Tuesday, 2017-10-03

*** dsariel has joined #openstack-manila00:01
*** sapcc-bot has quit IRC00:05
*** sapcc-bot has joined #openstack-manila00:06
*** sticker has joined #openstack-manila00:16
*** markstur has joined #openstack-manila00:32
*** wes_dillingham has quit IRC00:40
*** wes_dillingham has joined #openstack-manila00:41
*** sapcc-bot3 has joined #openstack-manila01:04
*** sapcc-bot has quit IRC01:05
*** raissa_ has quit IRC01:05
*** raissa_ has joined #openstack-manila01:05
*** FL1SK has quit IRC02:57
*** erlon has quit IRC03:02
*** gouthamr has joined #openstack-manila04:01
*** arnewiebalck_ has joined #openstack-manila04:05
*** FL1SK has joined #openstack-manila04:09
*** gregsfortytwo is now known as gregsfortytwo104:25
*** gouthamr has quit IRC04:50
*** ganso has quit IRC04:52
*** pcaruana has joined #openstack-manila05:01
*** markstur has quit IRC05:13
*** lpetrut has joined #openstack-manila05:16
*** arnewiebalck_ has quit IRC05:16
*** pcaruana has quit IRC05:23
*** jprovazn has joined #openstack-manila06:08
*** lpetrut has quit IRC06:09
*** hoonetorg has quit IRC06:27
*** ociuhandu has joined #openstack-manila06:38
*** hoonetorg has joined #openstack-manila06:41
*** pcaruana has joined #openstack-manila06:44
*** lpetrut has joined #openstack-manila06:56
*** lpetrut has quit IRC07:05
openstackgerritChris Yang proposed openstack/manila master: Add support for enhanced features to the QNAP Manila driver  https://review.openstack.org/50909707:07
*** sticker has quit IRC07:49
*** ociuhandu has quit IRC08:14
*** boris_42_ has quit IRC08:15
*** wes_dillingham has quit IRC08:27
*** wes_dillingham has joined #openstack-manila08:33
*** ociuhandu has joined #openstack-manila09:16
*** a-pugachev has joined #openstack-manila09:19
*** belmoreira has joined #openstack-manila09:39
*** vponomaryov has quit IRC09:45
*** vponomaryov has joined #openstack-manila09:45
*** lpetrut has joined #openstack-manila09:46
*** vivsoni_ has quit IRC09:54
*** lpetrut_ has joined #openstack-manila10:04
*** lpetrut has quit IRC10:06
*** vivsoni has joined #openstack-manila10:08
*** belmoreira has quit IRC10:16
*** cyberde has joined #openstack-manila10:38
*** drico_ has joined #openstack-manila10:45
drico_Hi10:45
drico_Could somebody tell me where I can change the absolute limits like maxTotalShareGigabytes ?10:45
*** wes_dillingham has quit IRC10:46
openstackgerritTom Barron proposed openstack/manila master: WIP: Use constants for microversion values  https://review.openstack.org/50915211:33
*** ganso has joined #openstack-manila11:54
drico_could it be cinder that put that limit on manila ?12:03
*** dustins has joined #openstack-manila12:07
*** jmlowe has quit IRC12:10
*** rraja has joined #openstack-manila12:11
drico_ok so I got my answer, cinder is limiting manila thru the service project and apply his quota on it12:13
drico_shame the documentation says :  An administrator can configure limits in the manila.conf file there https://docs.openstack.org/manila/pike/admin/shared-file-systems-quotas.html12:14
*** jmlowe has joined #openstack-manila12:14
*** tpsilva has joined #openstack-manila12:25
*** erlon has joined #openstack-manila12:32
*** catintheroof has joined #openstack-manila12:36
*** catintheroof has quit IRC12:36
*** catintheroof has joined #openstack-manila12:36
*** catintheroof has quit IRC12:41
*** catintheroof has joined #openstack-manila12:54
tbarrondrico_: that doc is accurate for manila resources but if you use the manila "generic" driver it uses cinder on the back end so cinder's limits apply to those resources13:01
tbarrondrico_: make sense?  ^^13:01
drico_yes now that I figured out it make sense :)13:02
*** chlong has joined #openstack-manila13:10
*** chlong_ has joined #openstack-manila13:10
*** dsariel has quit IRC13:15
*** eharney has joined #openstack-manila13:18
*** jmlowe has quit IRC13:27
*** dsariel has joined #openstack-manila13:29
*** lpetrut_ has quit IRC13:30
*** jmlowe has joined #openstack-manila14:01
gansoping bswartz, tbarron14:10
tbarronganso: hey, you home now?14:11
gansotbarron: actually at work, but here in Brazil xD14:12
tbarronganso: well, that's wheat I meant :)14:12
gansotbarron: yeah, in my home country14:12
tbarronganso: hope your trip was easy14:12
gansotbarron: well it was as good as all flights usually are: not good14:13
tbarronganso: yeah14:13
tbarronganso: anyways, what's up?14:13
*** jmlowe has quit IRC14:13
gansotbarron: I really don't want to wait until Thursday to discuss this, would prefer to gather some more opinions on this to move forward14:14
gansotbarron: I already talked to Goutham and I hope bswartz can chime in as well14:14
gansotbarron: so, I don't know if you ever saw this: https://github.com/openstack/manila/blob/master/manila/db/sqlalchemy/api.py#L350814:14
tbarronganso: i've seen it but I don't understand it14:15
gansotbarron: that comment is a false statement. Read_deleted does not work like that. It either reads only non-deleted, or only deleted14:15
*** jmlowe has joined #openstack-manila14:15
gansotbarron: None is the same as False, so it will never read the deleted entries14:15
gansotbarron: this causes line 3515 to never work in case of deleted entries, thus the line 3519 will never have the intended effect14:16
gansotbarron: one solution to this is to not do soft deletes when handling drivers private storage entries14:16
gansotbarron: in other words, doing hard deletes, which is something we do not do at all in manila14:17
gansotbarron: so what's your stance on this?14:18
tbarronganso: is there another potential fix?14:18
tbarronganso: offhand it seems there should be a way to do what is needed w/o getting rid of soft deletes14:19
gansotbarron: if we do not want to change to hard deletes, the alternative fix is to not use the generic model_query that always includes a 'deleted' field filter14:19
tbarronis there a downside to that alternative fix?14:19
gansotbarron: there is a way. Goutham is in favor of hard deletes. Myself am a bit uneasy of changing this behavior and having drivers private storage being a even more special snowflake14:20
gansotbarron: some code duplication14:20
gansotbarron: but I agree that soft_deletes do not make sense for drivers_private storage unless we really do want to audit everything the driver has done in the past14:21
tbarronwell I'd like to hear from gouthamr why he's in favor of hard deletes for private storage.  It isn't so much that I think we *need* soft deletes there but14:21
gansotbarron: based on u_glide's false assumption, it seems we never intended to really save or audit that because the updates can undelete entries14:22
tbarronganso: i see14:22
tbarronso if we use hard deletes can we just document that in code comments and maybe in devref for private storage?14:23
tbarronIf we can make it "unsurprising" and the end result is cleaner code then I'm OK with hard deletes.14:23
tbarronI don't think there's a use case for soft deletes for private storage, as you say, it's just the inconsistency that would need to be documented14:24
gansotbarron: yes14:24
*** jmlowe has quit IRC14:24
tbarronganso: so I'm ok with you and gouthamr doing hard deletes and documenting it in code and devref appropriately14:25
gansotbarron: ok thanks for your opinion on it :)14:25
*** chlong has quit IRC14:31
*** chlong_ has quit IRC14:31
*** chlong has joined #openstack-manila14:32
*** jmlowe has joined #openstack-manila14:50
*** eharney has quit IRC14:56
*** markstur has joined #openstack-manila15:02
*** vponomaryov has quit IRC15:05
*** vponomaryov has joined #openstack-manila15:05
*** eharney has joined #openstack-manila15:08
*** vponomaryov has quit IRC15:10
*** vponomaryov has joined #openstack-manila15:11
*** jmlowe has quit IRC15:19
*** pcaruana has quit IRC15:20
*** jmlowe has joined #openstack-manila15:23
*** chlong has quit IRC15:36
*** a-pugachev has quit IRC16:09
*** xyang1 has joined #openstack-manila16:19
*** jmlowe has quit IRC16:24
*** gouthamr has joined #openstack-manila17:06
*** ociuhandu has quit IRC17:15
*** rraja has quit IRC17:18
*** arnewiebalck_ has joined #openstack-manila17:40
*** ociuhandu has joined #openstack-manila17:41
*** jmlowe has joined #openstack-manila17:45
*** gregsfortytwo1 is now known as gregsfortytwo17:52
*** a-pugachev has joined #openstack-manila17:58
*** a-pugachev has quit IRC18:01
*** jmlowe has quit IRC18:07
*** jmlowe has joined #openstack-manila18:09
*** jmlowe has quit IRC18:17
*** jmlowe has joined #openstack-manila18:17
*** arnewiebalck_ has quit IRC18:32
*** markstur has quit IRC18:44
*** a-pugachev has joined #openstack-manila19:09
*** arnewiebalck_ has joined #openstack-manila19:12
*** arnewiebalck_ has quit IRC19:26
*** catintheroof has quit IRC19:32
*** catintheroof has joined #openstack-manila19:32
*** jprovazn has quit IRC19:39
*** sapcc-bot3 has quit IRC19:42
*** sapcc-bot has joined #openstack-manila19:42
*** chlong_ has joined #openstack-manila19:43
*** chlong has joined #openstack-manila19:44
*** chlong has quit IRC19:46
*** catintheroof has quit IRC20:10
*** catintheroof has joined #openstack-manila20:11
*** markstur has joined #openstack-manila20:25
*** eharney has quit IRC20:31
*** jmlowe has quit IRC20:34
*** catintheroof has quit IRC20:42
*** catintheroof has joined #openstack-manila20:42
*** jmlowe has joined #openstack-manila21:03
*** chlong_ has quit IRC21:03
*** boris_42_ has joined #openstack-manila21:05
*** dustins has quit IRC21:09
*** a-pugachev has quit IRC21:43
*** xyang1 has quit IRC22:02
*** gouthamr has quit IRC22:21
*** gouthamr has joined #openstack-manila22:50
*** jmlowe has quit IRC22:51
*** tpsilva has quit IRC22:52
*** jmlowe has joined #openstack-manila22:52
*** jmlowe has quit IRC22:53
*** jmlowe has joined #openstack-manila22:58
*** catintheroof has quit IRC23:01
*** wes_dillingham has joined #openstack-manila23:06
*** ganso has quit IRC23:56

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!