Tuesday, 2017-03-07

*** ianychoi has quit IRC00:10
*** ianychoi has joined #openstack-manila00:11
*** delatte has quit IRC00:13
*** gcb has quit IRC00:26
openstackgerritPeter Wang proposed openstack/manila master: Handle ssl for VNX manila driver  https://review.openstack.org/44036401:11
openstackgerritOpenStack Proposal Bot proposed openstack/manila-ui master: Updated from global requirements  https://review.openstack.org/44087201:58
*** mtanino_ has quit IRC02:09
*** erlon has quit IRC02:15
openstackgerritVictoria Martinez de la Cruz proposed openstack/manila-image-elements master: Adds support for NFS Ganesha  https://review.openstack.org/41150002:27
*** akapil has joined #openstack-manila02:40
*** akapil has quit IRC02:44
openstackgerritMerged openstack/puppet-manila master: Add service workers API config option  https://review.openstack.org/44188003:05
*** bswartz has quit IRC03:25
*** gouthamr has quit IRC03:55
*** bswartz has joined #openstack-manila04:36
*** akapil has joined #openstack-manila04:38
*** akapil has quit IRC04:42
*** dsariel has quit IRC05:03
*** dsariel has joined #openstack-manila05:37
*** gcb has joined #openstack-manila06:28
*** nkrinner_afk is now known as nkrinner06:40
*** tuanluong has joined #openstack-manila06:51
*** lpetrut has joined #openstack-manila06:58
openstackgerritTommyLike proposed openstack/manila master: Fix: [LVM] Can't extend share when snapshots exist  https://review.openstack.org/43131507:10
*** jprovazn has joined #openstack-manila07:16
*** lpetrut has quit IRC07:43
*** gcb has quit IRC07:45
*** pcaruana has joined #openstack-manila07:50
*** makowals_ has quit IRC07:55
*** gcb has joined #openstack-manila07:58
*** makowals has joined #openstack-manila07:58
*** akapil has joined #openstack-manila08:00
*** makowals has quit IRC08:01
*** akapil has quit IRC08:05
*** akapil has joined #openstack-manila08:05
*** lpetrut has joined #openstack-manila08:12
*** makowals has joined #openstack-manila08:14
*** makowals has quit IRC08:16
*** makowals has joined #openstack-manila08:41
*** makowals has quit IRC08:51
*** makowals has joined #openstack-manila09:04
*** akapil has quit IRC09:14
*** lpetrut has quit IRC09:15
*** akapil has joined #openstack-manila09:15
*** lpetrut has joined #openstack-manila09:17
*** a-pugachev has joined #openstack-manila09:20
openstackgerritkavitha h r proposed openstack/manila master: The python version is added Python 3 and 3.5 version was missing.  https://review.openstack.org/44233509:21
*** faizy has joined #openstack-manila09:22
*** belmoreira has joined #openstack-manila09:24
tommylikehuping vponomaryov09:37
tommylikehuhey vponomaryov, are you around?09:37
*** ociuhandu has joined #openstack-manila10:10
*** rraja has joined #openstack-manila10:22
*** gcb has quit IRC10:26
*** openstackgerrit has quit IRC10:33
*** ganso has joined #openstack-manila10:42
vponomaryovtommylikehu: pong10:44
*** akapil has quit IRC10:45
tommylikehuvponomaryov: thanks, could you take some time on this patch? https://review.openstack.org/#/q/is:starred10:47
vponomaryovьфниу нщг рфму ьщку зкщзук дштл,10:48
tommylikehu...10:48
vponomaryovmaybe you have more proper link?10:48
tommylikehusorry10:48
tommylikehuhttps://review.openstack.org/#/c/431315/10:48
gansotommylikehu: Hi TommyLike, could you please elaborate on your comment in the gerrit change above ^ ?10:52
gansotommylikehu: it seems you are trying to say that there is concurrency in that test10:52
tommylikehuganso: I think so, you can tell from the log, we collect all the snapshots and then detect the original volume one by one, but the failed snapshot is deleted just few seconds before, so that could happen10:54
*** akapil has joined #openstack-manila10:55
gansotommylikehu: that shouldn't be allowed to happen10:57
tommylikehuganso: what do you exactly mean by shouldn't be allowed to happen10:58
gansotommylikehu: a request to delete the LVM shouldn't be accepted if it is already in a reverting state, and vice-versa10:59
gansotommylikehu: shouldn't be allowed to accept a request to revert if it is in a deleting state10:59
tommylikehuganso: no10:59
tommylikehuganso: we are not reverting you mean deleting?11:00
tommylikehus/deleting/extending11:00
gansotommylikehu: sorry s/reverting/extending11:00
*** akapil has quit IRC11:00
gansotommylikehu: I had "revert-to-snapshot" on my mind11:01
tommylikehuok, wait a second11:01
gansotommylikehu: from what I see, there is currently no control of that because extending is an operation on a share, and deleting is an operation on the snapshot11:01
tommylikehuganso: please take a look at line 13211:02
tommylikehuhttps://review.openstack.org/#/c/431315/16/manila/share/drivers/lvm.py11:02
gansotommylikehu: so it looks a bit more plausible that the snapshot shouldn't be allowed to be deleted if the share is busy extending11:02
tommylikehuganso: actually we collect all the snapshots, not related to this volume11:02
tommylikehuand choose the correct ones by lvdisplay origin command11:03
tommylikehuso we could have other snapshots11:03
gansotommylikehu: oh I see, so, after you listed the snapshot, it goes missing, and the next command fails, correct?11:03
tommylikehuganso: you are right11:03
tommylikehuganso: this error is caused by your comments that we should not pass the related snapshots from manila to driver :)11:04
*** akapil has joined #openstack-manila11:05
gansotommylikehu: hold on I gotta refresh my memory11:05
gansotommylikehu: I need to remember why I said that11:05
gansotommylikehu: I am looking at previous patchsets11:05
*** akapil has quit IRC11:06
tommylikehuganso: I am just kidding, your comments are right11:06
*** akapil has joined #openstack-manila11:06
gansotommylikehu: lol, ok, then maybe a try/except block could work?11:06
gansotommylikehu: like, if the snapshot does not exist anymore, you shouldn't have to worry about it11:06
gansotommylikehu: if it is not related, that's fine, if it was related, it is gone and unmounted already11:07
gansotommylikehu: I don't like the idea of preventing snapshot deletion when share is extending, and we would still need the try/except block for the unrelated snapshots that may go missing11:07
tommylikehuganso: I am sorry where does your idea ' preventing snapshot deletion when share is extending' comes from, Is that one of my suggestions?11:09
gansotommylikehu: no, I said I do not like that idea very much11:10
gansotommylikehu: it came from me11:10
tommylikehuganso: ok~11:10
gansotommylikehu: what do you think about the try/except block?11:12
tommylikehuganso: to be clear the snapshots are different in this case. but your concern could be a bug for this feature.11:12
tommylikehuganso: a valid bug11:12
gansotommylikehu: the main problem about that idea is that it wouldn't make sense for other drivers11:12
gansotommylikehu: why a valid bug? are you saying we cannot address this bug in other ways?11:12
tommylikehuganso: If we call extending share and deleting snapshot of this share at the same time, does this could happen?11:14
gansotommylikehu: it could happen, yes, why prevent it?11:14
gansotommylikehu: it seems only LVM has this bug, we should address it in the driver if possible11:15
tommylikehuganso: any suggestion on this?11:25
gansotommylikehu: ok I had suggested a try/except block, but I see the command does not fail11:30
tommylikehuganso: try/excerpt block for what?11:31
gansotommylikehu: for a command that would fail11:31
gansotommylikehu: wait, I am very confused11:32
gansotommylikehu: the link you gave me http://logs.openstack.org/15/431315/15/check/gate-manila-tempest-minimal-dsvm-lvm-ubuntu-xenial/4e4ec52/logs/screen-m-shr.txt.gz?#_2017-03-07_02_44_27_92711:32
gansotommylikehu: shows that the command succeeded, but below the command fails11:32
tommylikehuganso lvremove command succeed and lvdisplay command failed11:33
gansotommylikehu: oh great! so we can use the try/except block11:33
gansotommylikehu: for the lvdisplay command11:33
gansotommylikehu: if it fails, you could just skip11:33
gansotommylikehu: what do you think about that?11:33
tommylikehuganso: you mean: https://review.openstack.org/#/c/431315/16/manila/share/drivers/lvm.py line110?11:34
gansotommylikehu: yes11:34
tommylikehuganso: it's already done11:34
gansotommylikehu: oh so that log is from a previous patchset11:35
*** openstackgerrit has joined #openstack-manila11:35
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [DNM] Squash SGS member and SS instances DB tables  https://review.openstack.org/44211511:35
tommylikehuganso: yes11:35
gansotommylikehu: I am seeing the diff now11:36
gansotommylikehu: ok cool11:36
tommylikehuhey vponomaryov I see a lot of patches with a DNM label. what does DNM mean?11:36
vponomaryovDo Not Merge11:36
tommylikehuvponomaryov: thanks11:37
openstackgerritMerged openstack/manila master: Start NFS and SMB services on fedora platforms  https://review.openstack.org/44207311:37
gansotommylikehu: I'd like tbarron's centOS patch to merge before +2'ing your patch11:37
gansotommylikehu: there we go ^11:37
gansotommylikehu: ok, I'll recheck your patch, and centOS job should pass11:38
tommylikehuganso: thanks11:38
gansotommylikehu: thanks for working on this fix11:38
vponomaryovtommylikehu:  about https://review.openstack.org/#/c/431315/16/manila_tempest_tests/tests/api/test_shares_actions.py11:46
vponomaryovtommylikehu: why do you use "STATUS_EXTENDING_ERROR"  there?11:46
vponomaryovtommylikehu: just to reduce wait time?11:46
vponomaryovtommylikehu: if error appears?11:47
vponomaryovif yes, then line 366 in https://review.openstack.org/#/c/431315/16/manila_tempest_tests/services/share/v2/json/shares_client.py already solves this case11:47
vponomaryovtommylikehu: so, for the moment, I do not see the reason of update of https://review.openstack.org/#/c/431315/16/manila_tempest_tests/services/share/v2/json/shares_client.py11:48
vponomaryovganso ^11:49
gansovponomaryov: he wants to accept ["Available", "Extending_error"]11:50
vponomaryovwhy?11:50
vponomaryovmy question was and is "why"11:50
gansovponomaryov: because of additional checks in the test11:54
gansovponomaryov: you could take a look at the previous patchset comments11:54
gansovponomaryov: as it was before, it was accepting only "available". So it the wait function was either going to fail the test if the status was "extending_error", or if status reaches available, there was no need for further validation11:55
gansovponomaryov: I suggested that tommylikehu could remove the validation, or add a way to the code to continue if status was "extending_error". He chose to implement the latter11:56
vponomaryovganso: validation of size is a must anyway11:58
vponomaryovsecond status is redundant11:58
vponomaryovchanges to client redundant too11:58
vponomaryovexpected success status only one - available11:58
vponomaryovas it is now11:58
vponomaryovganso: file https://review.openstack.org/#/c/431315/16/manila_tempest_tests/tests/api/test_shares_actions.py does not have comments related to my question11:59
gansovponomaryov: why is it a must? it is an API contract that if status changes from extending to available, the size will be changed regardless11:59
vponomaryovganso: in none of patch sets11:59
gansovponomaryov: it does https://review.openstack.org/#/c/431315/8/manila_tempest_tests/tests/api/test_shares_actions.py12:00
vponomaryovganso: in https://review.openstack.org/#/c/431315/8/manila_tempest_tests/tests/api/test_shares_actions.py your comment for line 593 is false12:01
vponomaryovit will not timeout12:01
vponomaryovso, this variant is correct12:01
gansovponomaryov: yes it will fail, because it has "error" in the status12:01
vponomaryovit will not TIMEOUT12:02
gansovponomaryov: Valeriy, if it fails, it will not proceed to the validations12:02
vponomaryovit will fail immediatly12:02
gansovponomaryov: yes I noticed it will not timeout, but it will fail, and will not proceed to the validations12:03
gansovponomaryov: so the only case it will proceed to the validations is of success, and in that case, API contract already guarantees size increase, and this is covered in other tests12:03
gansovponomaryov: so size validation would be redundant12:03
vponomaryovganso: why is that?12:04
vponomaryovganso: you check status first, then size as second12:04
vponomaryovganso: and fail on any of them12:04
gansovponomaryov: no need to check size if it transitions to "available"12:04
gansovponomaryov: it will not fail if wait function succeeds12:04
vponomaryovganso: success of status has nothing to prove about size12:05
vponomaryovat all12:05
vponomaryovit is 2 separate things12:05
gansovponomaryov: that behavior is covered in another test12:05
vponomaryovand comment for line 604 https://review.openstack.org/#/c/431315/8/manila_tempest_tests/tests/api/test_shares_actions.py is false too12:06
gansovponomaryov: it is part of the feature workflow to increase the size when transitioning from extending to available12:06
vponomaryovganso: it is test12:08
vponomaryovits goal is to validate12:08
gansovponomaryov: feel free to add comments in the patch to change it how you'd like then12:08
vponomaryovpresence of snapshot may touch different code and we can have bug only here12:08
vponomaryovso, check for size is a must and second status is redundant12:09
vponomaryovtommylikehu: also, you fix case with "mounted" snapshot12:10
vponomaryovtommylikehu: would be good to add such test12:10
*** tuanluong has quit IRC12:11
*** faizy has quit IRC12:12
*** dgonzalez has quit IRC12:22
*** catintheroof has joined #openstack-manila12:24
*** erlon has joined #openstack-manila12:43
*** chlong has joined #openstack-manila13:00
*** porrua has joined #openstack-manila13:01
*** faizy has joined #openstack-manila13:01
*** dustins has joined #openstack-manila13:02
*** porrua_ has joined #openstack-manila13:02
*** esker has joined #openstack-manila13:03
*** porrua has quit IRC13:06
*** dsariel has quit IRC13:06
*** esker has quit IRC13:08
*** eharney has quit IRC13:12
*** faizy has quit IRC13:20
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [Share Groups] Squash SGS member and SS instances DB tables  https://review.openstack.org/43385413:24
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [DNM] test share groups  https://review.openstack.org/43029613:28
*** tommylikehu_ has joined #openstack-manila13:37
*** gouthamr has joined #openstack-manila13:43
*** lpetrut has quit IRC13:49
*** crushil has joined #openstack-manila13:55
*** gouthamr has quit IRC13:59
*** gouthamr has joined #openstack-manila14:02
*** rraja has quit IRC14:03
*** lpetrut has joined #openstack-manila14:09
*** xyang1 has joined #openstack-manila14:15
*** rraja has joined #openstack-manila14:18
*** ociuhandu has quit IRC14:18
*** ociuhandu has joined #openstack-manila14:29
*** cknight has joined #openstack-manila14:47
*** mtanino has joined #openstack-manila14:49
*** Yogi1 has joined #openstack-manila14:49
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [Share Groups] Squash SGS member and SS instances DB tables  https://review.openstack.org/43385414:50
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [DNM] test share groups  https://review.openstack.org/43029614:51
*** sandanar has joined #openstack-manila14:59
*** pcaruana has quit IRC15:09
openstackgerritValeriy Ponomaryov proposed openstack/manila master: Update share replicas after promotion in proper order  https://review.openstack.org/43889215:13
*** pcaruana has joined #openstack-manila15:23
*** dgonzalez has joined #openstack-manila15:24
*** sandanar has quit IRC15:33
*** tommylikehu_ has quit IRC15:33
*** dsariel has joined #openstack-manila15:43
*** eharney has joined #openstack-manila15:45
*** jprovazn has quit IRC15:49
*** nkrinner is now known as nkrinner_afk15:53
*** lpetrut has quit IRC16:03
*** akapil_ has joined #openstack-manila16:17
*** akapil_ has quit IRC16:17
*** akapil has quit IRC16:20
*** belmoreira has quit IRC16:22
*** crushil has quit IRC16:22
*** pcaruana has quit IRC16:26
*** esker has joined #openstack-manila16:29
*** esker has quit IRC16:29
gouthamrvponomaryov: https://review.openstack.org/#/c/438892/3/manila/share/manager.py you live and you learn16:29
gouthamr:P16:29
*** rraja has quit IRC16:35
*** eharney_ has joined #openstack-manila16:36
*** eharney has quit IRC16:38
*** eharney_ is now known as eharney16:39
*** JoseMello has joined #openstack-manila16:39
gouthamrvponomaryov: mind if i submit the next patch?16:45
vponomaryovgouthamr: in my commit?16:45
gouthamrvponomaryov: yes.. we're fixing the bug eventually, want to do it right.. i don't agree with you that it's unrelated. the bug existed because of all the complicated things shoved into one long inseparable flow16:46
vponomaryovgouthamr: I mind, actually16:47
gouthamrvponomaryov: haha. okay. feel free to merge yours then, i'll commit the refactor16:47
*** crushil has joined #openstack-manila16:48
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [Tempest] Fix concurrency in test with listing share servers  https://review.openstack.org/44263317:23
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [Tempest] Fix concurrency in test with listing share servers  https://review.openstack.org/44263317:26
*** david-lyle_ has joined #openstack-manila17:52
*** david-lyle has quit IRC17:54
*** david-lyle_ is now known as david-lyle17:55
*** david-lyle has quit IRC18:04
*** david-lyle has joined #openstack-manila18:05
*** a-pugachev has quit IRC18:06
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [Tempest] Fix concurrency in test with listing share servers  https://review.openstack.org/44263318:22
*** david-lyle has quit IRC18:23
*** david-lyle has joined #openstack-manila18:25
*** david-lyle has quit IRC18:29
*** david-lyle has joined #openstack-manila18:36
*** eharney has quit IRC18:41
*** eharney has joined #openstack-manila18:45
*** david-lyle has quit IRC19:04
*** david-lyle has joined #openstack-manila19:06
*** crushil has quit IRC19:08
*** crushil has joined #openstack-manila19:18
openstackgerritAndrea Frittoli proposed openstack/manila master: Local copy of scenario test base class  https://review.openstack.org/44271919:18
*** markstur has joined #openstack-manila19:19
*** gouthamr has quit IRC20:01
*** dsariel has quit IRC20:07
*** dustins has quit IRC20:15
*** dsariel has joined #openstack-manila20:20
*** gouthamr has joined #openstack-manila20:29
*** JoseMello has quit IRC20:35
*** dustins has joined #openstack-manila20:55
*** lpetrut has joined #openstack-manila21:07
*** eharney has quit IRC21:13
*** eharney has joined #openstack-manila21:29
*** Yogi1 has quit IRC21:36
*** gouthamr has quit IRC21:37
*** cknight has quit IRC21:51
*** gouthamr has joined #openstack-manila21:59
*** crushil has quit IRC22:00
*** porrua_ has quit IRC22:06
*** cknight has joined #openstack-manila22:13
*** cknight1 has joined #openstack-manila22:19
*** cknight has quit IRC22:21
*** ganso has quit IRC22:25
*** lpetrut has quit IRC22:26
*** cknight1 has quit IRC22:28
openstackgerritAlex Schultz proposed openstack/puppet-manila master: Prepare Pike metadata  https://review.openstack.org/44282422:34
*** catintheroof has quit IRC22:41
*** catintheroof has joined #openstack-manila22:42
*** catintheroof has quit IRC22:47
*** dustins has quit IRC22:52
*** erlon has quit IRC23:05
*** eharney has quit IRC23:16
*** ociuhandu has quit IRC23:17
*** xyang1 has quit IRC23:27
*** jungleboyj has joined #openstack-manila23:31
*** catintheroof has joined #openstack-manila23:32
*** ociuhandu has joined #openstack-manila23:40

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