*** sharikapongubala has quit IRC | 00:02 | |
*** esmute_ has quit IRC | 00:18 | |
*** chlong has joined #openstack-trove | 00:27 | |
*** sharikapongubala has joined #openstack-trove | 00:59 | |
openstackgerrit | Ishita Mandhan proposed openstack/trove: Code refactoring for couchdb https://review.openstack.org/210170 | 01:01 |
---|---|---|
*** sharikapongubala has quit IRC | 01:04 | |
*** sharikapongubala has joined #openstack-trove | 01:05 | |
*** imandhan has joined #openstack-trove | 01:07 | |
*** sharikapongubala has quit IRC | 01:09 | |
openstackgerrit | octopuszhang proposed openstack/trove: Allow user to select zone for their trove volume https://review.openstack.org/209864 | 01:14 |
*** imandhan has quit IRC | 01:15 | |
*** sharikapongubala has joined #openstack-trove | 01:17 | |
*** sharikapongubala has quit IRC | 01:18 | |
*** erkules has joined #openstack-trove | 01:21 | |
*** erkules_ has quit IRC | 01:23 | |
*** sharikapongubala has joined #openstack-trove | 01:31 | |
*** sharikapongubala has quit IRC | 01:36 | |
*** sharikapongubala has joined #openstack-trove | 01:39 | |
*** sharikapongubala has quit IRC | 01:43 | |
openstackgerrit | octopuszhang proposed openstack/trove: Allow user to select zone for their trove volume https://review.openstack.org/209864 | 01:44 |
*** sharikapongubala has joined #openstack-trove | 01:46 | |
*** sharikapongubala has quit IRC | 01:51 | |
*** sharikapongubala has joined #openstack-trove | 01:52 | |
*** sharikapongubala has quit IRC | 01:54 | |
*** mingdang1 has joined #openstack-trove | 01:54 | |
*** sharikapongubala has joined #openstack-trove | 01:56 | |
*** sharikapongubala has quit IRC | 01:58 | |
*** sharikapongubala has joined #openstack-trove | 02:03 | |
*** sharikapongubala has quit IRC | 02:04 | |
*** sharikapongubala has joined #openstack-trove | 02:09 | |
*** sharikapongubala has quit IRC | 02:12 | |
*** pmackinn_ has quit IRC | 02:16 | |
*** sharikapongubala has joined #openstack-trove | 02:28 | |
*** sharikapongubala has quit IRC | 02:29 | |
*** sharikapongubala has joined #openstack-trove | 02:39 | |
*** sharikapongubala has quit IRC | 02:42 | |
*** sharikapongubala has joined #openstack-trove | 03:24 | |
*** rjaiswal has quit IRC | 03:40 | |
*** sharikapongubala has quit IRC | 03:40 | |
*** sharikapongubala has joined #openstack-trove | 04:01 | |
*** sharikapongubala has quit IRC | 04:05 | |
*** sgotliv_ has joined #openstack-trove | 04:11 | |
*** sharikapongubala has joined #openstack-trove | 04:21 | |
*** sharikapongubala has quit IRC | 04:24 | |
*** sharikapongubala has joined #openstack-trove | 04:36 | |
*** sharikapongubala has quit IRC | 04:40 | |
*** sharikapongubala has joined #openstack-trove | 04:40 | |
*** sgotliv_ has quit IRC | 04:48 | |
*** sharikapongubala has quit IRC | 04:50 | |
*** imandhan has joined #openstack-trove | 05:06 | |
*** imandhan has quit IRC | 05:10 | |
*** haomaiwang has quit IRC | 05:11 | |
*** haomaiwa_ has joined #openstack-trove | 05:11 | |
*** sharikapongubala has joined #openstack-trove | 05:16 | |
*** sharikapongubala has quit IRC | 05:20 | |
*** imandhan has joined #openstack-trove | 05:21 | |
*** sgotliv has joined #openstack-trove | 05:49 | |
openstackgerrit | Nikhil Manchanda proposed openstack/trove-integration: Fix openstack endpoint list usage for stable branches https://review.openstack.org/207193 | 05:55 |
*** k4n0 has joined #openstack-trove | 06:11 | |
*** imandhan has quit IRC | 06:19 | |
*** nexusz99 has joined #openstack-trove | 07:15 | |
*** dbennett has quit IRC | 07:59 | |
*** chlong has quit IRC | 08:04 | |
*** dbennett has joined #openstack-trove | 08:07 | |
*** erkules has quit IRC | 08:29 | |
*** erkules has joined #openstack-trove | 08:29 | |
*** dbennett has quit IRC | 08:54 | |
*** dbennett has joined #openstack-trove | 08:54 | |
*** tomblank has joined #openstack-trove | 09:00 | |
*** tosky has joined #openstack-trove | 09:32 | |
*** IanGovett has joined #openstack-trove | 09:45 | |
*** kmARC_ has joined #openstack-trove | 10:08 | |
*** kmARC has joined #openstack-trove | 10:08 | |
*** k4n0 has quit IRC | 10:43 | |
*** k4n0 has joined #openstack-trove | 10:43 | |
*** openstackgerrit has quit IRC | 10:45 | |
*** raginbajin has quit IRC | 10:45 | |
*** dhellmann has quit IRC | 10:45 | |
*** abramley has quit IRC | 10:45 | |
*** mingdang1 has quit IRC | 10:47 | |
*** openstackgerrit has joined #openstack-trove | 10:49 | |
*** raginbajin has joined #openstack-trove | 10:49 | |
*** dhellmann has joined #openstack-trove | 10:49 | |
*** abramley has joined #openstack-trove | 10:49 | |
*** kmARC has quit IRC | 11:00 | |
*** kmARC_ has quit IRC | 11:01 | |
*** pmackinn has joined #openstack-trove | 11:03 | |
*** kmARC_ has joined #openstack-trove | 11:05 | |
*** kmARC has joined #openstack-trove | 11:05 | |
*** kmARC has quit IRC | 11:18 | |
*** kmARC_ has quit IRC | 11:20 | |
*** k4n0 has quit IRC | 11:23 | |
*** haomaiwa_ has quit IRC | 11:37 | |
*** haomaiwa_ has joined #openstack-trove | 11:38 | |
*** saurabhs has joined #openstack-trove | 11:55 | |
*** chlong has joined #openstack-trove | 12:16 | |
*** radez is now known as radez_g0n3 | 12:23 | |
*** bapalm has joined #openstack-trove | 12:39 | |
*** logan2 has quit IRC | 13:11 | |
*** logan2 has joined #openstack-trove | 13:12 | |
*** radez_g0n3 is now known as radez | 13:18 | |
openstackgerrit | Zhao Lei proposed openstack/trove: Remove quotes from subshell call in tools/with_venv.sh https://review.openstack.org/210409 | 13:20 |
*** _crobertsrh is now known as crobertsrh | 13:46 | |
*** tomblank has quit IRC | 13:57 | |
*** mingdang1 has joined #openstack-trove | 14:27 | |
*** edmund has joined #openstack-trove | 14:29 | |
*** radez is now known as radez_g0n3 | 14:33 | |
*** edmund1 has joined #openstack-trove | 14:50 | |
*** edmund1 has quit IRC | 14:50 | |
*** edmund1 has joined #openstack-trove | 14:51 | |
*** edmund has quit IRC | 14:52 | |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add Redis backup/restore functionality https://review.openstack.org/185892 | 14:56 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add support for Redis replication https://review.openstack.org/203278 | 14:58 |
pmackinn | anyone care to explain test_volume_found to my small mind? | 15:01 |
*** sriram_tesora has joined #openstack-trove | 15:11 | |
*** mingdang1 has quit IRC | 15:21 | |
*** saurabhs has quit IRC | 15:26 | |
*** tomblank has joined #openstack-trove | 15:28 | |
SlickNik | pmackinn: are you asking about the int-test? | 15:31 |
pmackinn | SlickNik, yes...sems to fail for pgsql unless my setup is wrong somehow | 15:32 |
SlickNik | pmackinn: interesting — not sure why that'd happen. Do you have a stack trace of the failure? | 15:34 |
SlickNik | Maybe, it's possible that you have volume support turned off in your trove config under [pgsql], but not in your test config. | 15:35 |
pmackinn | SlickNik, paste.openstack.org/show/412196/ | 15:35 |
pmackinn | SlickNik, http://paste.openstack.org/show/412196/ | 15:36 |
*** logan2 has quit IRC | 15:38 | |
*** logan2 has joined #openstack-trove | 15:41 | |
SlickNik | pmackinn: interesting - from looking at the code, this should be getting set here https://github.com/openstack/trove/blob/master/trove/guestagent/dbaas.py#L103 | 15:42 |
pmackinn | retrying now | 15:43 |
* pmackinn pushes on the pull door | 15:43 | |
pmackinn | same deal | 15:44 |
SlickNik | But I wonder if there is a bug which is causing this test to fail — can you check your guestagent (or taskmanager) logs to see if there is an error corresponding to the call in https://github.com/openstack/trove/blob/d8d2d9f03aeac46e0e83c36c0f2487b4119371a5/trove/guestagent/datastore/experimental/postgresql/manager.py#L109 ? | 15:44 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add Redis backup/restore functionality https://review.openstack.org/185892 | 15:50 |
pmackinn | SlickNik, whiich conf should have volume_support = True? taskmanager? | 15:51 |
SlickNik | pmackinn: Yes, although from looking at the stack trace you sent me that doesn't seem to be the issue. | 15:54 |
pmackinn | SlickNik, are we saying it should be turned off for env+test for pgsql? or pgsql tests should work with it enabled? | 15:55 |
*** esmute_ has joined #openstack-trove | 16:09 | |
*** sgotliv has quit IRC | 16:12 | |
*** saurabhs has joined #openstack-trove | 16:27 | |
*** imandhan has joined #openstack-trove | 16:29 | |
SlickNik | pmackinn: The tests should work with pgsql + volumes enabled. | 16:30 |
*** bapalm has quit IRC | 16:31 | |
*** esmute_ has quit IRC | 16:33 | |
*** esmute_ has joined #openstack-trove | 16:33 | |
*** ashleighfarnham has joined #openstack-trove | 16:35 | |
saurabhs | SlickNik: upstream for trove integration projects http://logs.openstack.org/86/181186/15/check/gate-trove-functional-dsvm-mysql-kilo/84e4e58/ and http://logs.openstack.org/86/181186/15/check/gate-trove-functional-dsvm-mysql-juno/89c137e/ tests are failing is that a known problem or am I doing something wrong | 16:37 |
saurabhs | particularly this patch : https://review.openstack.org/#/c/181186/ | 16:37 |
SlickNik | saurabhs: They are failing on the stable branch tests | 16:37 |
SlickNik | You're not doing anything wrong. | 16:37 |
SlickNik | I'm looking into fixing the stable juno / kilo tests. | 16:38 |
SlickNik | saurabhs: https://review.openstack.org/#/c/207193/ still WIP | 16:38 |
saurabhs | SlickNik: thanks | 16:39 |
openstackgerrit | Sushil Kumar proposed openstack/trove-specs: Datastore Registration API spec https://review.openstack.org/188072 | 16:48 |
sushilkm | SlickNik, cp16net, vkmc I have updated the https://review.openstack.org/#/c/188072/ Datastore Registration API spec | 16:49 |
sushilkm | please review the same, thank you | 16:49 |
*** esmute_ has quit IRC | 16:51 | |
*** esmute_ has joined #openstack-trove | 16:53 | |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add Redis backup/restore functionality https://review.openstack.org/185892 | 16:54 |
*** sriram_tesora has quit IRC | 16:55 | |
*** esmute_ has quit IRC | 16:58 | |
*** esmute_ has joined #openstack-trove | 16:58 | |
*** exploreshaifali has joined #openstack-trove | 17:09 | |
*** radez_g0n3 is now known as radez | 17:15 | |
*** sharikapongubala has joined #openstack-trove | 17:15 | |
sharikapongubala | can i please have some reviews on these? :) https://review.openstack.org/#/c/196913/ https://review.openstack.org/#/c/197360/ | 17:16 |
*** rjaiswal has joined #openstack-trove | 17:19 | |
*** ashleighfarnham has quit IRC | 17:29 | |
edmondk | vkmc, we could go ahead and close this bug out, as it hasn't been reproduced: https://bugs.launchpad.net/trove/+bug/1395231 | 17:33 |
openstack | Launchpad bug 1395231 in Trove "Database Create Instance Limit does not work, shows Quota exceeded though quota is available" [Low,Invalid] - Assigned to Edmond Kotowski (ekotowski) | 17:33 |
vkmc | edmondk, yup :) | 17:33 |
vkmc | sushilkm, reviewed, thanks! | 17:33 |
*** radez is now known as radez_g0n3 | 17:42 | |
*** radez_g0n3 is now known as radez | 17:46 | |
*** ashleighfarnham has joined #openstack-trove | 17:47 | |
*** sriram_tesora has joined #openstack-trove | 18:03 | |
*** sharikapongubala has quit IRC | 18:03 | |
*** tosky has quit IRC | 18:03 | |
sushilkm | thank u vkmc | 18:03 |
*** saurabhs has quit IRC | 18:05 | |
crobertsrh | Any chance that any of you have managed to have the trove services start up in devstack by adding enable_plugin trove git://git.openstack.org/openstack/trove to local.conf? I have added it, it gets cloned, but I see no evidence of the processes trying to start. | 18:05 |
*** sharikapongubala has joined #openstack-trove | 18:06 | |
*** saurabhs has joined #openstack-trove | 18:07 | |
*** bhunter71 has joined #openstack-trove | 18:09 | |
*** kmARC has joined #openstack-trove | 18:11 | |
*** esmute_ has quit IRC | 18:11 | |
*** kmARC_ has joined #openstack-trove | 18:12 | |
*** bhunter71 has quit IRC | 18:13 | |
edmondk | pmackinn, I am getting the exact error on test_volume_found for mysql, looking into it now | 18:15 |
pmackinn | oh cool thx edmondk | 18:15 |
*** pmackinn is now known as pmackinn-afk | 18:16 | |
*** sharikapongubala has quit IRC | 18:18 | |
*** radez is now known as radez_g0n3 | 18:19 | |
*** radez_g0n3 is now known as radez | 18:21 | |
*** bapalm has joined #openstack-trove | 18:26 | |
*** sharikapongubala has joined #openstack-trove | 18:28 | |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add generic int-test classes https://review.openstack.org/203274 | 18:32 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add Redis backup/restore functionality https://review.openstack.org/185892 | 18:34 |
openstackgerrit | Petr Malik proposed openstack/trove: Add user actions int-tests https://review.openstack.org/210094 | 18:36 |
*** bapalm has quit IRC | 18:38 | |
*** ashleighfarnham has quit IRC | 18:38 | |
SlickNik | crobertsrh: I believe you need to enable the services explicitly as well — one sec, let me check on it. | 18:41 |
SlickNik | i.e. enable the plugin + enable the trove services. | 18:41 |
openstackgerrit | Merged openstack/trove-specs: Datastore Registration API spec https://review.openstack.org/188072 | 18:42 |
*** sriram_tesora has quit IRC | 18:47 | |
*** bapalm has joined #openstack-trove | 18:48 | |
*** bapalm has quit IRC | 18:48 | |
crobertsrh | ah, thanks SlickNik | 18:49 |
*** esmute_ has joined #openstack-trove | 18:50 | |
SlickNik | crobertsrh: you'll want to add something like https://github.com/openstack/trove-integration/blob/master/scripts/localrc.rc#L19 your localrc / local.conf | 18:53 |
crobertsrh | Right. Somehow, that wasn't clear from the docs I was looking at. | 18:54 |
*** sharikapongubala has quit IRC | 18:54 | |
*** sharikapongubala has joined #openstack-trove | 18:54 | |
*** kmARC has quit IRC | 18:55 | |
*** kmARC_ has quit IRC | 18:55 | |
*** radez is now known as radez_g0n3 | 18:56 | |
*** radez_g0n3 is now known as radez | 19:00 | |
*** ashleighfarnham has joined #openstack-trove | 19:00 | |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add support for Redis replication https://review.openstack.org/203278 | 19:02 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add generic int-test classes https://review.openstack.org/203274 | 19:02 |
*** sharikapongubala has quit IRC | 19:07 | |
*** sharikapongubala has joined #openstack-trove | 19:11 | |
*** bhunter71 has joined #openstack-trove | 19:11 | |
*** sharikapongubala has quit IRC | 19:14 | |
SlickNik | pmackinn-afk: Was looking at the volume bug we were talking about earlier with edmondk. Looks like we might have a bug upstream (race condition) that's causing this. Basically the instance may go to ACTIVE before the 'prepare' call has been run on the guest. Will | 19:19 |
cp16net | yikes | 19:22 |
cp16net | SlickNik: oh... i bet thats because the datastore is set to start on boot | 19:23 |
cp16net | maybe? | 19:23 |
cp16net | i noticed something where on boot percona would start up after the perpare had started and shutdown mysql. | 19:24 |
SlickNik | cp16net: Yes, that's what my guess was. | 19:24 |
cp16net | it would start it back up | 19:24 |
SlickNik | The heartbeat goes out as "Up" even before prepare has completed. | 19:24 |
cp16net | so in the image elements i added the line to disable mysql from starting | 19:24 |
cp16net | i could see that happening | 19:24 |
SlickNik | Causing the guest to flip to ACTIVE for a short time even before the volume is ready. | 19:25 |
cp16net | the guest starts mysql in the prepare any way so i figure thats good enough | 19:25 |
cp16net | if mysql isnt started on boot | 19:25 |
cp16net | its started at the end of install_nessesary_packages | 19:26 |
cp16net | after that i stopped running into that race condition | 19:26 |
vkmc | SlickNik, did ya run those tests in redstack? | 19:27 |
SlickNik | vkmc: Yes, in redstack. | 19:28 |
vkmc | SlickNik, I ran them and hit another issues, but not that one in particular | 19:29 |
vkmc | maybe it is not always reproducible? | 19:29 |
*** exploreshaifali has quit IRC | 19:34 | |
SlickNik | vkmc: Yes, it's a race condition — the issue manifests only sometimes. | 19:35 |
vkmc | ops yes | 19:36 |
*** esmute_ has quit IRC | 19:36 | |
SlickNik | basically dependent if the loopingcall which checks for ACTIVE happens to check at the time when the instance flips into ACTIVE erroneously (which happens for a short window) | 19:36 |
pmalik | Hi. Could we possibly push the configuraion race fix forward. I think it's already in a pretty good shape, just needing some pluses. :-) Thanks. P. | 19:51 |
pmalik | https://review.openstack.org/#/c/198891/ | 19:51 |
pmalik | ^^^^^ SlickNik peterstac cp16net vkmc _amrith_ | 19:51 |
openstackgerrit | Morgan Jones proposed openstack/python-troveclient: Redis Clustering Initial Implementation https://review.openstack.org/203701 | 19:56 |
openstackgerrit | Morgan Jones proposed openstack/trove: Redis Cluster Initial Implementation https://review.openstack.org/203702 | 19:56 |
*** chlong has quit IRC | 20:06 | |
*** pmackinn-afk is now known as pmackinn | 20:10 | |
*** sharikapongubala has joined #openstack-trove | 20:10 | |
pmackinn | SlickNik, i found my race cond w/ kilo pgsql...reproduces everytime with a home-rolled pgsql image but never saw it before w/ mariadb or mongodb | 20:19 |
saurabhs | sushilkm, cp16net, amrith Please review https://review.openstack.org/#/c/181186/. I think I have addressed most of the recent/old comments on this patch and was able to test it end to end. | 20:33 |
saurabhs | I added a seperate element to treat DATASTORE_PKG_LOCATION as file/directory/url so it should address the concerns everybody had about treating it as a file/directory | 20:34 |
*** logan2 has quit IRC | 20:56 | |
edmondk | pmackinn, we think we nailed down the issue that is happening | 20:57 |
edmondk | pmackinn, for mysql case mysql is started before prepare is finished which causes a fake ACTIVE state to appear, then mysql is stopped in prepare and prepare sets the status back to BUILD, then prepare finished and sets the status back to ACTIVE | 20:59 |
edmondk | pmackinn, for the trove volume test it is run when it the first fake ACTIVE status appears but when the test does the instances.get(id) the status is back to actual BUILD state which the volume['used'] won't show up | 21:00 |
pmackinn | edmondk, volume['used'] won't persist across these states i take it | 21:01 |
edmondk | SlickNik and I think this issue has been in since the conductor was added but has only shown up recently because of the oslo messaging addition | 21:01 |
edmondk | pmackinn, the volume does indeed exist but since it's in BUILD the trove API will not return the volume information | 21:02 |
edmondk | since it is a race condition the speed of your environment also will matter | 21:03 |
edmondk | I am running on a powerful system that is faster than our upstream integration test system which could be the reason we both see the issue locally and not on the gate | 21:03 |
pmackinn | edmondk, so none of http://bit.ly/1OVUJRM can come back through the API even though it must have been previously computed when ACTIVE | 21:04 |
edmondk | yeah I think so | 21:07 |
pmackinn | edmondk, gotcha | 21:08 |
*** crobertsrh is now known as _crobertsrh | 21:09 | |
edmondk | because we check http://paste.openstack.org/show/412234/ on the view | 21:09 |
edmondk | I have to check but when its in build volume_used maybe false | 21:09 |
pmackinn | yeah trying to find where that logic is | 21:16 |
pmackinn | SlickNik, edmondk: shouldn't the volume attributes ahead of 'used' also thrown an error? http://bit.ly/1OVXH8D | 21:21 |
SlickNik | pmackinn: No it's completely possible that they don't. | 21:25 |
SlickNik | pmackinn: The 'size' comes from the actual cinder volume which is created and provisioned. | 21:26 |
SlickNik | pmackinn: It's only 'used' (i.e. the actual used space on the volume) which needs to be queried from the guest and sent back. | 21:26 |
SlickNik | pmackinn: ANd since the guest is not fully active at this point, it's missing. | 21:26 |
pmackinn | SlickNik, considered missing :-) | 21:27 |
pmackinn | SlickNik, ok makes more sense | 21:27 |
pmackinn | 'used' is special in the API | 21:27 |
SlickNik | yeah, used' is special in that dict since we need to actively query the guest to get the value for it. | 21:28 |
pmackinn | so ACTIVE->BUILD and thrown away by ga, set to None, etc. | 21:29 |
*** logan2 has joined #openstack-trove | 21:39 | |
openstackgerrit | Craig Vyvial proposed openstack/trove: Percona Xtradb Cluster implementation https://review.openstack.org/208073 | 21:59 |
openstackgerrit | Craig Vyvial proposed openstack/trove: Percona Xtradb Cluster implementation https://review.openstack.org/208073 | 22:02 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add generic int-test classes https://review.openstack.org/203274 | 22:03 |
openstackgerrit | Ishita Mandhan proposed openstack/trove: Code refactoring for couchdb https://review.openstack.org/210170 | 22:04 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add Redis backup/restore functionality https://review.openstack.org/185892 | 22:05 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add user actions int-tests https://review.openstack.org/210094 | 22:06 |
openstackgerrit | Peter Stachowski proposed openstack/trove: Add support for Redis replication https://review.openstack.org/203278 | 22:07 |
pmackinn | whoa, peterstac ftw | 22:16 |
openstackgerrit | Ishita Mandhan proposed openstack/trove: Code refactoring for couchdb https://review.openstack.org/210170 | 22:24 |
openstackgerrit | Sharika Pongubala proposed openstack/trove: Modifying Vertica clusters to have a master node https://review.openstack.org/204271 | 22:35 |
*** radez is now known as radez_g0n3 | 22:36 | |
*** edmund1 has quit IRC | 22:39 | |
*** ashleigh_ has joined #openstack-trove | 22:42 | |
*** ashleighfarnham has quit IRC | 22:49 | |
*** imandhan has quit IRC | 22:49 | |
*** georgelorch has quit IRC | 22:49 | |
*** pmackinn has quit IRC | 22:49 | |
*** nexusz99 has quit IRC | 22:49 | |
openstackgerrit | Sharika Pongubala proposed openstack/trove: Modifying Vertica clusters to have a master node https://review.openstack.org/204271 | 22:50 |
*** nexusz99 has joined #openstack-trove | 22:52 | |
*** georgelorch has joined #openstack-trove | 22:53 | |
openstackgerrit | Petr Malik proposed openstack/trove: Redis Cluster Initial Implementation https://review.openstack.org/203702 | 22:55 |
edmondk | pmackinn, our suspicion was correct. I wrote a quick fix in the set_status on the guest to check when in building if it was the trove prepare method calling heartbeat of the datastore and I was able to get the volume test to pass | 22:57 |
edmondk | Here is the current quick fix I made: http://paste.openstack.org/show/412242/ | 22:58 |
edmondk | the first if status == instance.ServiceStatuses.Building: section | 22:58 |
edmondk | Discussing with SlickNik for a cleaner fix and I am going to try adding a check for if /var/lib/data exists send a heartbeat otherwise do not | 23:01 |
SlickNik | edmondk: Yeah, I think that's a bit cleaner | 23:03 |
SlickNik | You can read the data dir from the CONF file, so we don't have to explicitly rely on /var/lib/data | 23:04 |
SlickNik | But that way the hearbeat won't be explicitly sent out until the prepare call runs. | 23:05 |
edmondk | Created bug for issue: https://bugs.launchpad.net/trove/+bug/1482795 | 23:10 |
openstack | Launchpad bug 1482795 in Trove "On create instance the instance status alternates between BUILD->ACTIVE->BUILD->ACTIVE" [High,New] - Assigned to Edmond Kotowski (ekotowski) | 23:10 |
edmondk | I will test the /var/lib/data fix | 23:12 |
openstackgerrit | Sharika Pongubala proposed openstack/trove: Root enablement for Vertica clusters/instances https://review.openstack.org/196913 | 23:15 |
*** logan2 has quit IRC | 23:15 | |
openstackgerrit | Sharika Pongubala proposed openstack/trove: Modifying Vertica clusters to have a master node https://review.openstack.org/204271 | 23:16 |
openstackgerrit | Nikhil Manchanda proposed openstack/trove-integration: Fix openstack endpoint list usage for stable branches https://review.openstack.org/207193 | 23:26 |
*** nexusz99 has quit IRC | 23:35 | |
*** imandhan has joined #openstack-trove | 23:42 | |
openstackgerrit | Nikhil Manchanda proposed openstack/trove-integration: Fix openstack endpoint list usage for stable branches https://review.openstack.org/207193 | 23:53 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!