| *** chlong has quit IRC | 00:05 | |
| *** yangyape_ has quit IRC | 00:10 | |
| *** akshai has joined #openstack-manila | 00:40 | |
| *** akshai_ has joined #openstack-manila | 00:42 | |
| *** akshai has quit IRC | 00:45 | |
| *** dgonzalez has quit IRC | 00:48 | |
| *** dgonzalez has joined #openstack-manila | 00:50 | |
| *** mtanino has quit IRC | 00:52 | |
| *** suyash has quit IRC | 00:57 | |
| *** yangyapeng has joined #openstack-manila | 01:12 | |
| *** yangyapeng has quit IRC | 01:21 | |
| *** wangqun has joined #openstack-manila | 01:28 | |
| *** xyang1 has quit IRC | 01:29 | |
| *** chlong has joined #openstack-manila | 01:30 | |
| *** xiaoxi has joined #openstack-manila | 01:43 | |
| *** houming has joined #openstack-manila | 01:57 | |
| openstackgerrit | Merged openstack/manila: Fix Share status when driver migrates https://review.openstack.org/299332 | 02:18 |
|---|---|---|
| *** gouthamr_ has quit IRC | 02:20 | |
| *** jasonsb has quit IRC | 02:48 | |
| *** jasonsb has joined #openstack-manila | 02:48 | |
| *** absubram has joined #openstack-manila | 02:53 | |
| *** absubram_ has joined #openstack-manila | 02:55 | |
| *** absubram has quit IRC | 02:58 | |
| *** absubram_ is now known as absubram | 02:58 | |
| *** akshai_ has quit IRC | 03:15 | |
| *** jasonsb has quit IRC | 03:16 | |
| *** absubram has quit IRC | 03:26 | |
| *** yangyapeng has joined #openstack-manila | 03:28 | |
| *** amit213 has quit IRC | 03:33 | |
| *** amit213 has joined #openstack-manila | 03:33 | |
| *** jasonsb has joined #openstack-manila | 03:52 | |
| *** esker has quit IRC | 03:55 | |
| *** nidhimittalhada has joined #openstack-manila | 04:41 | |
| *** jseiler_ has joined #openstack-manila | 05:30 | |
| *** jseiler__ has quit IRC | 05:34 | |
| *** jasonsb has quit IRC | 06:08 | |
| *** jasonsb has joined #openstack-manila | 06:08 | |
| *** jasonsb has quit IRC | 06:13 | |
| *** rraja has joined #openstack-manila | 06:13 | |
| *** jasonsb has joined #openstack-manila | 06:16 | |
| openstackgerrit | NidhiMittalHada proposed openstack/python-manilaclient: manila list --all-tenants should display projectID https://review.openstack.org/312017 | 06:22 |
| openstackgerrit | Cedric Zhuang proposed openstack/manila: Add EMC Unity Driver for Manila https://review.openstack.org/312423 | 06:29 |
| openstackgerrit | Cedric Zhuang proposed openstack/manila: Add EMC Unity Driver for Manila https://review.openstack.org/312423 | 06:34 |
| *** nidhimittalhada has quit IRC | 06:35 | |
| *** cfey has quit IRC | 06:48 | |
| *** cfey has joined #openstack-manila | 06:49 | |
| *** cfey has quit IRC | 06:54 | |
| *** cfey has joined #openstack-manila | 06:56 | |
| openstackgerrit | NidhiMittalHada proposed openstack/manila: Delete Snapshot: status wrongly set when busy https://review.openstack.org/311996 | 07:05 |
| *** permalac has joined #openstack-manila | 07:06 | |
| vponomaryov | ameade, smcginnis, eharney: +1 for prefix as descriptive word like "volume", "share", "compute", etc... One letter is not suitable already. "M"anila overlaps with "M"agnum. | 07:13 |
| *** rraja is now known as rraja|Afk | 07:33 | |
| *** lpetrut has joined #openstack-manila | 07:36 | |
| *** openstackstatus has quit IRC | 07:38 | |
| *** openstackstatus has joined #openstack-manila | 07:39 | |
| *** ChanServ sets mode: +v openstackstatus | 07:39 | |
| *** lpetrut has quit IRC | 07:43 | |
| *** lpetrut has joined #openstack-manila | 07:45 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Container driver https://review.openstack.org/308930 | 07:53 |
| *** chlong has quit IRC | 07:54 | |
| *** dmk0202 has joined #openstack-manila | 08:08 | |
| *** lpetrut has quit IRC | 08:26 | |
| *** rraja|Afk is now known as rraja | 08:33 | |
| *** pcaruana has joined #openstack-manila | 08:36 | |
| *** sage has quit IRC | 08:56 | |
| *** sage has joined #openstack-manila | 08:57 | |
| *** aovchinnikov has quit IRC | 09:04 | |
| *** lpetrut has joined #openstack-manila | 09:14 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Container driver https://review.openstack.org/308930 | 09:37 |
| *** lpetrut has quit IRC | 09:42 | |
| *** houming has quit IRC | 09:48 | |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/manila-ui: Fix usages view on project/overview page https://review.openstack.org/313302 | 09:58 |
| *** ekarlso has quit IRC | 10:08 | |
| *** wangqun has quit IRC | 10:13 | |
| openstackgerrit | Cedric Zhuang proposed openstack/manila: Add EMC Unity Driver for Manila https://review.openstack.org/312423 | 10:15 |
| *** yangyape_ has joined #openstack-manila | 10:33 | |
| *** yangyapeng has quit IRC | 10:36 | |
| *** yangyape_ has quit IRC | 10:37 | |
| *** dmk0202 has quit IRC | 10:57 | |
| *** dmk0202 has joined #openstack-manila | 10:57 | |
| *** lpetrut has joined #openstack-manila | 11:01 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Container driver https://review.openstack.org/308930 | 11:08 |
| *** gouthamr has joined #openstack-manila | 11:21 | |
| *** gouthamr_ has joined #openstack-manila | 11:27 | |
| *** gouthamr has quit IRC | 11:31 | |
| *** yangyapeng has joined #openstack-manila | 11:49 | |
| *** jcsp has joined #openstack-manila | 11:58 | |
| openstackgerrit | Rodrigo Barbieri proposed openstack/manila: Fix HNAS error with unconfined filesystems https://review.openstack.org/307407 | 12:02 |
| *** xiaoxi has quit IRC | 12:06 | |
| *** jcsp has quit IRC | 12:13 | |
| *** ekarlso has joined #openstack-manila | 12:19 | |
| *** xyang1 has joined #openstack-manila | 12:38 | |
| bswartz | ameade: every time we use integer ID instead of UUIDs we eventually regret it and switch to UUIDs later | 12:42 |
| ameade | bswartz: the trouble here is we want them to be somewhat recognizable by admins | 12:43 |
| ameade | i think the namespaces may be enough | 12:44 |
| ameade | bswartz: ^ | 12:44 |
| bswartz | ameade: monotonically increasing integers then? each message gets the next integer? | 12:44 |
| ameade | bswartz: yeah atm | 12:45 |
| bswartz | you'll need locking around the increment unless you use DB autoincrement field | 12:45 |
| ameade | it's all back to back in one file, i think i can write a test to ensure they are unique | 12:45 |
| bswartz | I wonder how DB autoincrement fields work with DB clustering | 12:46 |
| ameade | put down on my TODO list to have a test that guarantees this | 12:46 |
| bswartz | wait | 12:47 |
| *** tpsilva has joined #openstack-manila | 12:47 | |
| bswartz | are the IDs of the messages like message type codes, or message instance IDs? | 12:47 |
| ameade | type codes | 12:47 |
| bswartz | OH! | 12:47 |
| ameade | the ids are def uuids | 12:47 |
| bswartz | well then yeah integers make more sense | 12:47 |
| bswartz | sorry I thought we were talking about instance ids | 12:48 |
| ameade | haha, yeah dont panic | 12:48 |
| bswartz | forget everything I said | 12:48 |
| ameade | funny story | 12:48 |
| ameade | i was involved in the initial id->uuid transition | 12:48 |
| ameade | I had a scrum story with 50+ tasks | 12:48 |
| ameade | each task was a sticky note | 12:49 |
| bswartz | btw: http://apigee.com/about/blog/technology/restful-api-design-what-about-errors | 12:49 |
| ameade | that hung at my desk for ages | 12:49 |
| bswartz | ^ I like what Twilio did here | 12:49 |
| ameade | ah yeah, i love it when they have that | 12:49 |
| ameade | angular does that to the next level | 12:49 |
| ameade | it actually gets specific error context passed in the post when you click the more info link | 12:50 |
| ameade | and tells you all about YOUR issue | 12:50 |
| ameade | very cool | 12:50 |
| ameade | thanks for the link | 12:50 |
| *** alejandrito has joined #openstack-manila | 12:52 | |
| vponomaryov | wait, integer can harm security | 12:52 |
| vponomaryov | it shows how many messages exists and what is next | 12:52 |
| *** timcl has joined #openstack-manila | 12:53 | |
| vponomaryov | usage of prefixes? - ok, but with not-integer-unique-values | 12:53 |
| *** akshai has joined #openstack-manila | 12:53 | |
| ameade | it's opensource? | 12:54 |
| bswartz | vponomaryov: the integers will be for the message codes (analogous to HTTP return codes) | 12:54 |
| *** alkhodos has joined #openstack-manila | 13:00 | |
| *** alkhodos has quit IRC | 13:07 | |
| *** mtanino has joined #openstack-manila | 13:17 | |
| *** pcaruana has quit IRC | 13:25 | |
| *** porrua has joined #openstack-manila | 13:27 | |
| *** mtanino has quit IRC | 13:27 | |
| *** akerr has joined #openstack-manila | 13:31 | |
| *** cknight has joined #openstack-manila | 13:33 | |
| openstackgerrit | Alex Meade proposed openstack/manila: WIP User Messages https://review.openstack.org/313549 | 13:33 |
| gouthamr_ | ameade: w00t | 13:34 |
| *** gouthamr_ is now known as gouthamr | 13:34 | |
| ameade | gouthamr: shhhh | 13:34 |
| gouthamr | :D | 13:35 |
| ameade | it works, got to port docs and tests | 13:35 |
| ameade | then do client work | 13:35 |
| ameade | when i get free time | 13:35 |
| gouthamr | ameade: help is only an irc message (or a short walk from your desk) away | 13:36 |
| *** absubram has joined #openstack-manila | 13:46 | |
| *** absubram_ has joined #openstack-manila | 13:47 | |
| *** absubram has quit IRC | 13:50 | |
| *** absubram_ is now known as absubram | 13:50 | |
| *** dustins has joined #openstack-manila | 13:56 | |
| *** akshai has quit IRC | 13:57 | |
| *** akshai has joined #openstack-manila | 13:59 | |
| *** martyturner has joined #openstack-manila | 14:00 | |
| *** rraja has quit IRC | 14:01 | |
| *** yangyapeng has quit IRC | 14:02 | |
| *** alejandrito has quit IRC | 14:05 | |
| *** catintheroof has joined #openstack-manila | 14:05 | |
| *** mtanino has joined #openstack-manila | 14:09 | |
| *** catintheroof has quit IRC | 14:15 | |
| -openstackstatus- NOTICE: Zuul is currently recovering from a large number of changes, it will take a few hours until your job is processed. Please have patience and enjoy a great weekend! | 14:15 | |
| *** gouthamr has quit IRC | 14:25 | |
| bswartz | dustins: I did a rebase of https://review.openstack.org/#/c/188137/ and the test failed -- I can't really tell why | 14:27 |
| *** catintheroof has joined #openstack-manila | 14:27 | |
| dustins | bswartz: I'll have a look! | 14:28 |
| dustins | Looking at the logs, it looks like a Property error | 14:31 |
| dustins | resources.EndpointMap.properties: Unknown Property ManilaApiVirtualIP | 14:31 |
| openstackgerrit | Valeriy Ponomaryov proposed openstack/manila-ui: Fix usages view on project/overview page https://review.openstack.org/313302 | 14:33 |
| openstackgerrit | Merged openstack/manila-ui: Use latest Manila API microversion https://review.openstack.org/310002 | 14:36 |
| *** absubram has quit IRC | 14:41 | |
| -openstackstatus- NOTICE: Zuul has been restarted. As a results, we only preserved patches in the gate queue. Be sure to recheck your patches in gerrit if needed. | 14:46 | |
| *** eharney has joined #openstack-manila | 14:57 | |
| *** akapil has joined #openstack-manila | 14:58 | |
| *** akshai has quit IRC | 15:01 | |
| ameade | vponomaryov: hey do you know how to turn quotas off via tempest for the manila tests? | 15:02 |
| ameade | vponomaryov: I can figure it out but thought you may know off the top of your head | 15:02 |
| vponomaryov | ameade: option "run_quota_tests=False" | 15:02 |
| *** smcginnis_ has joined #openstack-manila | 15:02 | |
| ameade | vponomaryov: I mean make quotas unlimited for the test user | 15:03 |
| vponomaryov | -1 is used for unlimited quota | 15:03 |
| ameade | so i can just set in manila.conf quota_gigabytes=-1 vponomaryov? | 15:05 |
| *** david-lyle has joined #openstack-manila | 15:06 | |
| vponomaryov | ameade: if you want tempest do it automatically and not in single specific test than it cannot do it | 15:06 |
| vponomaryov | s/than/then/ | 15:06 |
| *** ociuhandu has joined #openstack-manila | 15:07 | |
| vponomaryov | ameade: why do you need unlimited quota? | 15:08 |
| ameade | vponomaryov: hmm ok, I thought i remembered seeing somethign that set them all in devstack | 15:08 |
| ameade | vponomaryov: CI is hitting a quota | 15:08 |
| ameade | everyone once and awhile | 15:08 |
| ameade | and i dont want it to be an issue | 15:08 |
| ameade | every* | 15:09 |
| vponomaryov | ameade: we set Cinder quotas in our devstack | 15:09 |
| vponomaryov | https://github.com/openstack/manila/blob/85e930bc/devstack/plugin.sh#L106 | 15:10 |
| ameade | ah i see | 15:10 |
| vponomaryov | ameade: anyway, it is strange that using our tempest default Manila quotas are exceeded | 15:10 |
| vponomaryov | ameade: or you run in billion threads? )) | 15:11 |
| vponomaryov | s/in/it/ | 15:11 |
| ameade | vponomaryov: i'll check our concurrency, it hit ' ShareSizeExceedsAvailableQuota: Requested share exceeds allowed gigabytes quota.' for a CG test | 15:11 |
| vponomaryov | ameade: you can define any option via localrc | 15:11 |
| ameade | vponomaryov: do you know what happens if I set quota_driver to nothing? | 15:12 |
| ameade | i'll try it | 15:12 |
| ameade | thanks a lot, you saved me some time | 15:12 |
| vponomaryov | ameade: just define "quota_gigabytes" and other quotas | 15:13 |
| ameade | vponomaryov: yeah i may just do that, I wonder if there is a catch all solution in case more quotas are added | 15:13 |
| ameade | FTR, everything dies if quota_driver = nothing lol | 15:14 |
| vponomaryov | ameade: you are surprised? | 15:15 |
| ameade | nope, just optimistic | 15:15 |
| akerr | should make a quota_driver that just returns infinite for all quotas | 15:15 |
| akerr | and use that for tests | 15:15 |
| vponomaryov | it is really easier to define just 5 options to -1 | 15:15 |
| akerr | for now, but as ameade stated what about the new quotas | 15:16 |
| vponomaryov | they are added one per century | 15:16 |
| *** martyturner has quit IRC | 15:23 | |
| *** akshai has joined #openstack-manila | 15:25 | |
| tbarron | ameade: just as a point of curiosity, nova has a null quota driver available that would do what you want I think | 15:31 |
| *** dmk0202 has quit IRC | 15:33 | |
| ameade | tbarron: thanks, ill see if i can come up with something dead simple after lunch. | 15:33 |
| ameade | may just cave on the easy way out for now | 15:33 |
| *** dmk0202 has joined #openstack-manila | 15:34 | |
| tbarron | ameade: yeah, that wasn't intended as a practical suggestion for your problem. | 15:34 |
| *** martyturner has joined #openstack-manila | 15:35 | |
| tbarron | ameade: probably can just set the default quotas high or unlimited in manila.conf | 15:35 |
| *** catintheroof has quit IRC | 15:36 | |
| *** absubram has joined #openstack-manila | 15:36 | |
| openstackgerrit | Merged openstack/manila: Fix IPv6 standalone network plugin test https://review.openstack.org/313061 | 15:39 |
| *** dmk0202 has quit IRC | 15:39 | |
| *** akapil has quit IRC | 15:55 | |
| *** johnthetubaguy has quit IRC | 16:08 | |
| *** johnthetubaguy has joined #openstack-manila | 16:08 | |
| *** yangyapeng has joined #openstack-manila | 16:17 | |
| *** yangyapeng has quit IRC | 16:18 | |
| *** catintheroof has joined #openstack-manila | 16:27 | |
| *** dmk0202 has joined #openstack-manila | 16:40 | |
| *** yangyapeng has joined #openstack-manila | 16:42 | |
| *** yangyapeng has quit IRC | 16:47 | |
| *** sgotliv has joined #openstack-manila | 16:49 | |
| *** dmk0202 has quit IRC | 16:55 | |
| *** dmk0202 has joined #openstack-manila | 16:57 | |
| *** dmk0202 has quit IRC | 17:01 | |
| *** jseiler__ has joined #openstack-manila | 17:06 | |
| *** jseiler_ has quit IRC | 17:09 | |
| *** jasonsb has quit IRC | 17:11 | |
| ameade | vponomaryov: tbarron: lol we are running with 16 concurrency | 17:12 |
| *** akapil has joined #openstack-manila | 17:12 | |
| ameade | i'm sure that's why we hit it | 17:12 |
| *** jasonsb has joined #openstack-manila | 17:12 | |
| tbarron | ameade: 16 is nothing these days! | 17:13 |
| bswartz | dustins: pushed new patch | 17:14 |
| dustins | bswartz: Awesome, I'll keep an eye on it | 17:14 |
| tbarron | ameade: I wager we'll hit quota system bugs (in manila, cinder, nova, neutron) with high concurrency b/c of races, etc. around reservations. | 17:14 |
| ameade | tbarron: goutham keeps sneaking into the CI and upping the number | 17:14 |
| tbarron | ameade: one could potentially set expiration on reservations to very low time too. | 17:15 |
| ameade | tbarron: I'm gonna implement the RandomNumberQuotaDriver | 17:15 |
| tbarron | ameade: mostly just a thought experiment though | 17:15 |
| tbarron | ameade: complements bswartz's stochastic scheduler? | 17:16 |
| ameade | tbarron: ah yes, they would pair quite nicely | 17:17 |
| *** jasonsb has quit IRC | 17:17 | |
| *** cFouts is now known as nug | 17:31 | |
| *** nug is now known as Guest47652 | 17:32 | |
| *** Guest47652 is now known as cFouts | 17:41 | |
| *** lpetrut has quit IRC | 17:43 | |
| *** merooney has joined #openstack-manila | 17:55 | |
| *** vbellur has quit IRC | 18:02 | |
| *** vbellur has joined #openstack-manila | 18:03 | |
| *** gouthamr has joined #openstack-manila | 18:08 | |
| bswartz | tbarron: at one time there was a pure "chance" scheduler | 18:08 |
| *** suyash has joined #openstack-manila | 18:08 | |
| tbarron | bswartz: interesting | 18:08 |
| bswartz | honestly I can't see the value of a random quota driver | 18:09 |
| tbarron | bswartz: i think that was just a joke | 18:09 |
| bswartz | random chargeback could lead to interesting results though | 18:09 |
| tbarron | bswartz: i know your proposal is serious though | 18:09 |
| *** sgotliv has quit IRC | 18:10 | |
| bswartz | I know it was a joke -- I was trying trying to figure out if humorous outcomes could be obtained if we implemented it -- and I think not | 18:10 |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/manila: Fix share server info in CGs created from CGs https://review.openstack.org/312963 | 18:11 |
| tbarron | bswartz: I think the outcomes would be sad, not funny | 18:11 |
| bswartz | yes | 18:11 |
| bswartz | but random chargeback would be funny indeed | 18:11 |
| tbarron | maybe some kind of rally death test | 18:11 |
| tbarron | bswartz: not if you are getting the bills! | 18:12 |
| *** gouthamr_ has joined #openstack-manila | 18:13 | |
| bswartz | you would have to implement it like gambling -- everyone pays, and some people get lucky | 18:13 |
| tbarron | bswartz: I see you are making the subtle point that DOS protection is the prime purpose of quotas. | 18:16 |
| tbarron | Even if everyone consented to such a scheme for chargeback, unless the random limits were constrained then DOS attacks would be possible. | 18:17 |
| *** gouthamr has quit IRC | 18:17 | |
| tbarron | everyone could get very unlucky | 18:18 |
| tbarron | (still doesn't change the fact that quotas are often used for chargeback of course ...) | 18:18 |
| bswartz | well this fact irritates me ^ | 18:19 |
| bswartz | it clearly implies that we've failed to implement chargeback in a useful enough way | 18:19 |
| tbarron | bswartz: I know, that's why I brought it up :) | 18:19 |
| tbarron | bswartz: I think the right approach is to support other competing chargeback models. | 18:20 |
| bswartz | I need to dig into what people are doing today with cinder | 18:20 |
| bswartz | if they're also using quotas with cinder then the problem is bigger than I imagine | 18:20 |
| tbarron | bswartz: verizon charges me 15GB data a month even if I use less. Of course the quota is "soft", as they'll let me go over and pay more. | 18:23 |
| tbarron | bswartz: in aggregate these quotas serve as DOS protection (protect capacity vis a vis usage) and lead to more predictable billing and revenues. | 18:24 |
| tbarron | bswartz: I'm not arguing that it's the *best* chargeback system. | 18:24 |
| bswartz | companies that sell to the general public often have highly evolved pricing structures | 18:25 |
| tbarron | bswartz: I'm just pointing to a real use case. | 18:25 |
| bswartz | IT departments are comparably in the stone age | 18:25 |
| * gouthamr_ mic drop | 18:25 | |
| tbarron | the customer that has me looking at quotas is a telco who wants to run a public cloud | 18:26 |
| *** gouthamr_ is now known as gouthamr | 18:26 | |
| tbarron | I'll let you figure what epoch in history they fit into. | 18:26 |
| bswartz | well as the authors of manila it's not our job to tell deployers how to run their business -- it's our job to provide tools so they can run their business how they want | 18:26 |
| tbarron | +1 | 18:26 |
| bswartz | I guess I'm saying we need to support stone age pricing models AND highly evolved pricing models | 18:27 |
| tbarron | so I think we need both basis for quota based and usage based chargeback | 18:27 |
| tbarron | and usage-based has its own definitional issues of course. | 18:28 |
| tbarron | I think we are in violent agreement. | 18:28 |
| smcginnis | tbarron: I wouldn't consider that "chargeback". Seems like the wrong term to me. | 18:28 |
| smcginnis | At least how I've always used the term chargeback. | 18:29 |
| bswartz | I'm interested in the differences between steady state consumers and bursty consumers though | 18:29 |
| tbarron | s/chargeback/billing/ | 18:29 |
| smcginnis | :) | 18:29 |
| bswartz | some people use a flat amount of storage all day every day | 18:29 |
| tbarron | smcginnis: point taken ^^^ | 18:29 |
| bswartz | others consume large amounts of storage briefly then dispose of it | 18:29 |
| *** akapil has quit IRC | 18:30 | |
| bswartz | quotas fail for the second type of customer | 18:30 |
| bswartz | of course a deployer may charge both customers based on quotas | 18:30 |
| bswartz | but then he's only incentivising the bursty customer to be less bursty and to consume his maximum all the time | 18:31 |
| bswartz | that doesn't seem desirable | 18:31 |
| tbarron | if we provide both usage vs. time and quota info to cloud operator then they can choose their billing model. | 18:34 |
| *** catintheroof has quit IRC | 18:41 | |
| *** ociuhandu has quit IRC | 18:43 | |
| bswartz | tbarron: I see it differently -- we're supposed to provide the usage-based information, including when, what, how much, and for how long -- we're supposed to enforce any quotas that are given to us -- but there's little point in reporting the quotas back, since they're externally defined | 18:51 |
| bswartz | I see quotas as a write-only interface for admins | 18:51 |
| scottda | cknight: I'm looking at code, and I'm not clear why microversions requires that extensions be moved to core. Extensions still inherit from wsgi.Controller, so they should be able to use api_version object... | 18:55 |
| scottda | xyang1: Do you know? ^^^ | 18:56 |
| cknight | scottda: Probably true. But as originally envisioned, extensions weren't versioned so we had already decided to eliminate them in manila. | 18:56 |
| scottda | cknight: OK, thanks. | 18:57 |
| tbarron | bswartz: are you sayinig that 'manila quota-show ...' should go away? | 18:58 |
| bswartz | tbarron: the only valid use case I see for that command is debugging | 18:59 |
| bswartz | or for end users to see their quotas | 18:59 |
| tbarron | bswartz: so you disagree with how some customers might want to use it. I guess I am more laissez faire abot how our customers can/should bill. | 19:00 |
| bswartz | I guess in my mind there has to be an external system that does billing which stores its own state | 19:03 |
| bswartz | manila will never be that thing, and manila can never store all the state that thing would be interested in, unless it's basically stone-age and care about nothing but current quotas | 19:04 |
| bswartz | I guess you're saying stone-age billing systems can and do exist and we shouldn't judge them | 19:05 |
| tbarron | maybe that's the disconnect. I am thinking of billing system as external to OpenStack but OpenStack supplies inputs to such systems. If such a system wants quota at a time, fine. | 19:05 |
| tbarron | yeah, we can and should judge them as consumers, as cloud builders, maybe as citizens or political economists. | 19:06 |
| xyang1 | scottda: got an answer? | 19:06 |
| tbarron | but as OpenStack infra builders, I think not. | 19:06 |
| scottda | xyang1: Yes, I think cknight has answered ^^^, unless you've another opinion? | 19:07 |
| bswartz | tbarron: openstack should provide the inputs to the system, in the form of events or notifications -- manila at least shouldn't need to store historical information about usage | 19:07 |
| xyang1 | scottda: cknight knows more about it | 19:07 |
| scottda | xyang1: I'm thinking that we should be moving those extensions to core in Cinder that are fully supported, but not those that are not. That seems to be the original intent of extensions. | 19:07 |
| tbarron | bswartz: +1000 | 19:08 |
| xyang1 | scottda: sure | 19:08 |
| scottda | xyang1: thanks. I'll be working on that first one and ping you when ready for review. | 19:08 |
| xyang1 | scottda: ok | 19:08 |
| *** ChanServ changes topic to "OpenStack Shared File Systems | Manila | Also cinder spillover channel" | 19:10 | |
| *** ociuhandu has joined #openstack-manila | 19:12 | |
| *** ChanServ changes topic to "OpenStack Shared File Systems | Manila" | 19:22 | |
| tbarron | speaking of cinder issues .... | 19:23 |
| bswartz | nobody thought my joke was funny.... | 19:23 |
| tbarron | bswartz: I did! | 19:23 |
| bswartz | you cinder guys are more than welcome here btw | 19:23 |
| *** martyturner has quit IRC | 19:24 | |
| bswartz | does anyone use ceilometer for billing? | 19:26 |
| bswartz | is it even a suitable tool for that? | 19:26 |
| smcginnis | I haven't heard of anyone using it, but that seems to fall within its capabilities. | 19:27 |
| bswartz | smcginnis: are you aware of any other tool being used instead? | 19:28 |
| bswartz | I'd like to know what people are actually doing in that space | 19:28 |
| bswartz | I know we added notifications to cinder ostensibly for that purpose | 19:28 |
| smcginnis | I haven't heard of anything, but I wouldn't be surprised if that is being used. | 19:29 |
| smcginnis | I also thought we added the notifications sort of for that readon. | 19:29 |
| smcginnis | *reason | 19:29 |
| bswartz | yeah I guess my question is, did that work out or not? | 19:29 |
| smcginnis | No idea really. Would be nice to hear is someone is using it. | 19:30 |
| *** yangyapeng has joined #openstack-manila | 19:31 | |
| ganso | xyang1: ping | 19:31 |
| xyang1 | ganso: hi | 19:32 |
| ganso | xyang1: hi Xing, I have a quick question about consistency groups in Cinder | 19:32 |
| xyang1 | ganso: sure | 19:32 |
| ganso | xyang1: does Cinder core code have any hierarchy for consistency groups snapshots? like, does it allow snapshot 1 to be deleted if there is a snapshot2? | 19:34 |
| ganso | xyang1: I know it allows for regular snapshots, but I am wondering if this is different for CGs | 19:34 |
| xyang1 | ganso: hierarchy? take a snapshot of another snapshot? | 19:34 |
| *** absubram has quit IRC | 19:35 | |
| ganso | xyang1: no, like a sequence of snapshots | 19:35 |
| *** yangyapeng has quit IRC | 19:35 | |
| ganso | xyang1: take 2 CG snapshots, one after the other, and try to delete the first one without deleting the 2nd one | 19:35 |
| xyang1 | ganso: take snapshot at time0 and another at time1? | 19:35 |
| ganso | xyang1: yes | 19:36 |
| xyang1 | ganso: they are not dependent on each other | 19:36 |
| xyang1 | ganso: so you can delete one or the other | 19:36 |
| ganso | xyang1: oh ok, I thought that maybe could be different for CGs. Thanks! =) | 19:37 |
| xyang1 | ganso: no problem | 19:37 |
| bswartz | ganso: there are backends that have the limitation you describe | 19:38 |
| bswartz | but generally we don't force end users to cope with that | 19:38 |
| ganso | bswartz: hummm, do you think it is ok to fail the snapshot deleting operation and leave the snapshot in ERROR_DELETING? | 19:39 |
| bswartz | backends which cannot delete intermediate snapshots have the option of soft deleting them | 19:39 |
| bswartz | you can always NOT delete them and succeed anyway | 19:39 |
| bswartz | as long as you mark them for eventual deletion | 19:40 |
| *** timcl has quit IRC | 19:40 | |
| ganso | bswartz: yes, it seems the list of cgsnapshots that are always supplied are useful for that | 19:40 |
| ganso | s/are useful/is useful | 19:41 |
| *** akerr has quit IRC | 19:43 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Container driver https://review.openstack.org/308930 | 19:55 |
| *** sage has quit IRC | 19:58 | |
| *** sage has joined #openstack-manila | 20:00 | |
| openstackgerrit | Tom Barron proposed openstack/manila: Remove deprecated manila RequestBodySizeLimiter https://review.openstack.org/313683 | 20:03 |
| *** dmk0202 has joined #openstack-manila | 20:05 | |
| bswartz | dustins: oddly it passed one but failed 3 others: https://review.openstack.org/#/c/188137/ | 20:13 |
| dustins | Time to have another look | 20:16 |
| dustins | I saw in zuul that it failed the containers one | 20:16 |
| *** aovchinnikov has joined #openstack-manila | 20:31 | |
| *** openstackgerrit has quit IRC | 20:32 | |
| *** openstackgerrit has joined #openstack-manila | 20:33 | |
| openstackgerrit | Alexey Ovchinnikov proposed openstack/manila: Container driver https://review.openstack.org/308930 | 20:37 |
| *** aovchinnikov has quit IRC | 20:38 | |
| *** cknight has quit IRC | 20:45 | |
| *** porrua has quit IRC | 20:47 | |
| *** dmk0202 has quit IRC | 20:58 | |
| *** xyang1 has quit IRC | 21:16 | |
| *** dustins has quit IRC | 21:21 | |
| *** eharney has quit IRC | 21:24 | |
| openstackgerrit | Tom Barron proposed openstack/manila: Remove deprecated manila RequestBodySizeLimiter https://review.openstack.org/313683 | 21:25 |
| *** merooney has quit IRC | 21:33 | |
| *** cknight has joined #openstack-manila | 21:43 | |
| *** david-lyle has quit IRC | 22:35 | |
| *** david-lyle has joined #openstack-manila | 22:37 | |
| *** adrianofr has quit IRC | 22:52 | |
| *** tpsilva has quit IRC | 22:57 | |
| *** martyturner has joined #openstack-manila | 23:03 | |
| *** yangyapeng has joined #openstack-manila | 23:07 | |
| *** yangyapeng has quit IRC | 23:12 | |
| *** martyturner has quit IRC | 23:21 | |
| *** akshai has quit IRC | 23:35 | |
| *** absubram has joined #openstack-manila | 23:48 | |
| *** absubram_ has joined #openstack-manila | 23:49 | |
| *** merooney has joined #openstack-manila | 23:51 | |
| *** absubram has quit IRC | 23:53 | |
| *** absubram_ is now known as absubram | 23:53 | |
| *** akshai has joined #openstack-manila | 23:57 | |
| *** akshai_ has joined #openstack-manila | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!