Friday, 2017-10-06

smcginnisjungleboyj: Yeah, looks like the same thing:
jungleboyjRemote error: ProcessExecutionError Unexpected error while running command.\nCommand: sudo cinder-rootwrap /etc/cinder/rootwrap.conf lvchange -a y --yes -K stack-volumes-lvmdriver-1/_snapshot-6544f849-be4e-437e-9f65-c5cfaf78a6ff\nExit code: 5\nStdout: u''\nSt", "is_incremental": false, "id": "dd6bdd20-1b60-4782-92c8-c9a856a53019", "size": 1}}00:01
jungleboyjHe he.  Found that at the same time as you.00:02
jungleboyjOk, so that is the thin provisioning issue.  So the other one is the issues with cascade delete that I am intermittently seeing.00:02
jungleboyjDo you know if we have bugs open for either of those yet?00:03
smcginnisWell, it's just the one bug.00:11
*** ntpttr_laptop has quit IRC00:12
* jungleboyj is confused00:12
jungleboyjsmcginnis: So are both problems caused by the thin provisioning with LVM issues we think?00:12
smcginnisThe error "device-mapper: reload ioctl on (252:4) failed: Invalid argument". It's just bubbling up in multiple places.00:13
jungleboyjOk.  I misinterpreted what you had said earlier.00:14
*** hemna__ has quit IRC00:14
smcginnisjungleboyj: Seems to be a timing issue since we don't always see it.00:15
smcginnisTiming with itself, or timing between it and other concurrent operations that are going on at the time.00:15
jungleboyjYeah, I was trying to get a VM up on the PowerEdge box to try and recreate.00:16
jungleboyjGotta go feed the boys.  Back online later.00:17
*** crushil has joined #openstack-cinder00:19
*** crushil has quit IRC00:23
*** crushil has joined #openstack-cinder00:25
*** hemna__ has joined #openstack-cinder00:25
*** crushil has quit IRC00:26
*** savvas has joined #openstack-cinder00:38
*** Nel1x has quit IRC00:44
*** savvas has quit IRC00:58
*** crushil has joined #openstack-cinder01:01
*** markstur has quit IRC01:04
*** Apoorva has quit IRC01:11
*** openstackstatus has quit IRC01:14
*** openstackstatus has joined #openstack-cinder01:14
*** ChanServ sets mode: +v openstackstatus01:14
*** mriedem has quit IRC01:26
*** zigo has quit IRC01:27
*** dikonoor has joined #openstack-cinder01:29
*** zigo has joined #openstack-cinder01:31
*** markvoelker has joined #openstack-cinder01:52
*** markstur has joined #openstack-cinder01:57
*** wN has quit IRC02:04
*** sdake has quit IRC02:04
*** sweston has quit IRC02:04
*** sweston has joined #openstack-cinder02:05
*** wiggin15 has quit IRC02:05
*** wiggin15 has joined #openstack-cinder02:05
*** andymccr has quit IRC02:06
*** rha has quit IRC02:06
*** lyarwood has quit IRC02:06
*** wN has joined #openstack-cinder02:06
*** wN has quit IRC02:06
*** wN has joined #openstack-cinder02:06
*** twouters has quit IRC02:06
*** gerhardqux has quit IRC02:06
*** gerhardqux has joined #openstack-cinder02:06
*** mtreinish has quit IRC02:08
*** lyarwood has joined #openstack-cinder02:08
*** jistr has quit IRC02:08
*** john5223_ has quit IRC02:08
*** savvas has joined #openstack-cinder02:09
*** stephenfin has quit IRC02:09
*** dirk_ has joined #openstack-cinder02:09
*** enriquetaso_ has joined #openstack-cinder02:09
*** jistr has joined #openstack-cinder02:10
*** mtreinish has joined #openstack-cinder02:10
openstackgerritMerged openstack/cinder master: Remove QNAP volume driver
*** twouters has joined #openstack-cinder02:11
*** twouters has quit IRC02:11
*** twouters has joined #openstack-cinder02:11
*** stephenfin has joined #openstack-cinder02:11
*** enriquetaso has quit IRC02:12
*** dirk has quit IRC02:12
*** kambiz has quit IRC02:12
*** logan- has quit IRC02:12
*** enriquetaso_ is now known as enriquetaso02:12
*** dirk_ is now known as dirk02:12
*** sdake has joined #openstack-cinder02:13
*** andymccr has joined #openstack-cinder02:13
*** sdake is now known as Guest284902:13
*** savvas has quit IRC02:14
*** kambiz has joined #openstack-cinder02:14
*** rha has joined #openstack-cinder02:14
*** logan- has joined #openstack-cinder02:15
*** crushil has quit IRC02:17
*** crushil has joined #openstack-cinder02:22
*** mgagne has quit IRC02:23
*** mgagne has joined #openstack-cinder02:24
*** mgagne is now known as Guest6609802:24
*** markvoelker has quit IRC02:27
*** jwcroppe has joined #openstack-cinder02:30
*** diegows_ has joined #openstack-cinder02:32
*** dave-mccowan has quit IRC02:33
*** diegows has quit IRC02:35
*** sapd has joined #openstack-cinder02:38
*** sapd has quit IRC02:38
*** ericyoung has quit IRC02:38
*** ericyoung has joined #openstack-cinder02:41
*** crushil has quit IRC02:52
*** crushil has joined #openstack-cinder02:53
*** nicolasbock_ has quit IRC03:00
*** nicolasbock has quit IRC03:01
*** edmondsw has joined #openstack-cinder03:04
*** edmondsw has quit IRC03:09
*** felipemonteiro_ has joined #openstack-cinder03:12
openstackgerritMerged openstack/cinder master: Dell EMC PS: Fix over-subscription ratio stats
*** markvoelker has joined #openstack-cinder03:24
*** wes_dillingham has quit IRC03:29
*** wes_dillingham has joined #openstack-cinder03:30
*** wes_dillingham has quit IRC03:30
*** links has joined #openstack-cinder03:32
*** mriedem has joined #openstack-cinder03:35
*** abhitechie has quit IRC03:36
*** mriedem has quit IRC03:42
*** gouthamr has quit IRC03:42
*** crushil has quit IRC03:52
*** crushil has joined #openstack-cinder03:53
*** markvoelker has quit IRC03:57
*** gkadam has joined #openstack-cinder04:01
*** arnewiebalck_ has joined #openstack-cinder04:05
*** armax has quit IRC04:23
*** felipemonteiro_ has quit IRC04:49
openstackgerritMerged openstack/cinder master: docs: Fix typo error in blockstorage-driver-filter-weighing.rst
pooja_jadhavdiablo_rojo: Hi nelson, Please help me to review these patches. and
*** edmondsw has joined #openstack-cinder04:52
*** markvoelker has joined #openstack-cinder04:54
*** edmondsw has quit IRC04:57
*** markstur has quit IRC05:02
*** markstur has joined #openstack-cinder05:03
*** psachin has joined #openstack-cinder05:05
*** lpetrut_ has joined #openstack-cinder05:06
*** markstur has quit IRC05:08
*** arnewiebalck_ has quit IRC05:15
*** jwcroppe has quit IRC05:15
*** markvoelker has quit IRC05:27
*** kwathore has joined #openstack-cinder05:27
*** ChubYann has quit IRC05:34
*** crushil has quit IRC05:37
openstackgerritMerged openstack/cinder master: Doc: Configuration: Remove some outdated Ceph info
*** e0ne has joined #openstack-cinder05:55
*** spectr has joined #openstack-cinder05:56
*** e0ne has quit IRC06:02
*** jwcroppe has joined #openstack-cinder06:03
*** lpetrut_ has quit IRC06:06
*** hemna_ has joined #openstack-cinder06:07
*** hemna__ has quit IRC06:11
*** jwcroppe has quit IRC06:14
*** rcernin has joined #openstack-cinder06:17
openstackgerritSilvan Kaiser proposed openstack/cinder master: [WIP] Adds Overlay Volumes Created from Snapshots to Quobyte
*** markvoelker has joined #openstack-cinder06:24
*** pcaruana has joined #openstack-cinder06:24
*** pgadiya has joined #openstack-cinder06:27
*** pgadiya has quit IRC06:27
*** e0ne has joined #openstack-cinder06:31
*** rcernin has quit IRC06:41
*** links has quit IRC06:42
*** rcernin has joined #openstack-cinder06:43
*** e0ne has quit IRC06:46
*** isaacb has joined #openstack-cinder06:46
*** swamireddy has quit IRC06:55
*** markvoelker has quit IRC06:58
*** wanghao has joined #openstack-cinder06:59
*** links has joined #openstack-cinder06:59
*** swamireddy has joined #openstack-cinder07:04
*** jwcroppe has joined #openstack-cinder07:05
*** wanghao has quit IRC07:09
*** tesseract has joined #openstack-cinder07:25
*** wanghao has joined #openstack-cinder07:27
*** jwcroppe has quit IRC07:27
*** isaacb has quit IRC07:30
*** wanghao has quit IRC07:33
*** wanghao has joined #openstack-cinder07:33
*** wanghao has quit IRC07:34
*** wanghao has joined #openstack-cinder07:34
*** wanghao has quit IRC07:35
*** links has quit IRC07:46
*** e0ne has joined #openstack-cinder07:47
*** markvoelker has joined #openstack-cinder07:54
*** mlakat has quit IRC07:56
*** links has joined #openstack-cinder07:59
*** mlakat has joined #openstack-cinder08:28
*** markvoelker has quit IRC08:28
*** edmondsw has joined #openstack-cinder08:28
openstackgerritAlexandru Muresan proposed openstack/cinder master: RemoteFS: revert snapshot support
*** edmondsw has quit IRC08:33
*** psachin has quit IRC08:34
*** jwcroppe has joined #openstack-cinder08:49
*** jwcroppe has quit IRC08:58
*** dalgaaf has quit IRC09:18
*** ociuhandu has joined #openstack-cinder09:18
*** psachin has joined #openstack-cinder09:18
*** dalgaaf has joined #openstack-cinder09:19
*** ociuhandu has quit IRC09:21
*** markvoelker has joined #openstack-cinder09:25
*** spectr has quit IRC09:26
ArchiFleKsHi, I'm running into this in latest master when running cinder-manage db online_data_migrations (
ArchiFleKsTypeError: Message objects do not support addition.09:28
*** spectr has joined #openstack-cinder09:34
*** spectr has quit IRC09:34
openstackgerritNam Nguyen Hoai proposed openstack/cinder master: Cleanup test-requirements
*** spectr has joined #openstack-cinder09:36
*** spectr has quit IRC09:36
*** spectr has joined #openstack-cinder09:37
*** spectr has quit IRC09:38
*** spectr has joined #openstack-cinder09:40
*** spectr has quit IRC09:40
*** spectr has joined #openstack-cinder09:41
*** spectr has quit IRC09:43
*** spectr has joined #openstack-cinder09:43
*** kwathore_ has joined #openstack-cinder09:52
*** kwathore has quit IRC09:52
*** spectr has quit IRC09:56
*** spectr has joined #openstack-cinder09:57
*** spectr has quit IRC09:57
*** spectr has joined #openstack-cinder09:58
*** markvoelker has quit IRC09:58
*** stephenfin is now known as finucannot10:05
*** edmondsw has joined #openstack-cinder10:16
*** edmondsw has quit IRC10:21
*** sdague has joined #openstack-cinder10:28
*** Dinesh_Bhor has quit IRC10:35
*** jwcroppe has joined #openstack-cinder10:39
*** jwcroppe has quit IRC10:47
*** ganso has joined #openstack-cinder10:48
*** savvas has joined #openstack-cinder10:55
*** markvoelker has joined #openstack-cinder10:56
*** nicolasbock has joined #openstack-cinder11:10
*** nicolasbock_ has joined #openstack-cinder11:10
*** nicolasbock_ has quit IRC11:15
*** nicolasbock has quit IRC11:15
*** dave-mccowan has joined #openstack-cinder11:20
*** nicolasbock has joined #openstack-cinder11:27
*** nicolasbock_ has joined #openstack-cinder11:27
*** nicolasbock_ has quit IRC11:29
*** nicolasbock has quit IRC11:29
*** markvoelker has quit IRC11:29
*** takedakn has joined #openstack-cinder11:34
*** abishop has joined #openstack-cinder11:52
*** spectr has quit IRC12:02
*** spectr has joined #openstack-cinder12:03
*** edmondsw has joined #openstack-cinder12:05
*** wes_dillingham has joined #openstack-cinder12:07
*** edmondsw has quit IRC12:09
hemna_It's Friday peeps!12:17
*** tpsilva has joined #openstack-cinder12:19
*** MVenesio has joined #openstack-cinder12:21
*** nicolasbock has joined #openstack-cinder12:22
*** markvoelker has joined #openstack-cinder12:26
*** jwcroppe has joined #openstack-cinder12:28
*** spectr has quit IRC12:28
*** jdwidari has joined #openstack-cinder12:30
*** markvoelker has quit IRC12:31
*** markvoelker has joined #openstack-cinder12:31
*** jwcroppe has quit IRC12:36
*** spectr has joined #openstack-cinder12:39
*** nicolasbock has quit IRC12:39
*** mchlumsky has joined #openstack-cinder12:40
*** erlon has joined #openstack-cinder12:41
*** alexchadin has joined #openstack-cinder12:45
*** takedakn has quit IRC12:47
*** nicolasbock has joined #openstack-cinder12:51
*** mchlumsky has quit IRC12:56
*** takedakn has joined #openstack-cinder12:58
*** mchlumsky has joined #openstack-cinder13:01
*** crushil has joined #openstack-cinder13:02
e0nehemna_: it's almost Friday's evening!13:04
*** psachin has quit IRC13:05
*** nicolasbock has quit IRC13:05
*** nicolasbock has joined #openstack-cinder13:05
*** takedakn has quit IRC13:05
*** takedakn has joined #openstack-cinder13:07
*** eharney has joined #openstack-cinder13:12
*** lbragstad has joined #openstack-cinder13:15
*** dustins has joined #openstack-cinder13:17
*** takedakn has quit IRC13:18
*** gkadam has quit IRC13:18
*** dansmith is now known as superdan13:28
*** jmlowe_ has quit IRC13:29
*** jmlowe has joined #openstack-cinder13:30
openstackgerritSean McGinnis proposed openstack/cinder master: Remove Reduxio volume driver
*** jmlowe_ has joined #openstack-cinder13:33
openstackgerritSean McGinnis proposed openstack/cinder master: Remove Tegile volume driver
*** spectr has quit IRC13:35
*** jmlowe has quit IRC13:35
*** lbragstad has quit IRC13:36
*** lbragstad has joined #openstack-cinder13:36
*** spectr has joined #openstack-cinder13:36
*** mriedem has joined #openstack-cinder13:39
*** gouthamr has joined #openstack-cinder13:39
*** wes_dillingham has quit IRC13:45
*** wes_dillingham has joined #openstack-cinder13:46
*** links has quit IRC13:48
*** edmondsw has joined #openstack-cinder13:53
*** jdwidari has quit IRC13:54
*** jwcroppe has joined #openstack-cinder13:55
* hemna_ is ceph-ing13:57
*** edmondsw has quit IRC13:57
smcginnisI've heard of that.13:57
hemna_cephilus ?13:57
*** catintheroof has joined #openstack-cinder14:05
*** catinthe_ has joined #openstack-cinder14:06
*** catintheroof has quit IRC14:07
*** spectr has quit IRC14:11
*** alexchadin has quit IRC14:15
*** gouthamr has quit IRC14:16
*** savvas has quit IRC14:19
*** felipemonteiro_ has joined #openstack-cinder14:21
*** alexchadin has joined #openstack-cinder14:22
*** felipemonteiro__ has joined #openstack-cinder14:22
*** alexchadin has quit IRC14:23
*** felipemonteiro_ has quit IRC14:26
*** armax has joined #openstack-cinder14:26
jungleboyjTGIF ... been quite the week.14:36
* jungleboyj doesn't want to get cephilus from hemna_ 14:36
*** ntpttr_laptop has joined #openstack-cinder14:50
hemna_it spreads through irc probably already have it.14:51
*** markstur has joined #openstack-cinder14:52
smcginnisAnyone know under what conditions an LVM thin pool is "suspended"?14:55
smcginnisThinly provisioned, but shouldn't be overprovisioned. Happening in our tests, and I don't think we really have anything that actually writes much data.14:57
*** ntpttr_laptop has quit IRC14:58
twoutersgoogle says that the origin device is suspended during snapshots :)15:01
jungleboyjtwouters:  You googled better than I.15:02
smcginnistwouters: Oh, hmm. So maybe it is still in the process of creating the snapshot when we try to activate it. Thanks!15:02
smcginnisThat could be a good clue.15:02
smcginnisWe may need to add a check in there to wait for completion.15:02
smcginnisCould also explain the intermittent nature of the failures we've been seeing.15:03
twoutersjungleboyj: "lvm thin suspend" :p15:03
*** spectr has joined #openstack-cinder15:04
*** spectr has quit IRC15:04
*** links has joined #openstack-cinder15:07
*** gouthamr has joined #openstack-cinder15:07
*** mriedem is now known as ronlund15:09
*** MVenesio has quit IRC15:16
*** rcernin has quit IRC15:19
Swanson"cephilus" is genius.15:22
*** shvepsy has quit IRC15:25
*** shvepsy has joined #openstack-cinder15:25
*** jwcroppe has quit IRC15:28
*** jwcroppe has joined #openstack-cinder15:29
*** savvas has joined #openstack-cinder15:29
*** jwcroppe has quit IRC15:33
*** savvas has quit IRC15:34
*** jwcroppe has joined #openstack-cinder15:36
*** edmondsw has joined #openstack-cinder15:41
*** ildikov is now known as coffee_cat15:45
*** edmondsw has quit IRC15:45
*** markstur_ has joined #openstack-cinder15:59
openstackgerritSean McGinnis proposed openstack/cinder master: Add retries to LVM logical volume activation
smcginnisOptimistically marking review 510168 as "Closes-bug". If that merges and we still see this I will reopen.16:01
*** Apoorva has joined #openstack-cinder16:01
smcginnisjungleboyj, hemna_, e0ne, all: Was hoping for a more deterministic fix, but I think that might do it. ^^16:02
*** Apoorva has quit IRC16:02
smcginnisI'd like to get that pushed through and see if we observe any improvement.16:02
*** Apoorva has joined #openstack-cinder16:02
jungleboyjsmcginnis: Let me look.16:02
*** markstur has quit IRC16:02
jgriffithsmcginnis in the past we've introduced some forced delays to pin that down, might be worth trying again16:03
smcginnisWe can also check out the logs for any sign that it has been doing the retries and succeeding. That could give us a little more assurance it's actually working.16:03
smcginnisjgriffith: I thought about putting in some actual greenthread.sleep(), but figured with the retry we wouldn't be blocking if we didn't need to be.16:03
smcginnisjgriffith: But that may be another option.16:03
jgriffithspecifically the lvm calls rapid fire from tempest sometimes get a bit hay wire, so we would introduce a delay after a create of a few seconds16:03
jgriffithyeah, the retry thing is funky though with device-mapper16:04
jgriffithsometimes you bork device-mapper to the point where it doesn't recover, or it waits for it's time out in some cases before letting you do anything16:04
smcginnisjgriffith: We could if do "if lvm_type == thin: sleep"16:04
jgriffithwhich IIRC the default is like 30 seconds or something :(16:04
jgriffithsmcginnis that's not a bad idea16:04
jgriffiththe sleep if thin16:04
smcginnisHmm. I put the backoff delay to 2 seconds with 5 retries thinking that could be enough, but maybe we want to make that even biggerer if needed.16:05
jgriffithnah, I'd try that first16:05
smcginnisDo like a 10 second backoff. Then by the 5th attempt it could wait long enough for DM to get straight.16:05
jgriffithwell, you could start with what you suggested there and collect some data, then go up if needed.  Or start the other way and work down16:06
jgriffithyou're call, six one, half dozen the other I guess16:06
smcginnisjgriffith: I wanted to start a little conservative so we didn't introduce too big of a slow down if it wasn't actually needed. But yeah, let's see how it goes.16:07
jgriffithi don't think in this case you're hitting the dm or lvm internal timeout mechanisms though FWIW16:07
smcginnisYeah, if it's 30 seconds this is going to fail before then.16:07
jgriffithlvm activiation has been a bit of a pain ever since it was introduced16:07
smcginnisReally seems like it shouldn't be this difficult.16:08
jgriffithwe fumbled around like this quite a bit16:08
jgriffithI know right!!!16:08
smcginnisFingers crossed this will get us past it. If so, we should probably backport to pike.16:08
smcginnisNot really happy this has been around for so long and we've just bypassed it by having devstack switch things to thick provisioned. :/16:09
jungleboyjsmcginnis:  Thank you for digging into this.16:10
smcginnisMaybe we should take the entire local_dev/lvm file and put a retry with a 30 second delay on every call. :D16:10
smcginnisjungleboyj: I don't know if you've been getting any of it, but I've been getting more and more grumbling from others about the test failures, so I'm hoping we can get past this.16:11
jgriffithLet me know if you need help on it, I may get my stuff fixed and finished today16:11
smcginnisjgriffith: We're actually almost to that point if you scan through that file. :)16:11
jgriffithwe need to rewrite all of that :)16:11
smcginnisjgriffith: Cool, we'll see how it goes.16:11
smcginnisjgriffith: If we still get failures, I'd love to have your eyes on it with your LVM experience. I know it's much more than mine.16:12
jgriffithIt's too bad, once a year or so we have these crazy intermittent LVM issues that just kill us16:12
smcginnisWould be great if there was a good API for interacting with LVM. Other than running commands and parsing output.16:13
*** pcaruana has quit IRC16:13
*** jwcroppe has quit IRC16:13
smcginnisjgriffith: Have you played around with that at all? If it's the one I looked at awhile back, it didn't seem like it was very active.16:14
jgriffithThere are several I have found and messed with over the years, but none seemed that great16:14
smcginnisUh, yeah - Uploaded on 2012-06-0816:14
smcginnisI think LVM has had a few tweaks since then.16:14
jgriffithyeah, that's the problem... most are about 5 years old16:15
smcginnisI suppose we could fork it and make it into what we need it to be.16:15
smcginnisWould take some work - "the resize of Physical and Logical volumes is not implemented in the api yet"16:16
*** xyang1 has joined #openstack-cinder16:16
jungleboyjsmcginnis:  I haven't been getting that, but it clearly has become insane.16:16
*** xyang1 has quit IRC16:16
*** jwcroppe has joined #openstack-cinder16:16
*** xyang1 has joined #openstack-cinder16:17
smcginnisDefinitely picked up:
jungleboyjsmcginnis:  Yeah, I saw that yesterday.16:17
coffee_catsmcginnis: problem with forks is you need to maintain it, or convince jgriffith to do that for you as he's advocating on LVM needing some love from all of us :)16:17
jgriffithcoffee_cat or just *steal it* and adopt it as your own offspring16:18
coffee_catjgriffith: if it fits, then why not?16:19
smcginnisIt at least fits in and would be used by Cinder. But yeah, we don't really need to look for more things to maintain.16:19
* jgriffith has pretty much given up on the dream of many people caring about LVM16:19
coffee_catjgriffith: sadness :(16:19
jgriffiththere's definitely a few but not as many as there should be16:19
smcginnisWe need to start selling an LVM Cinder appliance. ;)16:20
jungleboyjI think I am going to be using LVM for the cloud I am putting together.  :-)16:20
coffee_catsmcginnis: it only takes good marketing :)16:20
smcginnisJust take one of these and preconfigure it with c-vol:
* coffee_cat is hearing jungleboyj volunteering for maintenance :)16:21
jgriffithsmcginnis +116:21
smcginnisPlug into your rack and perform a couple config steps. Cloud storage ready to go.16:21
coffee_catsmcginnis: that brings up old memories of changing a motherboard in a supermicro server and I couldn't find cables anywhere that would've fit, but at least everyone reminded me supermicro is terrible on the way... :)16:23
smcginniscoffee_cat: Really? I've had good luck with them.16:23
smcginnisMy worst was something like a 6U Sun server I maintained. Had to replace the motherboard.16:23
smcginnisHad to go out and find a super long screwdriver to be able to get to one of the screws.16:24
coffee_catsmcginnis: it was a success story at the end, and trying to find a cable the night before Christmas might've been a bit ambitious of me as well :)16:24
smcginniscoffee_cat: Haha, bad timing.16:24
coffee_catsmcginnis: lol, if even the screwdriver doesn't fit, that really sucks... :/ :)16:25
coffee_catsorry, didn't want to hijack the chat16:25
smcginnisEase of maintenance was definitely not on their minds when they designed it.16:26
coffee_catI don't think that's ever on the minds for real TBH...16:27
*** dikonoor has quit IRC16:28
smcginnisI've actually been impressed with some of them now with tool-less access and things.16:28
smcginnisEven rack rails have gotten so much better that just snap in.16:28
coffee_catoh nice!16:30
coffee_catwe should adapt that in software too :)16:30
*** dikonoor has joined #openstack-cinder16:38
SwansonI'll use LVM when I make my cloud running on the 12 unneeded cores of the thread ripper CPU I want to buy for no good reason.16:38
smcginnisWorking well with my Ryzen 1700. :)16:41
Swansonpuny 8 cores?16:41
smcginnisTwo threads per core though.16:41
smcginnisMy 32G RAM is the more limiting factor.16:42
SwansonYeah. I love when you get a 32 core server and they give you 32GB. Like, yay!16:44
openstackgerritMerged openstack/cinder master: Remove Reduxio volume driver
SwansonI've powered none of your stuff up in a month. It's like you never worked here.16:45
smcginnisIf no one wants that stack of servers, I can pull up to the loading dock in about 30 minutes.16:46
openstackgerritSorin Sbarnea proposed openstack/cinder master: bump pylint to recent version
*** wes_dillingham has quit IRC16:47
SwansonOh, when I get a chance I'm having your's and Mike's moved into a rack where I will install Openstack on a huge cluster and run one single instance of ubuntu on it.16:48
SwansonUnseen performance art is the new performance art.16:48
smcginnisThere's massive compute power sitting idle in that lab. I think you need to start mining for bitcoin or something.16:49
smcginnisOr process all of Seti@Home over the weekend.16:50
* hemna_ misses the days of using screen for devstack17:00
hemna_seriously hate systemd17:00
Swansonhemna, In openstack years devstack was 60 years ago.17:01
SwansonSorry, screen17:01
hemna_Swanson, you had me there for a second...I was like....did everyone move to containers for dev already?17:02
Swansonhemna, That's coming up in 60 openstack years (or 6 months).17:02
*** melwitt is now known as jgwentworth17:02
*** openstackstatus has quit IRC17:05
*** openstack has joined #openstack-cinder17:09
*** ChanServ sets mode: +o openstack17:09
*** ntpttr_laptop has joined #openstack-cinder17:17
*** edmondsw has joined #openstack-cinder17:29
*** links has quit IRC17:32
*** edmondsw has quit IRC17:33
*** MVenesio has joined #openstack-cinder17:40
*** armax has quit IRC17:44
openstackgerritMerged openstack/cinder master: Fix tempest test revet_client typo
*** oomichi_afk is now known as oomichi17:44
*** MVenesio has quit IRC17:45
*** ntpttr_laptop has quit IRC17:45
*** ntpttr_laptop has joined #openstack-cinder17:45
openstackgerritJohn Griffith proposed openstack/cinder master: Fix migration 112 to use live_data_migration API
jgriffithsmcginnis jungleboyj show some love:
jgriffithIt's good, I totally P-Promise18:13
jungleboyjjgriffith:  Your studder makes me nervous.  ;-)18:14
smcginnisjgriffith: Haha, looked fine from me, but I haven't had a chance to try it out. I mostly trust you though. :)18:14
jgriffithsmcginnis LOL... is EXACTLY why no one should trust me!!!18:15
jgriffithwell.. you should trust me on 510201!18:15
*** lbragstad has quit IRC18:16
jgriffithYou just shouldn't trust me due to the fact that I created the thing that needed fixing in the first place18:16
* jgriffith should probably just stop talking now18:16
smcginnisBut oh noes, what about all those people that have already deployed that using CD. We need to support that for the next 5 years and force everyone else to suffer because there might be one person out there that deployed it already.18:16
smcginnisToo snarky?18:16
jgriffithsmcginnis screw-em18:17
jgriffithalthough on the bright side, it won't hurt them, it just becomes a noop18:17
jgriffithsmcginnis and you can NEVER be "too snarky"18:17
jgriffithespecially on that particular topic :)18:18
jgriffithI didn't figure out the weird i18n thing though18:18
jgriffithwhich is annoying18:18
smcginnisjgriffith: Which was that?18:18
coffee_catjgriffith: what's the exact issue with that there?18:19
coffee_catjgriffith: just for my education18:19
jgriffithfor some reason if I put the i18n symbol in front of the prettyprint items (cmd/ it raises18:19
jgriffithI put a comment in the code on that patch with the error18:19
jgriffithand a "FIXME", to say "yeah, I know but it doesn't work and I don't know how to fix it right now"18:20
jgriffithmaybe next week18:20
coffee_catis that used anywhere else for those prettyprint tables?18:20
smcginnisOh right, saw that.18:20
jgriffithor next year18:20
coffee_catsmcginnis: that was a +2 of trust :)18:21
smcginnisEh, a +2 of "good enough". ;)18:21
jgriffithhey... hold on now, I worked hard on that :)18:22
coffee_catsmcginnis: I know, I read slow... :)18:22
smcginnisjgriffith: I'm guessing because _() doesn't actually return a string, so pprint is probably like what the hell is this?18:22
coffee_catjgriffith: no one said you didn't :)18:22
smcginnisI think.18:23
smcginnisBut meh.18:23
jgriffithsmcginnis indeed, but Nova made it work and it really bothered me18:23
smcginnisHmm, well we can't have that.18:24
jgriffithcoffee_cat to answer your earlier question "no" we're not using it elsewhere in Cinder, we do use in the client but we don't i18n it there either18:24
jgriffithsmcginnis :)18:24
openstackgerritJon Bernard proposed openstack/cinder-specs master: Update cheesecake promotion specification
*** lbragstad has joined #openstack-cinder18:25
jgriffithThink I'll get lunch before reworking the actual patches I wanted to get done this week :(18:25
*** markstur_ has quit IRC18:28
coffee_catjgriffith: it's weird though that you removed one column and renamed two more and i18 freaked out from that but not the original version...18:28
*** psachin has joined #openstack-cinder18:29
jgriffithcoffee_cat ?18:29
jgriffithNo... that's not why it errors out18:29
coffee_catI didn't fully put together the pieces yet18:30
coffee_catjust tried to follow some of the changes18:30
jgriffithcoffee_cat so line #313 - 31518:30
jgriffith_('Total Needed'),18:31
jgriffithin other words, sending the table headers through the translator18:31
jgriffithI believe I could work around the type error by translating them into a variable18:32
jgriffithbut honestly don't know if it's worth messing with18:32
jgriffithor if it's the "right" thing to do18:32
jgriffithand like I said, it's annoying that it works elsewhere and I don't know what's different18:33
coffee_catbut if you check the base on the review it looks like it was translated18:33
smcginnisI don't think we translate CLI stuff normally.18:33
jgriffithanyway, it's not really worth any more of my CPU cycles18:33
jgriffithsmcginnis that's good enough for me18:33
smcginnisObsess and speculate. It's the only way.18:33
jgriffithI shouldn't have mentioned it here or in the comments :)18:33
jgriffithbecause now it'll be a *thing* for at least 30 minutes18:34
coffee_catjgriffith: fair enough, go get lunch, I can let this go :)18:34
smcginnisOh, it was translated before. Hmm. Oh well.18:34
smcginnisGo eat. That's more important.18:34
coffee_catsmcginnis: yeah, that got me thinking as I would guess that the test should've caught that even if nothing else used this thing18:35
coffee_catsmcginnis: or maybe even that didn't...18:35
jgriffithsmcginnis wait what?18:37
jgriffithno... see so funny thing18:37
jgriffiththere was a bunch of online_data_migration code in there18:37
jgriffithbut it was never called18:37
jgriffithunit tests are "unit tests" and they cheat18:37
jgriffithbeause as per usual we just mock away all the things18:38
smcginnisShocked. :)18:38
jgriffithI know!!  It's hard to believe18:38
coffee_catjgriffith: yeah, that was my guess18:40
coffee_catjgriffith: sorry to keep you from eating because of this, my bad18:40
coffee_catas for the tests, sadness...18:40
smcginniscoffee_cat: He got donuts the other day. He has enough calories to survive for awhile. ;)18:40
coffee_catcoffee_cat: haha, you're right!18:41
coffee_catcoffee_cat: let me think then how we could increase jgriffith's heart rate to burn those extra calories :)18:42
coffee_catI shouldn't drink more coffee I guess, lol18:42
smcginnis_shouldn't_ drink more coffee? Are you OK?18:43
coffee_catsmcginnis: well, as I'm talking to myself in public now I guess I'm not... :)18:47
jungleboyjA healthy person shouldn't say they shouldn't drink more coffee.  ;-)18:47
*** savvas has joined #openstack-cinder18:48
openstackgerritDhinesh Balasubramaniam proposed openstack/cinder master: Hedvig Cinder driver implementation
openstackgerritDhinesh Balasubramaniam proposed openstack/cinder master: Review comments fixed
hemna_anyone see this trying to stand up cinder rbd driver ?19:05
*** chlong has joined #openstack-cinder19:06
openstackgerritDhinesh Balasubramaniam proposed openstack/cinder master: Hedvig Cinder driver implementation
*** mchlumsky has quit IRC19:12
*** mchlumsky has joined #openstack-cinder19:14
*** edmondsw has joined #openstack-cinder19:17
*** mchlumsky has quit IRC19:20
*** edmondsw has quit IRC19:22
*** wes_dillingham has quit IRC19:24
*** ChubYann has joined #openstack-cinder19:26
hemna_fwiw my ceph keyring file was empty19:26
hemna_that fixed it19:26
hemna_very bad error for what the problem was19:26
smcginnishemna_: Wow, not even a clue to point in the general right area to troubleshoot.19:28
openstackgerritJon Bernard proposed openstack/cinder master: RBD: support driver-assisted volume migration
smcginnisjgriffith, hemna_: Would be great to see if this gets rid of our LVM issues:
jungleboyjsmcginnis:  You want me to just merge it?19:30
smcginnisjungleboyj: That's good to me.19:30
smcginnisWe can always revert or try something in addition to this if it ends up not fixing the job failures.19:30
jungleboyjIt isn't going to make anything worse.  You reached out earlier.19:30
*** tesseract has quit IRC19:31
jungleboyjI just wasn't sure after the earlier discussion if there was more you wanted to do, but I am good starting here.19:31
smcginnisThings won't suck any more than they do now. ;)19:31
*** hemna_ has quit IRC19:31
smcginnisI'm hoping this clears up the failures. No immediate plans to do more unless we find the need to do so.19:31
smcginnisDouble whammy! :)19:32
jungleboyjHe he.19:33
jungleboyjProblem is John is being the strong silent type.19:33
smcginnisHis mouth is too full with donuts for lunch.19:33
jungleboyjHe he.19:34
jgriffithNo donuts today!!  I'm having a Kale, BlueBerry, Quiona and goat cheese salad19:34
jungleboyjOk, stupid question, but is there a way to get devstack to update the install requirements for me?19:35
* jungleboyj make a fancy oooooo sound19:35
smcginnisjgriffith: We're not buying it.19:35
jgriffithsmcginnis dang19:35
jgriffiththey're on to me19:35
jgriffithjungleboyj you mean cinders install requirements?19:35
jgriffithJust by overriding them in the checkout as far as I know19:35
jungleboyjYeah.  When I run devstack right now it keeps failing because I have old versions installed and I am having to install updates by hand which doesn't seem right.19:36 maybe?19:36
jungleboyjTried that.19:36
smcginnispip or apt requirements?19:37
jgriffithjungleboyj by hand?  You should just use the requirements file manually19:38
jungleboyjjgriffith: Teach me oh wise one.19:38
jgriffith"sudo pip install -r ./requirements.txt"19:38
smcginnisIt seems to get in weird states sometimes when packages are already installed.19:39
smcginnisIf it's failing on a specific one, manually remove the package, then run it again.19:39
coffee_catsmcginnis: doesn't --upgrade work for those?19:39
jungleboyjDear lord.  I didn't know I could just run the file like that.  Thank you.19:39
smcginniscoffee_cat: If it's actually upgrading.19:39
coffee_catsmcginnis: good point19:40
jgriffithSo... I need some git foo for my least favorite thing that I screw up EVERY time I try it19:41
jgriffithI have patch A in flight19:41
jgriffithI then created patch-B19:41
jgriffithIt's no proposed as well19:41
smcginnisMake B dependent of A?19:41
jgriffithMake A dep on B19:42
jgriffithI cerry picked B into A, that's fine19:42
jgriffithwill it now be dep on B though when I push it to gerrit?19:42
smcginnisReverse the dependencies?19:42
coffee_catcherry picked B in A?19:42
coffee_catyou need to cherry-pick A on top of B if you want them in a chain19:43
coffee_catwith A depending on B19:43
smcginnisBut if B is already on A you'll need to take a step back.19:43
jgriffithluckily it is not19:43
coffee_catB is proposed already so it's just a download with checkout19:44
jgriffithright... but!!! :)19:44
coffee_cator if B would be on A then a reset --hard19:44
smcginnisOK, checkout B's branch, git cherry-pick $A_HASH, clean up if needed, git review.19:44
jgriffithis the target here19:44
jgriffiththe trick is that has Service object changes that are no longer needed/good19:45
smcginnisJust amend that one with the second commit changes and don't bother?19:45
jgriffithI want the changes in
coffee_catsmcginnis: no, don't amend please19:45
jgriffithcoffee_cat no?  I use amend all the time19:45
coffee_catjgriffith: you can soft reset B and remove the files from the commit that contain those changes19:46
coffee_catjgriffith: then cherry-pick A on top of it19:46
coffee_catjgriffith: I thought smcginnis means making one patch out of the two19:47
coffee_catoh ok, I guess amend can work too19:47
smcginnisIf the second one is just fixing up things in the first one, then easier just merging them into one and not bothering with dependencies.19:47
jgriffithYou people are all crazy :)19:47
coffee_catsmcginnis: no, those patches are two separate things, don't merge them please19:48
* jgriffith just makes a new patch19:48
coffee_catjgriffith: why?19:48
smcginnisI'll stop talking. It's probably not helping that coffee_cat and I have different approaches.19:48
coffee_catjgriffith: why would you make a new patch?19:48
jgriffithcoffee_cat why are you crazy?  I dunno, could be something in your water, perhaps genetic, maybe just enviromental impacts... who knows19:48
openstackgerritMerged openstack/cinder master: Update BlockBox
coffee_catcoffee_cat: so you actually think I'm crazy19:49
coffee_catjgriffith: ^^19:49
jgriffithNo, I'm just messing with you19:49
coffee_catI log off for today19:49
jgriffithand actually if you scroll back you'll notice I said "YOU ALL"19:49
coffee_catjgriffith: but I can fix your git issue without creating a new patch19:49
*** markstur has joined #openstack-cinder19:50
coffee_catjgriffith: well, that includes me too :)19:50
jgriffithI'm just trying to be "inclusive"19:50
coffee_catoh how nice of you :)19:50
jgriffithok... so; what you're suggesting is that I checkout
jgriffithand I then cherry pick on top of it19:51
*** savvas has quit IRC19:51
jgriffithand then just clean up the disaster that is merge-conflicts that occur :)19:51
coffee_catand as you have the service uuid changes in the first one and that's where you want it now that I think of it it should work19:51
jgriffithok... I'll give it a go, what have I got to loose :)19:52
coffee_catgit might complains a bit when you cherry-pick you just fix it as you go19:52
coffee_catit shouldn't have that many disasters19:52
*** eharney has quit IRC19:52
*** arnewiebalck_ has joined #openstack-cinder19:54
*** markstur has quit IRC19:54
coffee_caton the other hand I hate that you do a simple rebase on a Nova patch and the test suddenly starts to complain that you're using the DB where you shouldn't...19:54
coffee_catis this really a Friday???19:55
jgriffithAHHHHH!!!  My changes, my beautiful changes they're all gone!!!!19:55
jgriffithJust kidding19:55
jgriffithseems like that worked :)19:55
jgriffiththanks coffee_cat19:55
coffee_catjgriffith: anytime :)19:55
jgriffithone thing missing though19:56
jgriffithif I were to do a "git review" as it is right now it's going to submit it as the original patch as an update no?19:56
jgriffithWhat I mean by original is if I currently do a git status, I'm on bug/1721837 branch stll19:57
coffee_catit will link the two together as the ChangeId's are the same19:57
jgriffiththat's no bueno19:57
jgriffithok, we'll find out in a bit :)19:57
coffee_catI mean when you push the changes19:57
smcginnisYeah, just don't get rid of the change ID in the commit.19:57
*** jwcroppe has quit IRC19:57
coffee_cator that's what's supposed to happen, but I don't use git review so it might be "smarter" than all of us... :)19:58
*** jwcroppe has joined #openstack-cinder19:58
*** jwcroppe_ has joined #openstack-cinder20:00
*** jwcroppe has quit IRC20:02
*** john5223_ has joined #openstack-cinder20:02
*** e0ne has quit IRC20:05
openstackgerritJohn Griffith proposed openstack/cinder master: Add shared_targets and backend_id to volumes
jgriffithexcellent!  Thanks for the coaching coffee_cat20:08
coffee_catjgriffith: my pleasure, happy it worked :)20:09
coffee_catjgriffith: and never listen to smcginnis ;)20:09
*** ChubYann has quit IRC20:10
*** gouthamr has quit IRC20:21
*** catinthe_ has quit IRC20:28
*** spectr has joined #openstack-cinder20:34
*** spectr has quit IRC20:36
*** david-lyle has quit IRC20:38
*** e0ne has joined #openstack-cinder20:39
openstackgerritMerged openstack/cinder master: Add ploop to parallels naming conversion
*** crushil has quit IRC20:48
*** ChubYann has joined #openstack-cinder20:50
*** markstur has joined #openstack-cinder20:51
*** dikonoor has quit IRC20:52
*** arnewiebalck_ has quit IRC20:57
*** e0ne has quit IRC20:57
*** abishop has quit IRC20:58
openstackgerritMerged openstack/cinder master: Add retries to LVM logical volume activation
*** savvas has joined #openstack-cinder21:02
*** jwcroppe_ has quit IRC21:03
*** edmondsw has joined #openstack-cinder21:05
*** jwcroppe has joined #openstack-cinder21:05
*** savvas has quit IRC21:06
*** edmondsw has quit IRC21:10
*** dustins has quit IRC21:14
*** crushil has joined #openstack-cinder21:16
*** david-lyle has joined #openstack-cinder21:19
*** xyang1 has quit IRC21:20
*** felipemonteiro__ has quit IRC21:21
*** wes_dillingham has joined #openstack-cinder21:23
*** wes_dillingham has quit IRC21:24
*** ronlund is now known as mriedem21:29
*** dklyle has joined #openstack-cinder21:39
*** tpsilva has quit IRC21:39
*** david-lyle has quit IRC21:41
*** dklyle has quit IRC21:41
*** dklyle has joined #openstack-cinder21:42
*** lbragstad has quit IRC21:43
*** gnufied has quit IRC21:46
*** wes_dillingham has joined #openstack-cinder21:51
*** dklyle has quit IRC21:52
*** wes_dillingham has quit IRC21:52
*** wes_dillingham has joined #openstack-cinder21:53
*** wes_dillingham has quit IRC21:53
*** wes_dillingham has joined #openstack-cinder21:54
*** wes_dillingham has quit IRC21:54
*** wes_dillingham has joined #openstack-cinder21:55
*** wes_dillingham has quit IRC21:56
*** wes_dillingham has joined #openstack-cinder21:57
*** wes_dillingham has quit IRC21:57
*** jwcroppe has quit IRC21:58
*** jwcroppe has joined #openstack-cinder21:59
*** Apoorva_ has joined #openstack-cinder22:00
*** Apoorva has quit IRC22:01
*** ntpttr_laptop has quit IRC22:02
*** Apoorva_ has quit IRC22:02
*** Apoorva has joined #openstack-cinder22:02
*** Apoorva has quit IRC22:03
*** jwcroppe has quit IRC22:03
*** Apoorva has joined #openstack-cinder22:03
*** jgwentworth is now known as melwitt22:08
*** chlong has quit IRC22:14
*** armax has joined #openstack-cinder22:23
*** gouthamr has joined #openstack-cinder22:24
*** armax has quit IRC22:28
*** erlon has quit IRC22:30
*** shvepsy_ has joined #openstack-cinder22:31
*** armax has joined #openstack-cinder22:32
*** armax has quit IRC22:33
*** shvepsy has quit IRC22:33
*** crushil has quit IRC22:41
jgriffithI'll listen to smcginnis before I listen to the voices inside my head :)22:43
jgriffithwell.. depending on what they say22:44
jgriffithbut usually22:44
*** markstur has quit IRC22:49
*** ntpttr_laptop has joined #openstack-cinder22:50
*** edmondsw has joined #openstack-cinder22:53
*** edmondsw has quit IRC22:58
jgriffithanybody still around?23:02
*** crushil has joined #openstack-cinder23:05
*** gouthamr has quit IRC23:10
mtreinishjgriffith: I am here23:11
jgriffithmtreinish hey... so I actually was hoping to ping you earlier23:11
* mtreinish wasn't expecting that23:11
jgriffithI forget :)23:11
jgriffithbut anyway :)23:12
jgriffithI was going to see if there was anybody with a running devstack install up (with cinder)23:12
jgriffithbecause I just noticed I can't do something simple that I thought I should be able to do23:12
jgriffithif you `mysql cinder`23:12
jgriffiththen do `select binary from services;`23:12
jgriffithit results in a syntax error, which is completely odd to me23:13
jgriffithonly thing I wondered is if "binary" is some sort of mysql keyword or something23:13
mtreinishjgriffith: yeah it's a type:
jgriffithAhh... yes, well how about that23:14
jgriffithwell that's inconvenient23:15
*** Apoorva has quit IRC23:16
*** Apoorva has joined #openstack-cinder23:17
*** ganso has quit IRC23:18
jgriffithso I wonder how I specify that I actually want to use a column name there23:18
jgriffithOh :)23:19
jgriffith`select services.binar from services`23:19
jgriffithmtreinish thanks for the help :)23:20
jgriffithmriedem if you get a chance
jgriffithmriedem I'm updating to build off of that as well23:22
jgriffithmtreinish oh.. I remember, stestr question23:23
jgriffithbut I need to run.... I'll hit you up later23:23
*** crushil has quit IRC23:24
*** jgriffith is now known as jgriffith_23:24
*** crushil has joined #openstack-cinder23:38
*** david-lyle has joined #openstack-cinder23:42
*** harlowja has quit IRC23:45
*** markstur has joined #openstack-cinder23:45
*** markstur has quit IRC23:49
*** ntpttr_laptop has quit IRC23:50
*** markvoelker has quit IRC23:54

Generated by 2.15.3 by Marius Gedminas - find it at!