| *** yangyapeng has quit IRC | 00:01 | |
| *** qeas has joined #openstack-manila | 00:10 | |
| qeas | Hi everyone, after a recent change to devstack I am getting this error when trying to stack.sh http://paste.openstack.org/show/490443/. Using the same local.conf that worked previously, any suggestions? Thanks in advance. | 00:11 |
|---|---|---|
| *** eharney has quit IRC | 00:18 | |
| *** vbellur has quit IRC | 00:20 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Ensure Snapshots across replicas https://review.openstack.org/292659 | 00:21 |
| *** docaedo has quit IRC | 00:31 | |
| *** docaedo has joined #openstack-manila | 00:32 | |
| *** qeas has quit IRC | 00:33 | |
| tbarron | qeas: I'm going offline in a moment, but: did devstack change? (2) did your manila.conf change? or (3) did the manila codebase change? | 00:35 |
| *** furlongm has joined #openstack-manila | 00:35 | |
| tbarron | ^^ looking at the stack trace, it seems to lead back to CONF.migration_tmp_location ... | 00:35 |
| *** jasonsb has joined #openstack-manila | 00:37 | |
| *** lpetrut has joined #openstack-manila | 00:38 | |
| *** lpetrut has quit IRC | 00:42 | |
| *** yangyapeng has joined #openstack-manila | 01:06 | |
| *** yangyapeng has quit IRC | 01:08 | |
| *** a_ta has joined #openstack-manila | 01:09 | |
| *** a_ta has quit IRC | 01:13 | |
| *** chlong has quit IRC | 01:26 | |
| *** yangyapeng has joined #openstack-manila | 01:28 | |
| *** baojg has joined #openstack-manila | 01:43 | |
| *** amrith is now known as _amrith_ | 01:49 | |
| *** vbellur has joined #openstack-manila | 02:01 | |
| *** houming has joined #openstack-manila | 02:31 | |
| *** yangyapeng has quit IRC | 02:52 | |
| *** yangyape_ has joined #openstack-manila | 02:52 | |
| *** a_ta has joined #openstack-manila | 02:53 | |
| *** houming_ has joined #openstack-manila | 02:55 | |
| *** houming has quit IRC | 02:56 | |
| *** houming_ is now known as houming | 02:56 | |
| *** a_ta has quit IRC | 02:59 | |
| *** merooney has quit IRC | 03:09 | |
| *** esker has quit IRC | 03:13 | |
| *** mtanino has quit IRC | 03:17 | |
| *** erlon has quit IRC | 03:27 | |
| *** chlong has joined #openstack-manila | 03:30 | |
| *** kaisers has quit IRC | 03:36 | |
| *** houming has quit IRC | 03:48 | |
| *** houming_ has joined #openstack-manila | 03:48 | |
| *** kaisers has joined #openstack-manila | 03:50 | |
| *** martyturner has joined #openstack-manila | 04:04 | |
| *** houming_ has quit IRC | 04:06 | |
| *** martyturner has quit IRC | 04:09 | |
| *** a_ta has joined #openstack-manila | 04:56 | |
| *** a_ta has quit IRC | 05:00 | |
| *** furlongm has quit IRC | 06:00 | |
| *** furlongm has joined #openstack-manila | 06:05 | |
| openstackgerrit | Ramana Raja proposed openstack/manila: glusterfs volume layout: take care of deletion of DOA shares https://review.openstack.org/292177 | 06:14 |
| *** sgotliv has joined #openstack-manila | 06:18 | |
| *** yangyapeng has joined #openstack-manila | 06:28 | |
| *** yangyape_ has quit IRC | 06:30 | |
| *** lpetrut has joined #openstack-manila | 06:43 | |
| *** rraja has joined #openstack-manila | 06:51 | |
| openstackgerrit | zhongjun proposed openstack/manila: Create share API without checking share network in DHSS=true mode https://review.openstack.org/256353 | 07:11 |
| *** yangyapeng has quit IRC | 07:18 | |
| *** nkrinner has joined #openstack-manila | 07:28 | |
| *** sgotliv has quit IRC | 07:28 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: LXC/LXD: Fix for share race condition https://review.openstack.org/292360 | 07:30 |
| *** lpetrut has quit IRC | 07:45 | |
| *** a_ta has joined #openstack-manila | 07:57 | |
| *** a_ta has quit IRC | 08:01 | |
| *** chlong has quit IRC | 08:32 | |
| openstackgerrit | Marc Koderer proposed openstack/manila: Fix keystone v3 issues in neutron client https://review.openstack.org/290511 | 08:33 |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Fix for missing neutron_host_id check https://review.openstack.org/291082 | 08:47 |
| *** aovchinnikov has joined #openstack-manila | 08:55 | |
| *** yangyapeng has joined #openstack-manila | 08:55 | |
| *** sgotliv has joined #openstack-manila | 08:57 | |
| *** sgotliv has quit IRC | 08:57 | |
| *** sgotliv has joined #openstack-manila | 08:58 | |
| *** lpetrut has joined #openstack-manila | 08:58 | |
| openstackgerrit | Marc Koderer proposed openstack/manila: Fix keystone v3 issues in neutron client https://review.openstack.org/290511 | 09:16 |
| *** sgotliv has quit IRC | 10:04 | |
| *** ociuhandu has quit IRC | 10:05 | |
| *** tpsilva has joined #openstack-manila | 10:11 | |
| *** yangyapeng has quit IRC | 10:16 | |
| *** sgotliv has joined #openstack-manila | 10:20 | |
| *** merooney has joined #openstack-manila | 10:39 | |
| *** sgotliv has quit IRC | 10:56 | |
| *** rraja has quit IRC | 10:56 | |
| *** ganso has joined #openstack-manila | 10:58 | |
| *** baojg has quit IRC | 11:05 | |
| *** nkrinner_ has joined #openstack-manila | 11:06 | |
| *** nkrinner_ has quit IRC | 11:07 | |
| *** nkrinner_ has joined #openstack-manila | 11:07 | |
| openstackgerrit | zhongjun proposed openstack/manila: Check share-network in 'share create' API https://review.openstack.org/256353 | 11:08 |
| *** rraja has joined #openstack-manila | 11:08 | |
| *** nkrinner has quit IRC | 11:09 | |
| -openstackstatus- NOTICE: Gerrit is going to be restarted | 11:13 | |
| -openstackstatus- NOTICE: Gerrit had to be restarted because was not responsive. As a consequence, some of the test results have been lost, from 08:30 UTC to 10:30 UTC approximately. Please recheck any affected jobs by this problem. | 11:33 | |
| *** alyson_ has joined #openstack-manila | 11:34 | |
| *** mlima has joined #openstack-manila | 11:35 | |
| -openstackstatus- NOTICE: Gerrit had to be restarted because was not responsive. As a consequence, some of the test results have been lost, from 09:30 UTC to 11:30 UTC approximately. Please recheck any affected jobs by this problem. | 11:36 | |
| *** merooney has quit IRC | 11:39 | |
| *** rraja has quit IRC | 11:43 | |
| *** rraja has joined #openstack-manila | 11:44 | |
| *** merooney has joined #openstack-manila | 11:46 | |
| *** erlon has joined #openstack-manila | 11:49 | |
| *** _amrith_ is now known as amrith | 11:54 | |
| *** fthiagogv has joined #openstack-manila | 11:59 | |
| *** dencaval has quit IRC | 12:02 | |
| *** ociuhandu has joined #openstack-manila | 12:05 | |
| *** sgotliv has joined #openstack-manila | 12:13 | |
| *** dencaval has joined #openstack-manila | 12:15 | |
| *** merooney has quit IRC | 12:27 | |
| *** merooney has joined #openstack-manila | 12:32 | |
| openstackgerrit | Marc Koderer proposed openstack/manila: Fix keystone v3 issues in neutron client https://review.openstack.org/290511 | 12:32 |
| *** fthiagogv has left #openstack-manila | 12:40 | |
| *** timcl has joined #openstack-manila | 12:46 | |
| *** amrith is now known as _amrith_ | 12:50 | |
| *** gouthamr has joined #openstack-manila | 12:50 | |
| *** eharney has joined #openstack-manila | 13:01 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: LXC/LXD: Fix for share race condition https://review.openstack.org/292360 | 13:07 |
| *** akshai has joined #openstack-manila | 13:14 | |
| *** martyturner has joined #openstack-manila | 13:16 | |
| *** ig0r__ has joined #openstack-manila | 13:19 | |
| *** cknight has joined #openstack-manila | 13:21 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: LXC/LXD: Fix for incorrect user handling bug https://review.openstack.org/292871 | 13:21 |
| *** ig0r_ has quit IRC | 13:22 | |
| *** chlong has joined #openstack-manila | 13:25 | |
| *** merooney has quit IRC | 13:26 | |
| *** jcsp has joined #openstack-manila | 13:29 | |
| cknight | ganso: ping | 13:29 |
| ganso | cknight: pong | 13:29 |
| *** dustins has joined #openstack-manila | 13:30 | |
| cknight | ganso: hnas-ci is failing on the update_access patches. Are you ready for those to merge? https://review.openstack.org/#/c/291279/ | 13:30 |
| ganso | cknight: no, my team and I are working to fix CI, we should get it running very soon | 13:31 |
| *** akerr has joined #openstack-manila | 13:31 | |
| cknight | ganso: OK, I'll hold off, then. Thanks. | 13:31 |
| ganso | cknight: thank you | 13:31 |
| *** aovchinnikov has quit IRC | 13:32 | |
| *** merooney has joined #openstack-manila | 13:34 | |
| *** jcsp has quit IRC | 13:41 | |
| *** jcsp has joined #openstack-manila | 13:41 | |
| *** akshai has quit IRC | 13:42 | |
| *** rraja has quit IRC | 13:43 | |
| *** akshai has joined #openstack-manila | 13:44 | |
| *** akshai has quit IRC | 13:45 | |
| *** akshai has joined #openstack-manila | 13:45 | |
| *** a_ta has joined #openstack-manila | 13:53 | |
| *** martyturner has quit IRC | 13:58 | |
| bswartz | akerr: ping | 14:01 |
| bswartz | akerr: will "run netapp-ci" cause the CI to run even if jenkins voted -1? | 14:02 |
| *** Guest41345 has quit IRC | 14:03 | |
| *** martyturner has joined #openstack-manila | 14:03 | |
| bswartz | vponomaryov: ping re https://bugs.launchpad.net/manila/+bug/1548222 | 14:08 |
| openstack | Launchpad bug 1548222 in Manila "Capacity calculation error in batch creation" [High,New] | 14:08 |
| *** yangyapeng has joined #openstack-manila | 14:08 | |
| bswartz | vponomaryov: does generic driver support >1 pool? | 14:08 |
| vponomaryov | bswartz: no | 14:09 |
| bswartz | vponomaryov: the log messages in the bug ^ seem to indicate 2 pools w/ generic driver | 14:10 |
| bswartz | I'm confused | 14:10 |
| vponomaryov | bswartz: there is Huawei driver | 14:11 |
| bswartz | why is hostname "ubuntu@generic1" then? | 14:11 |
| vponomaryov | bswartz: u'vendor_name': u'Huawei' | 14:11 |
| vponomaryov | bswartz: in report to scheduler | 14:11 |
| bswartz | u'share_backend_name': u'GENERIC1' | 14:12 |
| akerr | bswartz: yes | 14:12 |
| vponomaryov | bswartz: it is configurable | 14:12 |
| bswartz | okay I see Huawei name | 14:12 |
| bswartz | akerr: ty | 14:12 |
| vponomaryov | bswartz: I suspect they just didn't redefine default devstack options | 14:12 |
| bswartz | vponomaryov: yes, seems like a poor choice of backend name | 14:13 |
| vponomaryov | bswartz: can you please look here - https://review.openstack.org/#/c/288504/10/doc/source/devref/share_back_ends_feature_support_mapping.rst | 14:13 |
| bswartz | what am I looking for? | 14:14 |
| vponomaryov | bswartz: it is about other thing, unrelated to your question, look at comments | 14:15 |
| bswartz | this is change from dustins | 14:15 |
| vponomaryov | bswartz: yes | 14:15 |
| * dustins starts paying attention | 14:15 | |
| vponomaryov | bswartz: so, would be great to see your opinion about raised concern | 14:16 |
| bswartz | regarding the comment change with you and gouthamr: I think it misses the point | 14:17 |
| dustins | FWIW, I mentioned in the comments that I could add a footnote for settings that can only be True | 14:17 |
| bswartz | the value returned should not matter | 14:17 |
| * gouthamr hears his name | 14:17 | |
| bswartz | the question is whether it returns something accurate or not | 14:18 |
| bswartz | zfs does not currently return anything accurate, but it will after vponomaryov fixes bug | 14:18 |
| *** merooney has quit IRC | 14:18 | |
| vponomaryov | bswartz: it is not about only zfs | 14:18 |
| *** Yogi1 has joined #openstack-manila | 14:19 | |
| vponomaryov | bswartz: there is also, HNAS and GlusterFS, maybe more | 14:19 |
| gouthamr | vponomaryov: would you prefer two columns, thin_provisioning=True | thin_provisioning=False (and so on) | 14:19 |
| gouthamr | vponomaryov: or a footnote saying that the value is True or False, Always. | 14:20 |
| bswartz | the way the common capability is written, if the driver returns nothing, then thin_provisioning=False is assumed | 14:20 |
| bswartz | gouthamr: no I don't like that | 14:20 |
| vponomaryov | gouthamr: we are restricted to amount of columns | 14:20 |
| bswartz | a single column should be used, and all driver should support it, unless they have inaccurate information | 14:20 |
| bswartz | therefore any driver that doesn't do thin provisioning automatically support the extra spec | 14:21 |
| bswartz | and any driver that returns something valid also does | 14:21 |
| bswartz | the only ones that are wrong are ones that do thin provisioning but to not report it | 14:21 |
| bswartz | for example ZFS | 14:21 |
| bswartz | that's why a bugfix is needed | 14:22 |
| gouthamr | bswartz: that was kind of my point... if these aren't configurable, they're already the default values. | 14:22 |
| gouthamr | i think vponomaryov was asking if we should somehow say that thick_provisioning is not possible. | 14:22 |
| vponomaryov | gouthamr: yes, table should say explicitly either driver supports " thick_provisioning" and/or " thin_provisioning" | 14:24 |
| vponomaryov | or some of these | 14:24 |
| *** akshai has quit IRC | 14:25 | |
| gouthamr | vponomaryov: i think the footnote makes sense, no? | 14:25 |
| bswartz | I don't know if it's smart to actually include which values the driver returns | 14:26 |
| bswartz | because it may depend on the underlying hardware platform or the list of supported values may change over time | 14:26 |
| *** akshai has joined #openstack-manila | 14:26 | |
| bswartz | the manila driver's job is to communicate whichever value is correct | 14:27 |
| bswartz | if the driver is communicating the correct value, then the capability is "supported" | 14:27 |
| vponomaryov | bswartz: the point is in reporting "possible" values | 14:28 |
| vponomaryov | bswartz: in case of ZFSonLinux driver, it is impossible to have "thick" or full reservation | 14:29 |
| vponomaryov | bswartz: so, we can and should say explicitly, that only "thin" type is available | 14:29 |
| dustins | vponomaryov: I could explicitly call that out in a footnote in that table | 14:29 |
| *** merooney has joined #openstack-manila | 14:32 | |
| *** yangyapeng has quit IRC | 14:32 | |
| vponomaryov | dustins: why not in table? | 14:33 |
| vponomaryov | dustins: False(M), True(L) | 14:33 |
| *** xyang1 has joined #openstack-manila | 14:33 | |
| vponomaryov | dustins: FalseTrue (M) | 14:33 |
| *** yangyapeng has joined #openstack-manila | 14:34 | |
| vponomaryov | dustins: False\/True (M) | 14:34 |
| dustins | That could get messy, kinda like what happened with the DHSS True/False stuff in the share create/delete in the other table | 14:34 |
| vponomaryov | dustins: ok, do as you propose | 14:35 |
| vponomaryov | dustins: will see | 14:35 |
| dustins | Will do, there'll be a patch up shortly! | 14:35 |
| *** sgotliv has quit IRC | 14:35 | |
| dustins | And it's just for the ZFSonLinux driver for now, right? or is there another that falls under the same category of only supporting Thin Provisioning | 14:36 |
| gouthamr | dustins: hnas | 14:36 |
| dustins | Okay, I'll make note of that as well | 14:36 |
| dustins | thanks, gouthamr! | 14:36 |
| ganso | gouthamr, dustins: hnas is already reporting True | 14:39 |
| ganso | gouthamr, dustins: but yes, it is similar | 14:39 |
| dustins | ganso: And it's the case that not only is it reporting True but that True is the only thing it can report? | 14:40 |
| ganso | gouthamr, dustins: we changed it to report always True because it is always Thin at this moment, we were not sure if Thin_provisioning_support = None was meant to mean "not configurable" or False (which would mean Thick), so we changed it to always report True | 14:40 |
| dustins | ganso: Makes sense to me | 14:41 |
| dustins | I'll be sure to reflect that in the next patch set | 14:41 |
| ganso | gouthamr, dustins: it can report Thick, but would require more changes to driver, so "False" value is currently unsupported right now | 14:41 |
| dustins | thanks, ganso! | 14:41 |
| *** mtanino has joined #openstack-manila | 14:41 | |
| *** baojg has joined #openstack-manila | 14:47 | |
| dustins | ganso: Which release did thin provisioning start being true for hnas? | 14:52 |
| *** rraja has joined #openstack-manila | 14:53 | |
| *** martyturner has quit IRC | 14:56 | |
| -openstackstatus- NOTICE: Launchpad OpenID SSO is currently experiencing issues preventing login. The Launchpad team is working on the issue | 14:57 | |
| *** ChanServ changes topic to "Launchpad OpenID SSO is currently experiencing issues preventing login. The Launchpad team is working on the issue" | 14:57 | |
| openstackgerrit | Marc Koderer proposed openstack/manila: Fix keystone v3 issues in neutron client https://review.openstack.org/290511 | 14:57 |
| *** merooney has quit IRC | 14:58 | |
| *** martyturner has joined #openstack-manila | 15:01 | |
| *** merooney has joined #openstack-manila | 15:03 | |
| gouthamr | ganso: thanks! | 15:03 |
| mkoderer___ | vponomaryov: should I activate keystone v3 for patch https://review.openstack.org/290511 directly? | 15:03 |
| mkoderer___ | means I add IDENTITY_API_VERSION=3 in pre_test_hook.sh.. | 15:04 |
| vponomaryov | mkoderer___: I think yes | 15:09 |
| vponomaryov | mkoderer___: but doing so you need to fix all three clients | 15:10 |
| mkoderer___ | vponomaryov: you're right.. | 15:10 |
| mkoderer___ | vponomaryov: so I will add patches up for cinder and nova and then wee swtich to v3 | 15:11 |
| vponomaryov | mkoderer___: you want make them separate? | 15:11 |
| mkoderer___ | vponomaryov: I thought so (since I dislike huge patches) | 15:11 |
| *** merooney has quit IRC | 15:11 | |
| vponomaryov | mkoderer___: they are not usable if use separately | 15:12 |
| mkoderer___ | vponomaryov: they are.. e.g. netapp driver only usues neutron client | 15:12 |
| vponomaryov | mkoderer___: cinder and nova? | 15:13 |
| vponomaryov | mkoderer___: they both should be fixed at once | 15:13 |
| mkoderer___ | vponomaryov: yeah maybe I do everything with one | 15:13 |
| vponomaryov | mkoderer___: right now, it looks like you fix only single particular use case | 15:13 |
| mkoderer___ | vponomaryov: that's right.. I will add nova and cinder in the commit then | 15:14 |
| mkoderer___ | vponomaryov: makes sense to switch to v3 with one commit and not with 2 or 3 | 15:14 |
| *** _amrith_ is now known as amrith | 15:19 | |
| *** gouthamr has quit IRC | 15:25 | |
| *** baojg has quit IRC | 15:28 | |
| *** gouthamr has joined #openstack-manila | 15:29 | |
| *** ChanServ changes topic to "OpenStack Manila Project" | 15:30 | |
| -openstackstatus- NOTICE: Launchpad SSO is back to normal - happy hacking | 15:30 | |
| *** xyang1 has quit IRC | 15:31 | |
| *** aovchinnikov has joined #openstack-manila | 15:33 | |
| *** chlong has quit IRC | 15:33 | |
| *** a_ta has quit IRC | 15:36 | |
| *** a_ta has joined #openstack-manila | 15:37 | |
| tpsilva | dustins: mitaka | 15:38 |
| dustins | tpsilva: Got it, thanks! | 15:38 |
| tpsilva | dustins: actually it was always thin provisioning, but it started reporting in mitaka | 15:38 |
| tpsilva | dustins: so I don't know which value you should put there | 15:38 |
| tpsilva | ganso, gouthamr: do you know? ^ | 15:39 |
| tpsilva | my guess would be mitaka | 15:39 |
| dustins | tpsilva: I'm thinking I'll do when it started reporting, that's what I'm doing with the NetApp backend as well on cknight's recommendation | 15:39 |
| gouthamr | tpsilva: my vote would be mitaka... | 15:39 |
| gouthamr | tpsilva: for the reason dustins just pointed out | 15:40 |
| ganso | dustins, gouthamr: best to leave it as not-supported in Liberty, and Thin in Mitaka | 15:40 |
| tpsilva | dustins, gouthamr: yeah right... just remembered that recommendation from cknight | 15:40 |
| dustins | Once I get out of my meeting, I'll upload a new patch set with that in there, thanks tpsilva and gouthamr! | 15:41 |
| gouthamr | thank you dustins | 15:41 |
| *** a_ta has quit IRC | 15:41 | |
| tpsilva | dustins: thank you! | 15:42 |
| dustins | No problem! | 15:42 |
| *** baojg has joined #openstack-manila | 15:43 | |
| *** a_ta has joined #openstack-manila | 15:43 | |
| *** baojg has quit IRC | 15:48 | |
| *** rraja has quit IRC | 15:48 | |
| *** amrith is now known as _amrith_ | 15:49 | |
| *** absubram has joined #openstack-manila | 15:50 | |
| *** nkrinner_ has quit IRC | 15:57 | |
| *** akshai has quit IRC | 16:00 | |
| *** akshai has joined #openstack-manila | 16:01 | |
| *** _amrith_ is now known as amrith | 16:04 | |
| *** martyturner has quit IRC | 16:05 | |
| *** martyturner has joined #openstack-manila | 16:06 | |
| *** jasonsb has quit IRC | 16:09 | |
| krotscheck | Any cores around to add a +A to a 1-line mitaka bugfix? https://review.openstack.org/#/c/291912/ | 16:10 |
| krotscheck | We already have two +2's. | 16:10 |
| *** jasonsb has joined #openstack-manila | 16:10 | |
| *** jasonsb has quit IRC | 16:10 | |
| cknight | krotscheck: I was hoping to see my CI vote for that one. I'll kick it again. | 16:11 |
| krotscheck | cknight: boot it with great justice. | 16:12 |
| *** timcl has quit IRC | 16:14 | |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/manila: Fix report of ZFSonLinux driver capabilities https://review.openstack.org/293025 | 16:15 |
| *** amrith is now known as _amrith_ | 16:19 | |
| openstackgerrit | Alex O'Rourke proposed openstack/manila: 3PAR remove file tree on delete when using nested shares https://review.openstack.org/290209 | 16:21 |
| *** sgotliv has joined #openstack-manila | 16:26 | |
| *** merooney has joined #openstack-manila | 16:32 | |
| *** vbellur has quit IRC | 16:35 | |
| *** baojg has joined #openstack-manila | 16:44 | |
| *** aovchinnikov has quit IRC | 16:45 | |
| *** baojg has quit IRC | 16:50 | |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/manila: Fix report of ZFSonLinux driver capabilities https://review.openstack.org/293025 | 16:51 |
| *** merooney has quit IRC | 16:52 | |
| *** akshai has quit IRC | 16:55 | |
| *** merooney has joined #openstack-manila | 16:56 | |
| *** timcl has joined #openstack-manila | 16:56 | |
| vponomaryov | gouthamr: https://review.openstack.org/#/c/292659/ is bugfix? | 16:57 |
| *** akshai has joined #openstack-manila | 16:57 | |
| gouthamr | vponomaryov: i believe it is.. the snapshot semantics seemed different at the first go while fixing that bug. | 16:58 |
| *** akshai has quit IRC | 16:58 | |
| *** martyturner has quit IRC | 16:58 | |
| *** vbellur has joined #openstack-manila | 16:58 | |
| gouthamr | vponomaryov: then we decided to use snapshot instances, that were there, but unused | 16:59 |
| vponomaryov | gouthamr: it implements brand new interfaces | 16:59 |
| gouthamr | vponomaryov: i can't shove new data into the old interfaces :( | 16:59 |
| *** akshai has joined #openstack-manila | 17:00 | |
| vponomaryov | gouthamr: I would say that now we support "share replication", but not "snapshot replication", and implementation of the latter is a feature | 17:00 |
| vponomaryov | gouthamr: also, proper support requires snapshot-instances API | 17:01 |
| vponomaryov | bswartz: ^ | 17:02 |
| gouthamr | vponomaryov: i did think hard about that.. but what of users who have replicas and snapshots when they migrate from mitaka to newton.. | 17:02 |
| gouthamr | bswartz ^ | 17:02 |
| gouthamr | :P | 17:02 |
| gouthamr | vponomaryov: working on the APIs for Newton, with zhongjun | 17:03 |
| bswartz | hi | 17:03 |
| bswartz | what are the arrow pointing at? | 17:03 |
| gouthamr | bswartz: #link https://review.openstack.org/#/c/292659/ | 17:03 |
| openstackgerrit | Thomas Bechtold proposed openstack/manila: Use sslutils from oslo_service https://review.openstack.org/289382 | 17:03 |
| *** martyturner has joined #openstack-manila | 17:04 | |
| gouthamr | bswartz: vponomaryov suggests that we say DR as of mitaka does not support replication of snapshots | 17:04 |
| bswartz | replicated snapshots have to be supported | 17:04 |
| bswartz | to back out that functionality would be a significant reduction in the usefulness of replication | 17:05 |
| vponomaryov | gouthamr: I say it is feature and we do implement it than is should follow FFE approach | 17:05 |
| vponomaryov | s/we do/if we do/ | 17:05 |
| vponomaryov | s/than/then/ | 17:05 |
| gouthamr | vponomaryov: i'm okay with requesting an FFE. bswartz, can i? | 17:05 |
| bswartz | the original implemenetation of snapshots with replication which was merged before the feature freeze did actually work | 17:05 |
| bswartz | at the time I did not catch that snapshot instances were not created for each replica | 17:06 |
| bswartz | so I asked gouthamr to fix that | 17:06 |
| bswartz | If we added new APIs as part of this change I would agree it's a new feature | 17:07 |
| bswartz | but no UIs are changed -- the change is all internal, and the fix is how the original code should have been (IMO) | 17:07 |
| vponomaryov | bswartz: on driver side it is implementation of new interfaces | 17:07 |
| vponomaryov | bswartz: if we say so, then any driver change is bugfix | 17:07 |
| bswartz | there is additional logic to reflect the true state of the snapshots into the manila DB | 17:08 |
| bswartz | but the driver was already responsible for creating the snapshots on all replicas | 17:08 |
| bswartz | I don't disagree that this is a gray area | 17:08 |
| bswartz | we often fall into the gray area between bugs and features | 17:08 |
| bswartz | here is my specific concern | 17:09 |
| bswartz | if we don't change the behavior now, and ship mitaka as-is, then when we change the DB interface in newton there will be a rather ugly upgrade step from Mitaka to Newton | 17:10 |
| bswartz | I want to get the DB representation right before mitaka ships | 17:10 |
| bswartz | even if we're missing APIs that we should have | 17:10 |
| bswartz | yes drivers need to write new code, but it's less messy than dealing with an upgrade step in the future | 17:10 |
| vponomaryov | bswartz: DB schema is not changed | 17:11 |
| bswartz | not the schema -- the rows | 17:11 |
| bswartz | if we wait until newton to make the change, then the upgrade step will need to take one row and convert it to many, and the driver will HAVE to be involved with that step | 17:11 |
| bswartz | we've never had drivers involved with DB upgrades before | 17:12 |
| bswartz | I'd rather avoid it by fixing this now | 17:12 |
| bswartz | even if it appears to be a new feature from the driver writer's perspective | 17:12 |
| bswartz | there are only 2 affected drivers | 17:13 |
| bswartz | again if we wait, the number of affected drivers may also be more | 17:13 |
| *** openstackgerrit has quit IRC | 17:14 | |
| *** openstackgerrit_ is now known as openstackgerrit | 17:14 | |
| *** openstackgerrit has quit IRC | 17:15 | |
| *** openstackgerrit_ has joined #openstack-manila | 17:15 | |
| *** merooney has quit IRC | 17:15 | |
| *** openstackgerrit_ is now known as openstackgerrit | 17:16 | |
| *** openstackgerrit_ has joined #openstack-manila | 17:17 | |
| *** merooney has joined #openstack-manila | 17:23 | |
| *** pcaruana has quit IRC | 17:24 | |
| bswartz | dustins gouthamr vponomaryov: finally I added my opinion to https://review.openstack.org/#/c/288504/ | 17:29 |
| bswartz | after thinking about it over lunch I changed my mind from what i said earlier in the channel | 17:30 |
| vponomaryov | bswartz: so, looks like we are all in agreement | 17:32 |
| bswartz | on dustins's patch? yes I think so | 17:32 |
| bswartz | on the replicated snapshots feature/bug fix I hope you guys agree that the pain of doing it before mitaka is less than the pain of doing it after mitaka | 17:33 |
| bswartz | and I asked gouthamr to minimize the scope of the change so it doesn't affect the API or config options because I don't want to treat it as a feature | 17:34 |
| *** pcaruana has joined #openstack-manila | 17:36 | |
| *** lpetrut has quit IRC | 17:39 | |
| openstackgerrit | Merged openstack/manila: NetApp cDOT driver should not split clones https://review.openstack.org/292528 | 17:48 |
| gouthamr | bswartz, vponomaryov: i will push up a patch for the API changes (meant for newton) soon; working on other bugs right now. | 17:53 |
| bswartz | gouthamr: it's not a priority | 17:54 |
| bswartz | we have so many bugs left for mitaka | 17:54 |
| bswartz | where are those test and the changes to the netapp driver btw? | 17:54 |
| Yogi1 | bswartz I am pushing the tests in few minutes | 17:55 |
| bswartz | Yogi1: ty | 17:55 |
| * bswartz thinks the hail storm last night may have knocked out some of his lab machines | 17:55 | |
| akerr | bswartz: you leave the roof open in your attic? | 17:57 |
| *** merooney has quit IRC | 17:57 | |
| *** martyturner has quit IRC | 17:58 | |
| *** merooney has joined #openstack-manila | 17:59 | |
| vponomaryov | akerr: maybe there is no roof anymore? )) | 18:02 |
| vponomaryov | gouthamr: can you please look at comment in https://review.openstack.org/#/c/292659/1/manila/share/driver.py ? | 18:03 |
| vponomaryov | gouthamr: it influences driver implementation | 18:04 |
| gouthamr | vponomaryov: you're right. its a list of dicts | 18:04 |
| gouthamr | sorry about that. will fix it | 18:04 |
| vponomaryov | gouthamr: ok, then I will code considering it as list of dicts | 18:04 |
| *** martyturner has joined #openstack-manila | 18:05 | |
| gouthamr | vponomaryov: thanks. | 18:05 |
| gouthamr | vponomaryov: same with delete_replicated_snapshot | 18:05 |
| gouthamr | vponomaryov: a list of instances to update, if any deleted instances, set status to "deleted" | 18:06 |
| vponomaryov | gouthamr: instances of snapshots? | 18:06 |
| gouthamr | vponomaryov: list of snapshot instances.. | 18:07 |
| gouthamr | vponomaryov: just like that snapshot_list there. | 18:07 |
| gouthamr | sry, snapshot_instances_list | 18:07 |
| bswartz | akerr: power outages on unprotected equipment | 18:11 |
| akerr | bswartz: smh | 18:11 |
| akerr | bswartz: you run a datacenter without surge protectors? | 18:11 |
| bswartz | my UPSes seem to be misallocated | 18:12 |
| bswartz | moving them around is a pain because it involves downtime | 18:12 |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Ensure Snapshots across replicas https://review.openstack.org/292659 | 18:12 |
| vponomaryov | gouthamr: what statuses are expected for snapshot instances? | 18:14 |
| gouthamr | vponomaryov: same as snapshots: 'creating', 'deleting', 'error', 'error_deleting', 'available' | 18:15 |
| *** lpetrut has joined #openstack-manila | 18:15 | |
| vponomaryov | gouthamr: you sure? | 18:15 |
| gouthamr | vponomaryov: and 'error_deleting'. | 18:15 |
| vponomaryov | gouthamr: replica snapshot cannot have "in-sync"? | 18:15 |
| gouthamr | vponomaryov: xyang added 'managing' and related statuses, i am submitting a related change to disallow unmanaging a replicated snapshot | 18:16 |
| gouthamr | vponomaryov: i like that idea.. but it would be very different from our current implementation.. | 18:17 |
| vponomaryov | gouthamr: current implementation? | 18:17 |
| gouthamr | vponomaryov: snapshots as of today | 18:17 |
| vponomaryov | gouthamr: snapsot instances are not available via API | 18:17 |
| dustins | bswartz: ping about thin_provisioning column | 18:18 |
| gouthamr | vponomaryov: so instead of 'creating' and 'available' are you suggesting 'in_sync' and 'out_of_sync'? | 18:23 |
| gouthamr | vponomaryov: creating---> out_of_sync; available---> in_sync | 18:23 |
| vponomaryov | gouthamr: same approach as for replicas | 18:23 |
| gouthamr | vponomaryov: the difference is i don't have a replica_state field on the snapshots.. | 18:24 |
| vponomaryov | gouthamr: )) | 18:24 |
| vponomaryov | gouthamr: then DB should be updated | 18:25 |
| gouthamr | vponomaryov: then it becomes a feature | 18:25 |
| gouthamr | :P | 18:25 |
| vponomaryov | gouthamr: right )) | 18:25 |
| bswartz | dustins: yes | 18:28 |
| dustins | bswartz: So it sounds like you want me to report whether the driver reports True, False or Nothing, or Both for the thin_provisioning capability | 18:29 |
| bswartz | dustins: nothing == False | 18:31 |
| gouthamr | ^^boolean, yeah | 18:31 |
| dustins | bswartz: Right, just including for completeness | 18:31 |
| bswartz | I wouldn't distinguish though | 18:31 |
| dustins | I wasn't planning on it :) | 18:31 |
| bswartz | the capability doc explains that the default is False | 18:31 |
| bswartz | okay | 18:31 |
| dustins | So for everything that currently has a "-", replace with "False" or "F" | 18:32 |
| gouthamr | not F dustins | 18:32 |
| gouthamr | F = Folsum | 18:32 |
| dustins | gouthamr: That was before my time :P | 18:33 |
| gouthamr | or before manila | 18:33 |
| gouthamr | :) | 18:33 |
| dustins | :P | 18:33 |
| dustins | The ones that have a release, and are not ZFSonLinux or HDFS, get Both | 18:33 |
| dustins | And those two get True only | 18:33 |
| vponomaryov | dustins: GlusterFS is expected to be that kind of driver too | 18:37 |
| vponomaryov | dustins: better ask csaba or rraja | 18:37 |
| dustins | vponomaryov: Not a bad idea, it wasn't immediately clear from the driver source, so I'll have to ask them | 18:38 |
| dustins | (though I think rraja is offline and csaba is on PTO) | 18:38 |
| *** merooney has quit IRC | 18:38 | |
| bswartz | I'm not sure if listing the release adds any value | 18:40 |
| bswartz | it's something we inherited from cinder and it's always annoyed me | 18:40 |
| bswartz | if people want to know what was supported in past releases, that's what git history and stable branches are for | 18:41 |
| *** dalgaaf has joined #openstack-manila | 18:42 | |
| dustins | But is that viable or expected for an "average" user or installer? | 18:42 |
| bswartz | dustins: these docs are part of the source code.... | 18:43 |
| bswartz | really I would prefer for non-developer-facing information to be out of tree | 18:43 |
| bswartz | this page is the only remaining exception to that rule | 18:43 |
| dustins | Right, but I'd imagine that most users would go to the "rendered" documents on OpenStack's documentation site | 18:44 |
| bswartz | yeah | 18:44 |
| bswartz | openstack-manuals is the right home | 18:44 |
| dustins | bswartz: Yeah, glad I wasn't the only one who that that was odd | 18:44 |
| bswartz | it's a quirk of history | 18:44 |
| bswartz | if someone wanted to fix this by moving the doc entirely I would wholeheartedly support that | 18:45 |
| * bswartz wonders why he can't highlight anything with his mouse... | 18:45 | |
| dustins | Hmm...An idea for when I get bored :D | 18:45 |
| * bswartz silently curses XFCE | 18:45 | |
| dustins | And that's odd... | 18:45 |
| dustins | Gnome3 has gotten pretty good of late | 18:46 |
| * dustins ducks | 18:46 | |
| *** baojg has joined #openstack-manila | 18:46 | |
| *** _amrith_ is now known as amrith | 18:49 | |
| *** baojg has quit IRC | 18:51 | |
| bswartz | grrrr | 18:54 |
| bswartz | what's with the schism between gnome 2 and gnome 3? | 18:54 |
| bswartz | I don't even follow that little war | 18:55 |
| *** martyturner has quit IRC | 18:57 | |
| openstackgerrit | Merged openstack/manila: register the config generator default hook with the right name https://review.openstack.org/291912 | 19:01 |
| *** martyturner has joined #openstack-manila | 19:04 | |
| *** timcl has quit IRC | 19:04 | |
| *** timcl has joined #openstack-manila | 19:06 | |
| *** timcl has quit IRC | 19:08 | |
| dustins | I dunno, but Gnome 2 lives on through Mate! | 19:09 |
| dustins | I ran that for a bit on the workstation | 19:09 |
| *** porrua has quit IRC | 19:10 | |
| *** gouthamr has quit IRC | 19:11 | |
| *** krotscheck is now known as kro_death_by_sin | 19:14 | |
| *** kro_death_by_sin is now known as kro_sinus_hell | 19:14 | |
| *** gouthamr has joined #openstack-manila | 19:16 | |
| *** xyang1 has joined #openstack-manila | 19:19 | |
| openstackgerrit | Merged openstack/manila: Remove default values for update_access() https://review.openstack.org/291153 | 19:26 |
| *** ociuhandu has quit IRC | 19:32 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Ensure Snapshots across replicas https://review.openstack.org/292659 | 19:34 |
| *** lpetrut has quit IRC | 19:43 | |
| aorourke | bswartz, thank you for reevaluating the 3PAR nested shares bug and granting the ffe. I definitely think this is the way to fix the issue | 19:45 |
| *** baojg has joined #openstack-manila | 19:50 | |
| *** ociuhandu has joined #openstack-manila | 19:53 | |
| *** baojg has quit IRC | 19:57 | |
| *** martyturner has quit IRC | 20:01 | |
| *** martyturner has joined #openstack-manila | 20:02 | |
| openstackgerrit | Yogesh proposed openstack/manila: Add tests to ensure snapshots across replicas https://review.openstack.org/293113 | 20:03 |
| *** ig0r_ has joined #openstack-manila | 20:04 | |
| *** ig0r__ has quit IRC | 20:06 | |
| bswartz | aorourke: sorry to make you do extra work | 20:06 |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Add tests to ensure snapshots across replicas https://review.openstack.org/293113 | 20:07 |
| *** esker has joined #openstack-manila | 20:14 | |
| *** akerr has quit IRC | 20:23 | |
| *** bswartz has quit IRC | 20:29 | |
| *** bswartz has joined #openstack-manila | 20:30 | |
| *** cknight has quit IRC | 20:38 | |
| *** esker has quit IRC | 20:38 | |
| *** esker has joined #openstack-manila | 20:38 | |
| *** baojg has joined #openstack-manila | 20:54 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Ensure Snapshots across replicas https://review.openstack.org/292659 | 20:59 |
| *** baojg has quit IRC | 20:59 | |
| *** eharney has quit IRC | 21:09 | |
| *** vbellur has quit IRC | 21:20 | |
| *** esker has quit IRC | 21:20 | |
| *** esker has joined #openstack-manila | 21:20 | |
| *** martyturner has quit IRC | 21:21 | |
| *** esker has quit IRC | 21:26 | |
| *** martyturner has joined #openstack-manila | 21:26 | |
| openstackgerrit | Alex O'Rourke proposed openstack/manila: 3PAR remove file tree on delete when using nested shares https://review.openstack.org/290209 | 21:32 |
| *** amrith has quit IRC | 21:35 | |
| *** dustins has quit IRC | 21:35 | |
| *** amrith has joined #openstack-manila | 21:35 | |
| *** eharney has joined #openstack-manila | 21:36 | |
| *** pcaruana has quit IRC | 21:38 | |
| *** pcaruana has joined #openstack-manila | 21:52 | |
| *** baojg has joined #openstack-manila | 21:55 | |
| *** baojg has quit IRC | 21:59 | |
| *** xyang1 has quit IRC | 22:00 | |
| *** martyturner has quit IRC | 22:04 | |
| *** martyturner has joined #openstack-manila | 22:06 | |
| *** tpsilva has quit IRC | 22:07 | |
| *** david-lyle has quit IRC | 22:13 | |
| *** a_ta has quit IRC | 22:15 | |
| *** absubram has quit IRC | 22:21 | |
| *** sgotliv has quit IRC | 22:23 | |
| *** Yogi1 has quit IRC | 22:26 | |
| *** gouthamr has quit IRC | 22:28 | |
| *** vbellur has joined #openstack-manila | 22:45 | |
| *** martyturner has quit IRC | 22:47 | |
| *** baojg has joined #openstack-manila | 22:56 | |
| *** baojg has quit IRC | 23:00 | |
| *** chlong has joined #openstack-manila | 23:28 | |
| *** jasonsb has joined #openstack-manila | 23:50 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Fix Hitachi HNAS driver version https://review.openstack.org/293175 | 23:55 |
| *** baojg has joined #openstack-manila | 23:56 | |
| *** dalgaaf has quit IRC | 23:58 | |
| *** jasonsb has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!