| *** ramashri has joined #openstack-trove | 18:06 | |
| *** eghobo has quit IRC | 18:07 | |
| *** eghobo has joined #openstack-trove | 18:09 | |
| *** yidclare has quit IRC | 18:17 | |
| *** saju_m has quit IRC | 18:30 | |
| *** digambar has quit IRC | 18:41 | |
| *** tattabbum has quit IRC | 18:52 | |
| *** sriram_tesora has joined #openstack-trove | 18:53 | |
| denis_makogon | merged patch doesn't fix anything, gate still failing https://bugs.launchpad.net/trove/+bug/1291516 | 18:56 |
| *** datsun180b has quit IRC | 18:59 | |
| SlickNik | what review is that for? | 19:02 |
| denis_makogon | SlickNik, https://review.openstack.org/#/c/77160/ | 19:02 |
| denis_makogon | SlickNik, i though submitted patch fixed that | 19:03 |
| denis_makogon | SlickNik, but as i can see, its not | 19:04 |
| SlickNik | denis_makogon: What's the problem then, do you know? | 19:04 |
| denis_makogon | SlickNik, its incremental backup tests | 19:05 |
| denis_makogon | SlickNik, i aslo added link at bug description | 19:05 |
| SlickNik | denis_makogon: I can't investigate now, I need to run to a meeting. | 19:05 |
| SlickNik | denis_makogon: Will look when I get out. | 19:05 |
| denis_makogon | SlickNik, cool | 19:05 |
| *** NehaV1 has joined #openstack-trove | 19:07 | |
| *** NehaV has quit IRC | 19:08 | |
| denis_makogon | amcrn, you said you would replay to our convo | 19:08 |
| amcrn | abramley denis_makogon: as to your conversation above | 19:08 |
| denis_makogon | amcrn, would be cool to hear your thoughts | 19:08 |
| amcrn | https://review.openstack.org/#/c/78054/ should be amended to use raise exception.DatastoreOperationNotSupported() in the existing managers, but the refactoring part should be reverted as it's overstepping the bounds of the problem statement. after that has merged, any refactoring requires a new blueprint, and said blueprint must be approved. | 19:09 |
| *** rramirez has quit IRC | 19:09 | |
| denis_makogon | amcrn, how https://review.openstack.org/#/c/80020/ overstepping the bounds of https://review.openstack.org/#/c/78054/ ? | 19:10 |
| amcrn | you misunderstand my point | 19:10 |
| amcrn | the bug is to have the guestagent properly handle an unsupported operation | 19:11 |
| denis_makogon | yes | 19:11 |
| amcrn | https://review.openstack.org/#/c/78054/ fixes said bug, but it goes a bit too far in terms of refactoring | 19:11 |
| amcrn | although the intentions were good, it seems there is some conflict between how to approach the refactoring | 19:11 |
| denis_makogon | yes, thats what i said | 19:11 |
| denis_makogon | i mean wrote at comments | 19:12 |
| amcrn | and therefore, the simple solution to this is to amend https://review.openstack.org/#/c/78054/ | 19:12 |
| abramley | amcrn - I wouldn't say it was 'refactoring'- it was simply the best and least error prone way of adding these improved exceptions into the datastores | 19:12 |
| amcrn | to revert the refactoring, but keep the essential raise exception.DatastoreOperationNotSupported() logic | 19:12 |
| denis_makogon | amcrn, agreed | 19:12 |
| denis_makogon | amcrn, there's another one patch https://review.openstack.org/#/c/80020/ | 19:13 |
| amcrn | abramley: i agree with you, but since there's a differing of opinions on how to handle the base class, etc, the easiest way to ensure the bug fix lands is to keep it as minimally viable as possible | 19:13 |
| amcrn | denis_makogon: https://review.openstack.org/#/c/80020/ will be deferred until Juno | 19:13 |
| abramley | amcrn I would suggest then that the improvements to the exceptions should wait until after a base class exists - since anything else is throw away work | 19:13 |
| denis_makogon | amcrn, this one make code a bit easy and makes mysql/redis guest use less memory | 19:13 |
| denis_makogon | amcrn, of course | 19:14 |
| denis_makogon | amcrn, because it doesn't fits to FF request | 19:14 |
| amcrn | abramley: not sure i follow, the basis of your patch still fixes the issue | 19:14 |
| denis_makogon | abramley, i think you should split patches into the two | 19:15 |
| abramley | amcrn - it does fix the bug with a lot of copy / paste of code across all the datastores - if that is what you want I will revert my changes and do it that way - just seems wrong somehow though | 19:15 |
| denis_makogon | abramley, one for the exception handling, another one for the manager base class | 19:16 |
| vgnbkr | amcrn: I think what abramley is saying is that it doesn't make sense to do the exception fixes, if we will then do the refactoring and remove the exception fixes. | 19:16 |
| denis_makogon | abramley, dependent patches is everything you need | 19:16 |
| amcrn | abramley: given the point of the lifecycle we're at, and given there is a bit of contention on how baseline manager functionality should be shared, the least controversial option is to avoid changing the structure of the code. | 19:17 |
| denis_makogon | vgnbkr, that's why manager base class should come firts | 19:17 |
| abramley | I am clearly new to the process and have a lot to learn - I can raise another patch request for 'reduce duplication of code in datastores with a manager baseclass' or something like that | 19:17 |
| amrith | abramley: +<several> for that | 19:17 |
| vgnbkr | I suspect the issue is in calling it a "base class". It's really just a mixin class to make error handling more consistent. | 19:18 |
| denis_makogon | abramley, AFAIK, it will work for all of us | 19:18 |
| denis_makogon | vgnbkr, i think naming it base class is fine since it contains possible API implementation | 19:19 |
| denis_makogon | vgnbkr, my point that BaseManager wasn't made only for the making exception handling easier | 19:20 |
| denis_makogon | vgnbkr, it describes API contract for the new income managers | 19:20 |
| vgnbkr | My Python is still rudimentary, but I don't really see the point of API defining base classes in a dynamically typed language. | 19:21 |
| denis_makogon | vgnbkr, there a good word for such classes in Java - interface, but since its the python is duck-typing lang, you don't need to implement all methods | 19:21 |
| vgnbkr | If you don't have a compiler that enforces the API, it's just extra code to maintain. | 19:22 |
| denis_makogon | vgnbkr, agreed, but its very good document for writing new managers | 19:22 |
| denis_makogon | vgnbkr, it was not so easy to write cassandra/mongo manager from the mysql manager | 19:22 |
| vgnbkr | Then that is a major change that needs bp discussion. | 19:22 |
| denis_makogon | vgnbkr, probably you're right | 19:23 |
| denis_makogon | vgnbkr, but i still stand for such change, its required | 19:23 |
| denis_makogon | IMHO | 19:23 |
| denis_makogon | ViswaV, ping | 19:23 |
| denis_makogon | ViswaV, hope you saw my comments at https://blueprints.launchpad.net/trove/+spec/config-paramters-mongodb | 19:24 |
| denis_makogon | ViswaV, please do better search before submitting new blueprint | 19:24 |
| ViswaV | denis_makogon: ? | 19:25 |
| ViswaV | Where do you see my name in that BP link you posted above? | 19:26 |
| denis_makogon | ViswaV, sorry wrong ping | 19:26 |
| denis_makogon | ViswaV, tried to ping Khyati | 19:26 |
| ViswaV | np. | 19:26 |
| denis_makogon | ViswaV, i've got question to you also | 19:27 |
| ViswaV | yes | 19:27 |
| denis_makogon | ViswaV, whats the status of mongo backup/restore ? | 19:27 |
| ViswaV | https://review.openstack.org/#/c/78339/ | 19:27 |
| ViswaV | Waiting for Juno… | 19:27 |
| *** datsun180b has joined #openstack-trove | 19:27 | |
| khyati | denis_makogon as for https://blueprints.launchpad.net/trove/+spec/config-paramters-mongodb I've replied to your comment | 19:28 |
| khyati | could you please re-frame the description and ill dup my bp | 19:28 |
| khyati | rephrase** | 19:28 |
| denis_makogon | khyati, done | 19:29 |
| denis_makogon | ViswaV, i'm gonna review it soon, and response to it | 19:29 |
| ViswaV | ok. | 19:29 |
| khyati | okay thanks denis_makogon | 19:29 |
| *** michael-yu has quit IRC | 19:30 | |
| denis_makogon | khyati, so, do you want to pick that task ? | 19:30 |
| khyati | yes I'm working on it actively | 19:30 |
| *** SnowDust has quit IRC | 19:31 | |
| khyati | denis_makogon could you assign it to me ? | 19:33 |
| SlickNik | denis_makogon: Looks like I'm going to be in this meeting for a while. Can you take a look at what might be the issue with https://bugs.launchpad.net/trove/+bug/1291516? | 19:34 |
| SlickNik | Thanks! | 19:36 |
| denis_makogon | khyati, yes, i'll do that | 19:40 |
| denis_makogon | ViswaV, done | 19:41 |
| denis_makogon | ViswaV, the main problem is that you place whole command into the one exec, command should be divided as it done at https://review.openstack.org/#/c/77551/6/trove/guestagent/strategies/backup/mysql_impl.py | 19:43 |
| *** tanisdl has quit IRC | 19:43 | |
| denis_makogon | ViswaV, and also, i'd not suggest you to add another one conf value into the each oslo group, this makes configuration more complext, Trove should have single module for all backup/restore strategies | 19:44 |
| *** rramirez has joined #openstack-trove | 19:45 | |
| ViswaV | denis_makogon: I tried both ways and submitted the one with single cmd since I saw a precedent for that (https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/mysql_impl.py#L34-L42) .. either way , any particular reason why it SHOULD NOT be one command? | 19:45 |
| denis_makogon | ViswaV, that is why it _should_ be divided https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L135-L139 | 19:46 |
| denis_makogon | ViswaV, and only | 19:47 |
| denis_makogon | ViswaV, backup base class was re-factored to support non-streaming backup strategies (cassandra, mondo, couchbase, redis) | 19:48 |
| denis_makogon | ViswaV, in _pre_backup you collect data into the one object, run_backup streams it via "cat file", _post_backup deletes that file | 19:49 |
| denis_makogon | ViswaV, easy | 19:49 |
| ViswaV | yeah…that had not landed when I submitted... | 19:49 |
| ViswaV | Also, I see it as a convenient hook to use (if you need it) and not enforced ….but yes, good to break up the flow (the cmd) that way. Will look at it when I have to rebase. | 19:49 |
| denis_makogon | ViswaV, now i'd suggest you to update implementation, and also take a deep look at comments about gate error | 19:49 |
| denis_makogon | ViswaV, you need to mock correctly subprocess | 19:50 |
| ViswaV | Also, https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L71 that _run_post_backup() should be moved out of there to https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L75 | 19:50 |
| openstackgerrit | Ramashri Umale proposed a change to openstack/trove: Added Backup/Restore validations https://review.openstack.org/80010 | 19:50 |
| ViswaV | Yes, denis_makogon, I ran into some mock erros when some python commands were trying to be executed on the Jenkins and their subsystem caught it and threw error. | 19:51 |
| denis_makogon | ViswaV, no, it should not be moved | 19:51 |
| ViswaV | I modified my tests to mock some more additional methods to avoid that. | 19:51 |
| denis_makogon | ViswaV, but the gate failed | 19:51 |
| *** saurabhs has quit IRC | 19:52 | |
| denis_makogon | ViswaV, it means that that something still mocked incorrect | 19:52 |
| ViswaV | yup.. will look into it. | 19:53 |
| ViswaV | _run_post_backup() should be called when the "with runner…." has completed and is about to exit. | 19:55 |
| ViswaV | which is intercepted by the __exit__ method in https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L75 | 19:55 |
| denis_makogon | ViswaV, no, it shouldn't | 19:56 |
| ViswaV | That is why the _run_post_backup() should be moved into the __exit__ method. It's not enough to simply wrap the run() method with pre and post | 19:56 |
| denis_makogon | ViswaV, it somehow breaks the percona backuping | 19:57 |
| denis_makogon | ViswaV, so, _pre_backup and _post_backup are called in execute_backup | 19:57 |
| denis_makogon | and its fine | 19:57 |
| amcrn | denis_makogon: on a different note before i go to lunch, since https://review.openstack.org/#/c/77160/ requires a rebuild anyway, change reqiurest => requires and propesed => proposed in your commit msg. | 19:58 |
| *** demorris has quit IRC | 19:58 | |
| amcrn | thanks :) | 19:59 |
| ViswaV | denis_makogon: https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L102 , the "with runner" bounds the backup begin and backup ended. and not just the https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L70 | 20:00 |
| ViswaV | So until the "with runner" execution is done , you can't be sure that backup phase is done and post_backup () can be called. | 20:00 |
| denis_makogon | ViswaV, in this case - yes, but it should be done as separate patch | 20:01 |
| ViswaV | Which is why post_backup should be called in the __exit__() method. since the __enter__ and __exi()___ bound the "with runner" code. | 20:01 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add timeout on prepare call per datastore https://review.openstack.org/77160 | 20:01 |
| *** NehaV1 has quit IRC | 20:02 | |
| denis_makogon | ViswaV, IMHO, https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/base.py#L67-L72 - its fine | 20:02 |
| *** NehaV has joined #openstack-trove | 20:02 | |
| denis_makogon | ViswaV, i don't see the problem, guest will lauch _post_backup only if run() will finish successfully | 20:03 |
| denis_makogon | ViswaV, so, its fine, 100% | 20:03 |
| *** Ranjitha has quit IRC | 20:04 | |
| ViswaV | Denis, run() will only generate the data that needs to be uploaded to swift, that run() gets executed as soon as the "with runner…" block is entered (which calls __enter__ which invokes pre_backup() and run() ) | 20:05 |
| ViswaV | But the save to swift part is here https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L106 | 20:05 |
| ViswaV | post_backup() cannot happen until all that is finished .... | 20:06 |
| *** ramashri has quit IRC | 20:06 | |
| ViswaV | as it exists today per your changes the sequence would be enter the "with runner" ---> triggers __enter__() method --> pre, run, post ---> then move further down in the "with runner" block where save() happens ---> block ends and __exit__() gets invoked. | 20:07 |
| ViswaV | What if in your post_backup() you clean up the data generated by backup cmd that needs to be uploaded to swift….you will now be deleting that before the save() step is reached. | 20:08 |
| ViswaV | IT has to be __enter__() { pre_backup, run() } , ……remaining logic in "with runner" block …….block ends , __exit__() { post_backup} | 20:09 |
| ViswaV | If you do otherwise, you will run into serious race conditions possibly. | 20:09 |
| ViswaV | denis_makogon: ^^ | 20:09 |
| *** saurabhs has joined #openstack-trove | 20:20 | |
| *** datsun180b has quit IRC | 20:26 | |
| *** NehaV has quit IRC | 20:32 | |
| *** radez is now known as radez_g0n3 | 20:43 | |
| *** michael-yu has joined #openstack-trove | 20:45 | |
| *** Ranjitha has joined #openstack-trove | 20:46 | |
| *** datsun180b has joined #openstack-trove | 20:46 | |
| *** atomic77 has quit IRC | 20:47 | |
| *** khyati has quit IRC | 20:57 | |
| *** NehaV has joined #openstack-trove | 20:57 | |
| *** saurabhs has quit IRC | 20:59 | |
| *** NehaV has quit IRC | 20:59 | |
| *** NehaV has joined #openstack-trove | 20:59 | |
| *** saurabhs has joined #openstack-trove | 21:02 | |
| *** yidclare has joined #openstack-trove | 21:02 | |
| *** amytron has quit IRC | 21:08 | |
| *** amytron has joined #openstack-trove | 21:08 | |
| *** denis_makogon_ has joined #openstack-trove | 21:14 | |
| *** datsun180b has quit IRC | 21:19 | |
| *** denis_makogon has quit IRC | 21:21 | |
| *** denis_makogon_ is now known as denis_makogon | 21:21 | |
| *** dmakogon_ has joined #openstack-trove | 21:21 | |
| *** datsun180b has joined #openstack-trove | 21:23 | |
| denis_makogon | ViswaV, at the run() data is being streamed into the remote storage, _pre_backup - prepares the data, _post_backup drops it | 21:24 |
| *** Esteban has joined #openstack-trove | 21:25 | |
| denis_makogon | ViswaV, so, i still don't see the problem in the way its being implemented, i don't think if you move _post_backup into the __exit__ it'll give a huge valueadd into the guest perfomance | 21:25 |
| denis_makogon | ViswaV, got your email, this chat is being logged, so i can see all messages, | 21:26 |
| denis_makogon | ViswaV, i don't think that this is the problem in the way backup execution organized. | 21:27 |
| ViswaV | the point is when the post_backup drops it… it can't drop until save() (upload to storage) finishes…and not just wait only until run() finishes | 21:27 |
| denis_makogon | mat-lowery, ping | 21:32 |
| *** yidclare has quit IRC | 21:32 | |
| mat-lowery | denis_makogon: hi | 21:33 |
| denis_makogon | mat-lowery, hi | 21:33 |
| denis_makogon | mat-lowery, saw your comments at https://review.openstack.org/#/c/80010/ | 21:33 |
| denis_makogon | mat-lowery, i think you should copy them into the actual patchset, since commiter doesn't took into account your suggestions | 21:34 |
| denis_makogon | ViswaV, please provide the example | 21:35 |
| denis_makogon | ViswaV, real-life example | 21:35 |
| mat-lowery | denis_makogon: I commented on an old patch set but I work with the committer so I just pointed them out to her. Thanks though. | 21:36 |
| mat-lowery | denis_makogon: Is that what you mean? That I commented on a now out-of-date patch set? | 21:37 |
| *** saurabhs has quit IRC | 21:37 | |
| denis_makogon | mat-lowery, yeah, thats all | 21:38 |
| denis_makogon | mat-lowery, also, could please point the commiter that we have already registered blueprint for this patch, so, bug-report should be marked as duplicated | 21:39 |
| denis_makogon | mat-lowery, https://blueprints.launchpad.net/trove/+spec/validation-on-restore-by-the-backup-strategy | 21:39 |
| denis_makogon | mat-lowery, would be nice if you help with that | 21:40 |
| ViswaV | denis_makogon: We are talking about logic flow (based on how python "with…block" operates) and what is the appropriate bounds to place _pre and _post at… What real life example do you want me to offer? This is about potential race conditions | 21:40 |
| ViswaV | http://docs.python.org/2/whatsnew/2.6.html#pep-343-the-with-statement | 21:40 |
| ViswaV | The object’s __enter__() is called before with-block is executed and therefore can run set-up code. It also may return a value that is bound to the name variable, if given. (Note carefully that variable is not assigned the result of expression.) | 21:40 |
| denis_makogon | ViswaV, i know how "with" works | 21:41 |
| denis_makogon | ViswaV, i asked for the example when current implementation would not work | 21:41 |
| *** saurabhs has joined #openstack-trove | 21:41 | |
| mat-lowery | denis_makogon: Good point. I saw that. The bug addresses a bit more than just the blueprint. I'll let amcrn decide on the dupe or no dupe since he created the bug. | 21:41 |
| ViswaV | So do you see how for example , https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L105 can not be reached before everything in __enter__() finishes? | 21:41 |
| ViswaV | meaning pre, run and post() would have to finish before you get to storage.save() as implemented currently... | 21:42 |
| ViswaV | that is why you have to split pre,run before with block (in __enter__) and post after with block (in __exit__) | 21:43 |
| *** thedodd has quit IRC | 21:44 | |
| *** khyati has joined #openstack-trove | 21:47 | |
| denis_makogon | ViswaV, somehow i agree, but from the other side, it depends how to interpretate whole flow | 21:47 |
| denis_makogon | basically, the mail question when to clean-up | 21:48 |
| denis_makogon | *main | 21:48 |
| ViswaV | denis_makogon: as it exists today, my vote would be in the __exit__method of the runner. | 21:49 |
| ViswaV | Also look at https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L170 | 21:49 |
| ViswaV | the restore implementation is cleaner because there is no complexity of "with block " and enter exit methods and the whole flow of when something happens or not. | 21:49 |
| *** yidclare has joined #openstack-trove | 21:49 | |
| ViswaV | Not sure why backup was implemented this way. May be there are reasons that I am not aware of. May be thats why pre and post steps weren't even there until you added them. To me the restore implementation looks cleaner. | 21:50 |
| denis_makogon | ViswaV, "with" was removed because it causes percona incremental restore errors | 21:50 |
| ViswaV | Got it. May be similar flow interoperation errors were made there. | 21:51 |
| *** achampion has quit IRC | 21:51 | |
| ViswaV | But if you use with, which the backup logic is using, then we got to align with __enter__() and __exit__() | 21:51 |
| denis_makogon | ViswaV, as for me its fine | 21:51 |
| denis_makogon | ViswaV, its just the another way to interact with objects | 21:52 |
| denis_makogon | nothing else | 21:52 |
| ViswaV | denis_makogon: 'fine' meaning you are ok moving the _run_post_backup() to __exit__() method? | 21:52 |
| ViswaV | I see no harm in doing that. But potential harm in keeping the current mechanism as is. | 21:52 |
| denis_makogon | ViswaV, i guess yes | 21:52 |
| denis_makogon | ViswaV, its the easiest way to make clean up when it should happen | 21:53 |
| *** Barker has quit IRC | 21:53 | |
| denis_makogon | ViswaV, i'll take it | 21:53 |
| denis_makogon | ViswaV, thanks for the convo | 21:53 |
| ViswaV | Cool. We can tackle it during any of the backup patch sets submission (cassandra, mongo etc). No one else is using them yet. | 21:53 |
| ViswaV | Sure. np | 21:54 |
| denis_makogon | ViswaV, it'll be the separate patch | 21:54 |
| ViswaV | sure | 21:54 |
| denis_makogon | ViswaV, anyway this way of backuping is required by the non-streaming backup tools | 21:55 |
| denis_makogon | like mongoumn | 21:55 |
| denis_makogon | *mongodump | 21:55 |
| denis_makogon | or nodetool | 21:55 |
| ViswaV | yup | 21:55 |
| ViswaV | where you have to simulate streaming via tar.. | 21:56 |
| *** kevinconway has quit IRC | 21:57 | |
| *** NehaV has quit IRC | 21:58 | |
| denis_makogon | amcrn, thanks for updating the BP whiteboard | 21:58 |
| amcrn | np | 21:59 |
| *** Esteban has quit IRC | 22:01 | |
| *** NehaV has joined #openstack-trove | 22:01 | |
| amcrn | denis_makogon: yeah, so if you wouldn't might writing up the specs (taking the comments into consideration), then we could review the bp on an upcoming monday | 22:01 |
| denis_makogon | amcrn, review which one ? | 22:02 |
| amcrn | https://blueprints.launchpad.net/trove/+spec/validation-on-restore-by-the-backup-strategy | 22:03 |
| denis_makogon | cool | 22:04 |
| amcrn | the scope of this blueprint being the long-term solution | 22:04 |
| *** yidclare has quit IRC | 22:04 | |
| amcrn | (as compared to the in-flight https://review.openstack.org/#/c/80010/ which will only check datastore-version-id) | 22:05 |
| denis_makogon | amcrn, yes, agreed, unfortunately i'm going to skip BP-meeting and weelky meeting due to traveling | 22:05 |
| amcrn | ah, vacation or work? | 22:05 |
| denis_makogon | amcrn, vacation | 22:05 |
| amcrn | niiice | 22:05 |
| denis_makogon | amcrn, far-far away from the internet | 22:06 |
| denis_makogon | i hope | 22:06 |
| amcrn | denis_makogon: that's always a good plan every once in awhile, decompress and enjoy the natural surroundings :) | 22:06 |
| *** sriram_tesora has quit IRC | 22:06 | |
| denis_makogon | amcrn, yeah, agreed =) | 22:06 |
| denis_makogon | amcrn, if you have time would be nice to hear your feedback on my feature designs (dblog, point in time recovery, data volume snapshot) | 22:07 |
| denis_makogon | amcrn, links are available at wiki.openstack.org/wiki/Trove#Development | 22:08 |
| amcrn | i'll do my best to give them a look by friday | 22:08 |
| *** timfreund has quit IRC | 22:08 | |
| amcrn | coffee break, bbiab | 22:09 |
| denis_makogon | amcrn, have a nice day, Auston =)) | 22:09 |
| *** yidclare has joined #openstack-trove | 22:17 | |
| *** yogesh has quit IRC | 22:27 | |
| *** harlowja has quit IRC | 22:30 | |
| *** harlowja has joined #openstack-trove | 22:31 | |
| *** NehaV has quit IRC | 22:41 | |
| *** timfreund has joined #openstack-trove | 22:42 | |
| *** Ranjitha has quit IRC | 22:47 | |
| *** jcru has quit IRC | 22:49 | |
| *** eguz has joined #openstack-trove | 22:56 | |
| *** datsun180b has quit IRC | 22:57 | |
| *** grapex has quit IRC | 22:58 | |
| *** eghobo has quit IRC | 23:00 | |
| *** eguz has quit IRC | 23:00 | |
| *** jmontemayor has quit IRC | 23:11 | |
| *** rramirez has quit IRC | 23:17 | |
| *** amytron has quit IRC | 23:24 | |
| *** saurabhs has quit IRC | 23:24 | |
| *** achampion has joined #openstack-trove | 23:25 | |
| *** saurabhs has joined #openstack-trove | 23:27 | |
| *** denis_makogon has quit IRC | 23:32 | |
| *** yidclare has quit IRC | 23:40 | |
| *** atomic77 has joined #openstack-trove | 23:40 | |
| *** saurabhs has quit IRC | 23:47 | |
| *** saurabhs has joined #openstack-trove | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!