| *** dviroel is now known as dviroel|out | 00:20 | |
| opendevreview | yuval proposed openstack/cinder master: lightos - bugfix compression stats should be True https://review.opendev.org/c/openstack/cinder/+/830538 | 00:20 |
|---|---|---|
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for volume os-reset_status https://review.opendev.org/c/openstack/cinder/+/773985 | 04:19 |
| *** bhagyashris_ is now known as bhagyashris | 06:28 | |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 07:26 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for backup os-reset_status https://review.opendev.org/c/openstack/cinder/+/778193 | 07:27 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 08:47 |
| opendevreview | yuval proposed openstack/cinder master: lightos - bugfix compression stats should be True https://review.opendev.org/c/openstack/cinder/+/830538 | 08:52 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 09:15 |
| *** bhagyashris_ is now known as bhagyashris | 09:25 | |
| yuval | dear pewp your reconnections are spamming the chat - consider using a different client | 10:33 |
| tosky | hemna: it looks like pewp is a bot of yours, or at least an instance of a bot developed by you :) maybe it should change its hosting? | 10:35 |
| *** dviroel|out is now known as dviroel | 11:26 | |
| opendevreview | yuval proposed openstack/cinder master: lightos - bugfix compression stats should be True https://review.opendev.org/c/openstack/cinder/+/830538 | 13:03 |
| hemna | tosky, yah my internets has been odd lately | 13:22 |
| hemna | my isp is ipv6 only basically and it's not playing nice at times | 13:23 |
| rosmaita | hemna: i put pewp on the agenda for today's cinder meeting, can you explain what it's for? It pre-dates my involvement in cinder, and afaik, only you and jungleboyj use it intentionally ... maybe if people understood it better, they will cut it some slack with its recent connectivity problems | 13:35 |
| hemna | it's really nothing important | 13:35 |
| rosmaita | my impression is that it's kind of a fun thing? | 13:35 |
| hemna | yah it's just a fun thing that we used | 13:35 |
| jungleboyj | He he. rosmaita wants to know about pewp . | 13:35 |
| hemna | lolz | 13:35 |
| jungleboyj | @h | 13:36 |
| jungleboyj | I can't remember how to use pewp. | 13:36 |
| * hemna looks at pewp logs | 13:36 | |
| rosmaita | rofl, jungleboyj can't remember how to pewp | 13:36 |
| rosmaita | i think it's an autonomic function, if you think about it too much, you can't do it | 13:36 |
| jungleboyj | *rofl* | 13:37 |
| jungleboyj | Since we have a video meeting this week I have a conflict ... So, won't make it. :-( | 13:38 |
| rosmaita | ok | 13:38 |
| hemna | sopel.irc.backends ERROR - Server timeout detected after 121.266824s; closing. | 13:38 |
| jungleboyj | We can't pewp ? | 13:39 |
| hemna | some connectivity issue. I've had to learn ipv6 quickly with my new firewall at home and I'm not sure if I've got it setup correctly just yet. It's quite complicated | 13:39 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 13:40 |
| jungleboyj | Your ISP has switched to IPv6. | 13:40 |
| hemna | I shut it down for now | 13:41 |
| hemna | I'll mess with it later | 13:41 |
| hemna | well, yes and no. they only give out ipv4 BOGON ips, which aren't routable, but give out an ipv6 /64 subnet | 13:41 |
| hemna | ipv4 behind a NAT, which sux0rs | 13:42 |
| jungleboyj | Ugh, I had to pay for a non-nated IP address to avoid that mess. | 13:45 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 13:51 |
| TusharTgite | rosmaita: updated the patch https://review.opendev.org/c/openstack/cinder/+/773985 | 13:52 |
| rosmaita | TusharTgite: ack | 13:53 |
| rosmaita | reminder: today's cinder meeting is on video + #openstack-meeting-alt at 1400 UTC | 13:59 |
| rosmaita | connection info: https://bluejeans.com/3228528973 | 13:59 |
| rosmaita | jungleboyj rosmaita smcginnis tosky whoami-rajat m5z e0ne geguileo eharney walshh_ jbernard sfernand enriquetaso hemna fabiooliveira yuval ^^ | 13:59 |
| *** whoami-rajat__ is now known as whoami-rajat | 13:59 | |
| whoami-rajat | thanks rosmaita | 14:00 |
| geguileo | rosmaita: thanks | 14:00 |
| hemna | jungleboyj I tried to get the isp to sell me a static ipv4, but they refused. I just think ipv4 is pretty much out of addresses for new ISPs now. | 14:01 |
| hemna | fibre to the home 1gbps up and down...so I had to try and make it work | 14:01 |
| jungleboyj | hemna: Wow, that is crazy. Yeah, that is what I have as well. But have to pay an extra $5 a month for a static IP or something. | 14:02 |
| hemna | hah $5 lol. yah that's a no brainer | 14:02 |
| jungleboyj | Exactly, to avoid all the issues that can come with having a NATed IP. | 14:02 |
| * enriquetaso bj freeze all the time | 14:04 | |
| jungleboyj | *sad trombone.wav* | 14:05 |
| whoami-rajat | geguileo, you're not in the meeting channel so https://review.opendev.org/c/openstack/cinder/+/822040 | 14:10 |
| geguileo | whoami-rajat: oh, my bad | 14:12 |
| hemna | https://github.com/hemna/sopel-tableflip | 14:31 |
| hemna | The help for it. https://github.com/hemna/sopel-tableflip/blob/master/sopel_modules/tableflip/tableflip.py#L84-L100 | 14:31 |
| coreycb | Hi enriquetaso, by any chance would you be able to revisit https://review.opendev.org/c/openstack/cinder/+/815560 ? | 14:37 |
| enriquetaso | sure coreycb, I'll do it after the meetings | 14:43 |
| coreycb | enriquetaso: thank you, appreciate it | 14:43 |
| enriquetaso | #startmeeting cinder_bs | 15:00 |
| opendevmeet | Meeting started Wed Feb 23 15:00:13 2022 UTC and is due to finish in 60 minutes. The chair is enriquetaso. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
| opendevmeet | The meeting name has been set to 'cinder_bs' | 15:00 |
| enriquetaso | 13 new bugs reported this period. | 15:00 |
| enriquetaso | Full list of bugs: | 15:00 |
| enriquetaso | #link http://lists.openstack.org/pipermail/openstack-discuss/2022-February/027391.html | 15:00 |
| fabiooliveira | o/ | 15:00 |
| yuval | my favorite meeting :) | 15:00 |
| enriquetaso | https://bugs.launchpad.net/os-brick/+bug/1961102 | 15:01 |
| enriquetaso | NVMe-oF doesn't disconnect | 15:01 |
| enriquetaso | #topic NVMe-oF doesn't disconnect | 15:01 |
| enriquetaso | #link https://bugs.launchpad.net/os-brick/+bug/1961102 | 15:01 |
| enriquetaso | #topic Request ID | 15:03 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1960329 | 15:03 |
| enriquetaso | Wrong request ID on API middleware filters. ^ | 15:03 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1960021 | 15:03 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1960020 | 15:03 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1960019 | 15:04 |
| enriquetaso | yuval++ | 15:05 |
| enriquetaso | The idea is to get them in yoga and backport them. | 15:08 |
| enriquetaso | #endmeeting | 15:09 |
| opendevmeet | Meeting ended Wed Feb 23 15:09:33 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:09 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/cinder_bs/2022/cinder_bs.2022-02-23-15.00.html | 15:09 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/cinder_bs/2022/cinder_bs.2022-02-23-15.00.txt | 15:09 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/cinder_bs/2022/cinder_bs.2022-02-23-15.00.log.html | 15:09 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for group-snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804757 | 15:10 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for group-snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804757 | 15:12 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Fix request id mess on Cinder-API https://review.opendev.org/c/openstack/cinder/+/827860 | 15:13 |
| geguileo | eharney: ^ done | 15:13 |
| geguileo | rosmaita: Could you review it again? I only removed 1 comment | 15:13 |
| rosmaita | geguileo: sure, no problem | 15:27 |
| geguileo | rosmaita: thanks | 15:27 |
| rosmaita | (was outside, that mouse is still stuck in my window well) | 15:27 |
| opendevreview | Eric Harney proposed openstack/cinder master: Don't init backup API code for each create volume request https://review.opendev.org/c/openstack/cinder/+/830632 | 15:27 |
| yuval | hey, small question in volume type property we allow a property to have value "True" or "<is> True" | 15:28 |
| yuval | why like this? | 15:28 |
| yuval | I mean the "<is> True" forcing us to parse the value "understand" it | 15:29 |
| rosmaita | i think because they are really strings, this way there's no confusion | 15:29 |
| yuval | is that some legacy thing | 15:29 |
| rosmaita | what i mean is bool("False") is True in python | 15:30 |
| yuval | yes, but the "<is> " is cinder convention right? its something python support | 15:31 |
| yuval | ? | 15:31 |
| rosmaita | yuval: yes, it's a cinder thing | 15:32 |
| yuval | the result are functions like this:https://opendev.org/openstack/cinder/src/branch/master/cinder/volume/drivers/qnap.py#L281 | 15:32 |
| jungleboyj | A Cinder thing that has always baffled me to some extent. :-) | 15:36 |
| opendevreview | Merged openstack/cinder master: Remove the need for project_id from API endpoints https://review.opendev.org/c/openstack/cinder/+/822040 | 15:40 |
| rosmaita | \o/ | 15:40 |
| opendevreview | Eric Harney proposed openstack/cinder master: Only init message API in create volume when needed https://review.opendev.org/c/openstack/cinder/+/830634 | 15:41 |
| rosmaita | geguileo: looks like https://review.opendev.org/c/openstack/cinder/+/827860/ has gone into merge conflict | 15:43 |
| geguileo | rofl rofl | 15:44 |
| rosmaita | you win some, you lose some | 15:44 |
| rosmaita | whoami-rajat: if you're still around, alan's project_id patch has merged | 15:46 |
| geguileo | it's abishop's patch to not require the project id | 15:46 |
| geguileo | rosmaita: that's precisely the one that broke mine | 15:46 |
| geguileo | lol | 15:46 |
| rosmaita | :D | 15:46 |
| hemna | https://github.com/openstack/cinder/blob/master/cinder/scheduler/filters/capacity_filter.py#L131-L132. shouldn't that be provisioned + requested / available_total | 15:48 |
| hemna | where available_total is total - (total * reserved %) | 15:49 |
| whoami-rajat | rosmaita, great, will update | 15:49 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Fix request id mess on Cinder-API https://review.opendev.org/c/openstack/cinder/+/827860 | 15:49 |
| hemna | instead of subtracting the reserved space from the reported free space | 15:49 |
| geguileo | rosmaita: eharney ^ let's try again without the merge conflict | 15:49 |
| hemna | reserved_percent is defined as 'The percentage of backend capacity is reserved', so why is it being applied to free space. The total available is the capacity, not the currently free. | 15:54 |
| hemna | the filter scheduler's algorithm is not right IMHO | 15:56 |
| yuval | hemna what you mean? | 15:57 |
| hemna | the calculations for the overprovisioning and available space isn't right I think | 15:57 |
| hemna | and it's being done differently in a few places within cinder | 15:58 |
| yuval | ot: sorry for missing the meeting today | 15:58 |
| rosmaita | yuval: ack | 15:59 |
| rosmaita | hemna: i got sidetracked yesterday and never found that doc about how cinder computes available space | 15:59 |
| *** gmann is now known as gmann_afk | 15:59 | |
| hemna | well, I'm looking at the code and it seems quite weird and inconsistent | 16:00 |
| hemna | I'm working on another gist with some of my findings. | 16:00 |
| rosmaita | cool, that will be helpful | 16:02 |
| rosmaita | i guess we need to abstract these computations out to common utility code | 16:02 |
| hemna | I think the current way of calculating the provisioned ratio leads to under reporting of the ratio because it uses the total space instead of the available space (total - reserved) | 16:02 |
| hemna | heh | 16:02 |
| opendevreview | Harsh Ailani proposed openstack/cinder master: [SVf]:Fix multiple lsvdisk calls for GMCV create volume operation. https://review.opendev.org/c/openstack/cinder/+/829110 | 16:02 |
| hemna | you mean like this? https://review.opendev.org/c/openstack/cinder/+/830456 | 16:03 |
| rosmaita | :D | 16:03 |
| hemna | anyway, the provisioned ration for one of my pools is reporting 0.81 | 16:03 |
| hemna | when it's actually over provisioned already | 16:04 |
| hemna | > 1 | 16:04 |
| hemna | because it's allocated_capacity / total | 16:04 |
| hemna | intead of allocated / (total - reserved) | 16:04 |
| hemna | and the free space is free reported - reserved | 16:04 |
| hemna | reserved is supposed to be applied to the total capacity, not the currently free, which changes | 16:05 |
| hemna | the reason I'm going over this, is because I'm seeing provisioning requests going through in my scheduler that thinks there is enough free space, but there isn't. | 16:06 |
| hemna | so the create fails | 16:06 |
| hemna | so at least I'd like to get a consistent answer for what it means to be provisioning ratio, virtual free, actual available, etc | 16:06 |
| hemna | it's probably been overlooked for years because not many have run into low available pools deployments | 16:07 |
| rosmaita | i suspect you are correct about that | 16:07 |
| hemna | we tend to run our individual pools down to low left, then alert with our tooling on that and then create new pools when needed. | 16:08 |
| *** gmann_afk is now known as gmann | 16:17 | |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for group os-reset_status https://review.opendev.org/c/openstack/cinder/+/804735 | 16:58 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for group os-reset_status https://review.opendev.org/c/openstack/cinder/+/804735 | 16:58 |
| whoami-rajat | rosmaita, hey, was rebasing on top of Alan's patch and found one issue (non-fatal) on his patch | 17:10 |
| whoami-rajat | rosmaita, it's missing an entry in the microversions.py file | 17:10 |
| rosmaita | oops | 17:10 |
| whoami-rajat | i think one way to go about this is to propose a quick patch to update it and we can ninja approve? | 17:11 |
| rosmaita | yes | 17:11 |
| rosmaita | i can't believe i missed that one | 17:12 |
| rosmaita | we have too many weird places you need to update when adding a mv | 17:12 |
| whoami-rajat | it's easy to miss since we use that reference in our code to allow/disallow things and in case of Alan's patch we don't have any condition check for mv.SUPPORT_PROJECT_ID (or something) | 17:13 |
| rosmaita | ... and that's the file that tells you where all those places are! | 17:13 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 17:13 |
| rosmaita | whoami-rajat: do you have time to propose the patch or do you want me to do it? | 17:14 |
| opendevreview | Tushar Trambak Gite proposed openstack/cinder master: Reset state robustification for snapshot os-reset_status https://review.opendev.org/c/openstack/cinder/+/804035 | 17:14 |
| whoami-rajat | rosmaita, I can do it | 17:15 |
| rosmaita | ok, thanks | 17:15 |
| rosmaita | whoami-rajat: PROJECT_ID_OPTIONAL_IN_URL | 17:16 |
| rosmaita | (call it that) | 17:16 |
| whoami-rajat | thanks, i was thinking about the right name to give | 17:16 |
| rosmaita | :) | 17:16 |
| opendevreview | Rajat Dhasmana proposed openstack/cinder master: Add missing microversion entry for 3.67 https://review.opendev.org/c/openstack/cinder/+/830669 | 17:22 |
| whoami-rajat | rosmaita, ^ | 17:22 |
| rosmaita | ty | 17:22 |
| rosmaita | whoami-rajat: is going to fail pep8, left a comment on the patch | 17:27 |
| whoami-rajat | ah blank space, will update it thanks | 17:29 |
| rosmaita | hemna: finally found that doc, don't know if it will be helpful, but it has the "Cinder" definitions of some terminology: https://specs.openstack.org/openstack/cinder-specs/specs/queens/provisioning-improvements.html | 17:31 |
| opendevreview | Rajat Dhasmana proposed openstack/cinder master: Add missing microversion entry for 3.67 https://review.opendev.org/c/openstack/cinder/+/830669 | 17:34 |
| rosmaita | whoami-rajat: ty! i went ahead and workflowed it, but since it touches a .py file, it's going to go through full check and gate, so probably won't merge for a few hours | 17:39 |
| whoami-rajat | rosmaita, np, I hope we could do something about it in the future (at least not run the full job stack for small changes) | 17:40 |
| whoami-rajat | I will go ahead and update my patch | 17:40 |
| whoami-rajat | and rebase on top of this in the morning | 17:40 |
| eharney | that missing microversion in microversions.py looks like a fun thing to add a unit test for, i might poke at that | 17:41 |
| whoami-rajat | eharney++ | 17:41 |
| rosmaita | eharney: i was thinking about that, maybe checking that the max reported in the versions response is represented in the constants in microversions.py | 17:43 |
| eharney | rosmaita: yup, that was my plan | 17:43 |
| rosmaita | cool | 17:43 |
| opendevreview | Brian Rosmaita proposed openstack/cinder master: Update doc8 ignore-path https://review.opendev.org/c/openstack/cinder/+/830674 | 17:59 |
| opendevreview | Rajat Dhasmana proposed openstack/cinder master: Support volume re-image https://review.opendev.org/c/openstack/cinder/+/606346 | 18:00 |
| opendevreview | Eric Harney proposed openstack/cinder master: Add a unit test to keep microversions.py up to date https://review.opendev.org/c/openstack/cinder/+/830676 | 18:18 |
| eharney | rosmaita: whoami-rajat: that does it ^ | 18:18 |
| whoami-rajat | great! | 18:19 |
| opendevreview | yuval proposed openstack/cinder master: lightos - bugfix compression stats should be True https://review.opendev.org/c/openstack/cinder/+/830538 | 18:38 |
| hemna | rosmaita thanks, checking it out | 19:17 |
| hemna | heh only 4 or 5 drivers report provisioned_capacity_gb, even though that doc says it's required | 19:21 |
| rosmaita | i guess not everyone got the memo | 19:22 |
| hemna | ah yah and that document describes the reserved percentage too, which is what I thought it would be | 19:28 |
| hemna | but it's not implemented that way in the capacity filter | 19:28 |
| hemna | so provisioned capacity is the allocated capacity in cinder | 19:29 |
| hemna | which is often incorrect at startup time | 19:29 |
| hemna | that value is only calculated by the volume service when it starts from available and in-use volumes | 19:29 |
| hemna | it doesn't take into account migrating, attaching, reserved, deleting, detaching, maintenance | 19:30 |
| hemna | so what I'm seeing quite often is that allocated capacity going negative | 19:31 |
| opendevreview | Rajat Dhasmana proposed openstack/python-cinderclient master: Add volume reimage command https://review.opendev.org/c/openstack/python-cinderclient/+/606891 | 19:31 |
| opendevreview | Eric Harney proposed openstack/cinder master: Add a unit test to keep microversions.py up to date https://review.opendev.org/c/openstack/cinder/+/830676 | 20:56 |
| *** dviroel is now known as dviroel|out | 21:18 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!