| *** rcernin has joined #openstack-glance | 00:03 | |
| *** hugokuo has quit IRC | 00:11 | |
| *** hugokuo has joined #openstack-glance | 00:13 | |
| *** itlinux has quit IRC | 00:17 | |
| *** gyee has quit IRC | 01:08 | |
| *** btully has joined #openstack-glance | 01:16 | |
| *** edmondsw has joined #openstack-glance | 01:17 | |
| *** btully has quit IRC | 01:21 | |
| *** edmondsw has quit IRC | 01:21 | |
| *** markvoelker has joined #openstack-glance | 01:43 | |
| *** markvoelker_ has quit IRC | 01:44 | |
| *** markvoelker_ has joined #openstack-glance | 01:46 | |
| *** markvoelker has quit IRC | 01:48 | |
| *** openstack has joined #openstack-glance | 02:47 | |
| *** ChanServ sets mode: +o openstack | 02:47 | |
| *** edmondsw has joined #openstack-glance | 03:05 | |
| *** edmondsw has quit IRC | 03:09 | |
| *** Sukhdev has joined #openstack-glance | 03:27 | |
| openstackgerrit | wangxiyuan proposed openstack/python-glanceclient master: Skip quote '=' for token header https://review.openstack.org/585937 | 03:34 |
|---|---|---|
| openstackgerrit | wangxiyuan proposed openstack/python-glanceclient master: Skip quote '=' for token header https://review.openstack.org/585937 | 03:37 |
| wxy | rosmaita: jokke_ smcginnis : really sorry that I missed a case here: https://review.openstack.org/585937 can it be merged quick before release the client? Thanks | 03:37 |
| wxy | i.e a case that if the token is encoded by base64, the token will be end with some '=' chars | 03:39 |
| openstackgerrit | wangxiyuan proposed openstack/python-glanceclient master: Skip quote '=' for token header https://review.openstack.org/585937 | 03:41 |
| *** udesale has joined #openstack-glance | 03:48 | |
| *** dalgaaf has quit IRC | 03:49 | |
| *** dalgaaf has joined #openstack-glance | 03:50 | |
| *** itlinux has joined #openstack-glance | 03:58 | |
| *** itlinux has quit IRC | 04:11 | |
| *** rosmaita has quit IRC | 04:12 | |
| *** tshefi has quit IRC | 04:35 | |
| *** Sukhdev has quit IRC | 04:44 | |
| *** Sukhdev has joined #openstack-glance | 04:47 | |
| *** edmondsw has joined #openstack-glance | 04:53 | |
| *** btully has joined #openstack-glance | 04:54 | |
| *** btully has quit IRC | 04:58 | |
| *** edmondsw has quit IRC | 04:58 | |
| *** Tahvok has quit IRC | 05:06 | |
| *** Tahvok has joined #openstack-glance | 05:10 | |
| *** abhishekk has joined #openstack-glance | 05:17 | |
| *** ratailor has joined #openstack-glance | 05:30 | |
| *** Sukhdev has quit IRC | 05:31 | |
| *** ratailor_ has joined #openstack-glance | 05:47 | |
| *** links has joined #openstack-glance | 05:48 | |
| *** ratailor has quit IRC | 05:50 | |
| *** Luzi has joined #openstack-glance | 06:04 | |
| *** udesale has quit IRC | 06:04 | |
| *** udesale has joined #openstack-glance | 06:04 | |
| *** edmondsw has joined #openstack-glance | 06:41 | |
| *** ratailor_ has quit IRC | 06:43 | |
| *** ratailor_ has joined #openstack-glance | 06:44 | |
| *** edmondsw has quit IRC | 06:47 | |
| *** hoonetorg has quit IRC | 06:57 | |
| *** hoonetorg has joined #openstack-glance | 07:14 | |
| *** pcaruana has joined #openstack-glance | 07:25 | |
| *** AlexeyAbashkin has joined #openstack-glance | 07:33 | |
| *** tshefi has joined #openstack-glance | 07:43 | |
| *** dtantsur|afk is now known as dtantsur | 07:46 | |
| *** alexchadin has joined #openstack-glance | 07:48 | |
| *** Guest7752 has quit IRC | 07:52 | |
| *** pdr_ has joined #openstack-glance | 07:53 | |
| *** alexchad_ has joined #openstack-glance | 08:05 | |
| *** alexchadin has quit IRC | 08:07 | |
| *** edmondsw has joined #openstack-glance | 08:30 | |
| *** edmondsw has quit IRC | 08:34 | |
| *** tesseract has joined #openstack-glance | 08:42 | |
| *** udesale_ has joined #openstack-glance | 08:44 | |
| *** udesale has quit IRC | 08:46 | |
| *** alexchadin has joined #openstack-glance | 08:59 | |
| *** alexchad_ has quit IRC | 09:01 | |
| openstackgerrit | OpenStack Release Bot proposed openstack/python-glanceclient master: Update reno for stable/rocky https://review.openstack.org/586011 | 09:01 |
| *** d0ugal has joined #openstack-glance | 09:21 | |
| *** ratailor_ has quit IRC | 09:23 | |
| openstackgerrit | Michal Arbet proposed openstack/glance master: Fix RuntimeError when running under py37 - StopIteration https://review.openstack.org/586034 | 09:50 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance-specs master: Modify operator image workflow specs https://review.openstack.org/580568 | 09:51 |
| jokke_ | wxy: no worries ... lets make sure we have bug for it and we can always backport | 09:52 |
| *** d0ugal has quit IRC | 09:52 | |
| *** udesale__ has joined #openstack-glance | 10:08 | |
| *** alexchadin has quit IRC | 10:08 | |
| *** udesale_ has quit IRC | 10:10 | |
| *** edmondsw has joined #openstack-glance | 10:18 | |
| *** alexchadin has joined #openstack-glance | 10:18 | |
| *** btully has joined #openstack-glance | 10:20 | |
| *** abhishekk has quit IRC | 10:20 | |
| *** edmondsw has quit IRC | 10:23 | |
| *** btully has quit IRC | 10:24 | |
| *** alexchadin has quit IRC | 10:29 | |
| *** d0ugal has joined #openstack-glance | 10:32 | |
| *** d0ugal has quit IRC | 10:34 | |
| *** alexchadin has joined #openstack-glance | 10:38 | |
| *** edmondsw has joined #openstack-glance | 10:49 | |
| *** alexchad_ has joined #openstack-glance | 10:49 | |
| *** alexchadin has quit IRC | 10:51 | |
| *** linkmark has joined #openstack-glance | 10:58 | |
| *** rcernin has quit IRC | 11:04 | |
| *** abhishekk-mbl has joined #openstack-glance | 11:09 | |
| *** udesale__ has quit IRC | 11:25 | |
| *** rosmaita has joined #openstack-glance | 11:26 | |
| openstackgerrit | Michal Arbet proposed openstack/glance master: Fix RuntimeError when running under py37 - StopIteration https://review.openstack.org/586050 | 11:31 |
| smcginnis | wxy, jokke_: We already released and created the stable/rocky branch, but I think we're still in good shape since today is the deadline. | 12:08 |
| smcginnis | wxy, jokke_: We should be able to get that backported to stable/rocky and do one more release today to still get it as part of the official rocky deliverables. | 12:08 |
| *** MattMan has quit IRC | 12:08 | |
| smcginnis | After today we would just have to wait and do it as soon as the freeze is lifted, but we've got some time. | 12:09 |
| *** MattMan has joined #openstack-glance | 12:09 | |
| rosmaita | smcginnis i'll look at the patch, but i think we need a more thorough fix for this | 12:16 |
| rosmaita | but i see your point about getting it into the rocky deliverables | 12:17 |
| *** abhishekk has joined #openstack-glance | 12:21 | |
| smcginnis | rosmaita: I had appoved that. Do you want me to remove my +W and let you have some time with it first? | 12:23 |
| rosmaita | smcginnis just saw that | 12:23 |
| rosmaita | no, i think it's ok | 12:23 |
| smcginnis | OK, great. :) | 12:24 |
| rosmaita | at least it will handle the most common cases | 12:24 |
| rosmaita | i think we should get it in | 12:24 |
| wxy | smcginnis: rosmaita: ++ yeah, tbh, my patch is just a quick fix for some Public Cloud usage (They still use old token format). | 12:24 |
| rosmaita | so thanks for approving | 12:24 |
| rosmaita | wxy hello | 12:24 |
| rosmaita | are there any other chars we should worry about? | 12:24 |
| rosmaita | :) | 12:25 |
| wxy | since I don't want to break RFC, I just skip some chars and still let the header be ascii encode. | 12:25 |
| abhishekk | rosmaita, thank you for detail explanation on hidden images patch | 12:25 |
| rosmaita | wxy: part of the problem is that some of the RFCs are a bit vague | 12:25 |
| rosmaita | i think i went overboard in my fix with urlencoding everything | 12:26 |
| rosmaita | only need to urlencode if any char in the string is ord(char) > 128 | 12:26 |
| rosmaita | so technically, + is ok | 12:26 |
| rosmaita | i think | 12:27 |
| rosmaita | not sure how to handle it though, because how is a consumer going to know if a string is urlencoded or not? | 12:27 |
| rosmaita | safer to assume that it always is | 12:28 |
| rosmaita | i don;'t know if there is a library out there that handles this correctly, i could not find one | 12:28 |
| rosmaita | anyway, at least we can get this fixed for the auth headers that might contain the old token format | 12:29 |
| rosmaita | abhishekk np, but it distracted me from completing my review of your patch! | 12:29 |
| abhishekk | ohh | 12:32 |
| smcginnis | Surprisingly strong opinions on that os_hidden naming. | 12:35 |
| rosmaita | yes indeed | 12:36 |
| abhishekk | yeah, and it might has caused more confusion with my reply :( | 12:36 |
| *** openstackgerrit has quit IRC | 12:36 | |
| rosmaita | abhishekk: not really, you pretty much said what was said last week at the glance meeting | 12:38 |
| rosmaita | smcginnis got a reno question for you ... i think we need a relnote for new glanceclient release ... do I do that in master and backport, or directly in stable/rocky ? | 12:39 |
| rosmaita | i'm pretty sure the last time i did this i did it wrong :( | 12:40 |
| smcginnis | rosmaita: Yeah, that's where things get weird with reno. | 12:40 |
| rosmaita | is this a dhellmann question? | 12:40 |
| smcginnis | rosmaita: If you do it in master then backport to stable/rocky it will end up in the stein release notes. | 12:40 |
| smcginnis | rosmaita: So even though it would cause a difference between the two, it would be best to propose that directly to the stable/rocky branch. | 12:41 |
| rosmaita | i agree, it's just going to say that a followup was done on the bug listed as fixed | 12:41 |
| rosmaita | in previous release | 12:42 |
| rosmaita | no need for stein to say that | 12:42 |
| rosmaita | ok | 12:42 |
| smcginnis | rosmaita: Yeah, that sounds like a good idea. | 12:42 |
| abhishekk | rosmaita, I think wxy has fixed it against the same bug which was included in release note | 12:42 |
| rosmaita | abhishekk yes, but i think the new release needs to say something about why there's a new release | 12:42 |
| abhishekk | got it | 12:42 |
| *** links has quit IRC | 12:46 | |
| *** Sukhdev has joined #openstack-glance | 12:56 | |
| rosmaita | not sure what i did wrong with https://review.openstack.org/#/c/586201/1 ... i checked out stable/rocky, created a feature branch from that, but gerrit is showing branch as master | 12:57 |
| abhishekk | Also we did not get notification here after you pushed a patch | 12:59 |
| rosmaita | yeah, i figured that;s because it was proposed against a stable branch | 13:00 |
| rosmaita | weird | 13:00 |
| abhishekk | yeah | 13:00 |
| jokke_ | rosmaita: that's because the stable branching utility patches by the bot has not been merged yet | 13:00 |
| jokke_ | one of them sets the default review branch to stable | 13:00 |
| rosmaita | ahhhh ... a bit of magic i was taking for granted | 13:00 |
| *** Sukhdev has quit IRC | 13:01 | |
| jokke_ | just +2W'd them three | 13:01 |
| jokke_ | they come as granted as long as we just merge the patches from the bot :D | 13:02 |
| rosmaita | guess i should abandon this patch and re-push my change after they land? | 13:02 |
| jokke_ | or you can just go git review stable/rocky | 13:02 |
| rosmaita | same review? | 13:03 |
| jokke_ | you might need to abandon it from gerrit first but you should be able to use the same commit | 13:03 |
| rosmaita | ok | 13:03 |
| *** alexchad_ has quit IRC | 13:09 | |
| rosmaita | that did it | 13:10 |
| jokke_ | yup | 13:10 |
| *** jistr is now known as jistr|mtg | 13:29 | |
| *** r-daneel has joined #openstack-glance | 13:40 | |
| *** alexchadin has joined #openstack-glance | 13:44 | |
| smcginnis | I see we didn't get the patch notification because apparently the gerrit bot got kicked out as part of freenode trying to stop all this spam. | 13:45 |
| smcginnis | Just fyi | 13:45 |
| rosmaita | ty, i was wondering | 13:45 |
| rosmaita | 'freenodegate' is getting to be a PITA | 13:45 |
| jokke_ | Courtesy meeting reminder on #openstack-meeting-4: flwang1, jokke_, mclaren, mfedosin, nikhil_k, rosmaita, GB21, bpoulos, harshs, abhishekk, dshakhray, kairat, alex_bash, smcginnis, arcolife, Brin, Roamer` | 14:00 |
| *** McClymontS has joined #openstack-glance | 14:01 | |
| *** weshay_PTO has quit IRC | 14:07 | |
| *** weshay has joined #openstack-glance | 14:08 | |
| *** Luzi has quit IRC | 14:09 | |
| *** abhishekk-mbl has quit IRC | 14:38 | |
| *** abhishekk has quit IRC | 14:43 | |
| *** McClymontS has quit IRC | 14:43 | |
| rosmaita | jokke_ could use a +2A on the glanceclient wxy backport: https://review.openstack.org/#/c/586232/ | 14:48 |
| rosmaita | actually, it has not finished gate yet, sorry | 14:49 |
| rosmaita | jokke_ https://review.openstack.org/#/c/586232/ is ready, though | 14:49 |
| jokke_ | yeah, been keeping eye on that ... will do as soon as mr. Zuul agrees | 14:49 |
| rosmaita | ok, cool | 14:49 |
| jokke_ | I'll +2W them both at the same time but not gonna merge the reno yet in case we can't get the fix through the check for some reason | 14:50 |
| *** Sukhdev has joined #openstack-glance | 14:50 | |
| jokke_ | shouldn't be problem but you never know | 14:50 |
| rosmaita | jokke_ sounds good, thanks | 14:51 |
| *** imacdonn has quit IRC | 14:51 | |
| *** imacdonn has joined #openstack-glance | 14:52 | |
| *** jistr|mtg is now known as jistr | 14:56 | |
| *** tbarron has quit IRC | 15:01 | |
| *** tbarron has joined #openstack-glance | 15:08 | |
| *** alexchadin has quit IRC | 15:08 | |
| *** Sukhdev has quit IRC | 15:17 | |
| *** tshefi has quit IRC | 15:25 | |
| *** pcaruana has quit IRC | 15:38 | |
| *** btully has joined #openstack-glance | 15:47 | |
| *** btully has quit IRC | 15:52 | |
| *** edmondsw has quit IRC | 15:57 | |
| *** edmondsw has joined #openstack-glance | 15:57 | |
| *** edmondsw has quit IRC | 15:59 | |
| *** gyee has joined #openstack-glance | 15:59 | |
| *** itlinux has joined #openstack-glance | 16:01 | |
| *** AlexeyAbashkin has quit IRC | 16:10 | |
| *** abhishekk-mbl has joined #openstack-glance | 16:16 | |
| *** dtantsur is now known as dtantsur|afk | 16:38 | |
| *** tesseract has quit IRC | 16:39 | |
| jokke_ | rosmaita: gating the client changes now | 16:42 |
| rosmaita | you are following closely! | 16:42 |
| jokke_ | yeah, focusing the hidden images glance change now | 16:44 |
| *** abhishekk has joined #openstack-glance | 16:52 | |
| *** d0ugal has joined #openstack-glance | 16:55 | |
| *** d0ugal has quit IRC | 16:55 | |
| *** d0ugal has joined #openstack-glance | 16:55 | |
| *** r-daneel_ has joined #openstack-glance | 17:10 | |
| *** r-daneel has quit IRC | 17:10 | |
| *** r-daneel_ is now known as r-daneel | 17:10 | |
| *** Sukhdev has joined #openstack-glance | 17:12 | |
| *** r-daneel has quit IRC | 17:14 | |
| *** d0ugal has quit IRC | 17:21 | |
| *** r-daneel has joined #openstack-glance | 17:22 | |
| *** btully has joined #openstack-glance | 17:35 | |
| *** btully has quit IRC | 17:39 | |
| *** mvk_ has quit IRC | 17:47 | |
| *** edmondsw has joined #openstack-glance | 17:48 | |
| rosmaita | jokke_ taking a quick lunch break ... left a comment on hidden images patch, i think there's a problem, but i will continue reviewing soon as i get back | 17:57 |
| *** DuncanT_ has left #openstack-glance | 17:57 | |
| jokke_ | good catch, I missed that and was just gonna +2 it | 18:11 |
| jokke_ | rosmaita: I think you're right with that comment | 18:11 |
| jokke_ | rosmaita: that's on db simple api, can we not to gate the new patch for next six hours and fix that logic error after it has merged? | 18:21 |
| abhishekk | jokke_, rosmaita yes, that change is unreachable (I am really not able to recollect why I have added that :() | 18:23 |
| abhishekk | But after removing that code I can see all tests are passing | 18:23 |
| rosmaita | yeah, i am ok with a follow up fix | 18:24 |
| abhishekk | it just needs this line in db/simple/api.py https://review.openstack.org/#/c/578755/12/glance/db/simple/api.py@233 | 18:24 |
| jokke_ | abhishekk: I don't think simple db is either gated nor supported in production. Thus I was hoping not need to spin that change around | 18:24 |
| rosmaita | i just get obsessed with logic | 18:24 |
| abhishekk | should I put a dependent patch? | 18:24 |
| jokke_ | rosmaita: and that is good | 18:25 |
| abhishekk | yes, really good | 18:25 |
| jokke_ | abhishekk: don't do anything just yet | 18:25 |
| abhishekk | ack | 18:25 |
| jokke_ | so rosmaita if nothing else is found that requires new PS are you willing to get fix for that in follow up (My main concern is how slow check and gate queues are atm.) | 18:26 |
| rosmaita | abhishekk with sqlalchemy, with default = False, will it initialize the entire column, or is that only for new entries? i guess it must initialize since nullable is also false | 18:26 |
| rosmaita | jokke_ yes i am willing to do that | 18:26 |
| jokke_ | cool | 18:27 |
| rosmaita | should have the rest reviewed in the next 20 min | 18:27 |
| rosmaita | yeah, gate is really slow today | 18:27 |
| abhishekk | it will initialize entire column | 18:27 |
| jokke_ | rosmaita: yeah it should initialize the whole db due to the combination of the two | 18:27 |
| rosmaita | abhishekk excellent, just wanted to adk | 18:27 |
| rosmaita | *ask | 18:27 |
| abhishekk | yeah got your point | 18:28 |
| abhishekk | glanceclient both patches merged | 18:31 |
| rosmaita | great, will put up release patch now | 18:32 |
| rosmaita | https://review.openstack.org/586319 | 18:35 |
| rosmaita | jokke_ or abhishekk ^^ please verify the hash | 18:35 |
| abhishekk | rosmaita, looks good | 18:38 |
| rosmaita | ty | 18:38 |
| rosmaita | abhishekk nice work on this hidden images patch , btw | 18:38 |
| abhishekk | thank you, | 18:38 |
| abhishekk | for your thorough review as well | 18:38 |
| rosmaita | ok, i just +2'd https://review.openstack.org/#/c/578755/12 | 19:01 |
| jokke_ | ok, I'll throw it to gate then so we get release ready today | 19:02 |
| abhishekk | Thank you, I will be around till v1 gets merged | 19:03 |
| jokke_ | that might take a while :P | 19:04 |
| abhishekk | incase hide images patch goes merged conflict I will rebase it and do the simple db related change | 19:04 |
| jokke_ | oh yeah there is that | 19:04 |
| jokke_ | it shouldn't | 19:05 |
| jokke_ | but ... | 19:05 |
| rosmaita | yeah, i hope not | 19:05 |
| abhishekk | so will stick around next couple of hours, just in case | 19:05 |
| rosmaita | abhishekk isn't it like past midnight where you are? | 19:05 |
| abhishekk | I am also sure it will not, but multi-store might definitely go in merge conflicts | 19:05 |
| abhishekk | rosmaita, just past 12 here | 19:06 |
| abhishekk | I will try to sleep lit bit extra tomorrow :D | 19:06 |
| rosmaita | it is 15:00 where i am ... maybe i should volunteer to rebase if necessary | 19:06 |
| jokke_ | 20:00 for me ... I might not be much around tomorrow once all this is sorted tonight :P | 19:07 |
| rosmaita | by "maybe" i mean definitely | 19:07 |
| jokke_ | :D | 19:07 |
| abhishekk | haha | 19:07 |
| abhishekk | rosmaita, thank you | 19:08 |
| rosmaita | np, hopefully it won't happen1 | 19:08 |
| abhishekk | yep, going offline now, have a nice time ahead | 19:08 |
| jokke_ | night abhishekk | 19:08 |
| rosmaita | ok, talk to you tomorrow | 19:08 |
| abhishekk | (will be available on mobile though) | 19:09 |
| *** abhishekk has quit IRC | 19:09 | |
| *** pcaruana has joined #openstack-glance | 19:17 | |
| *** r-daneel has quit IRC | 19:21 | |
| *** r-daneel has joined #openstack-glance | 19:22 | |
| *** btully has joined #openstack-glance | 19:24 | |
| *** btully has quit IRC | 19:28 | |
| *** openstackgerrit has joined #openstack-glance | 19:34 | |
| openstackgerrit | Rick Bartra proposed openstack/glance master: Add Policy enforcement for several Metadata Definition delete APIs https://review.openstack.org/584530 | 19:34 |
| openstackgerrit | Brian Rosmaita proposed openstack/glance_store master: Refactor periodic "tips" jobs https://review.openstack.org/586334 | 19:35 |
| *** pcaruana has quit IRC | 19:45 | |
| imacdonn | rosmaita: My "plan C" now is to make a dedicated VM to run the glance service, and install the Queens version of glance there, whilst the rest of the deployment would be Rocky .... other than longevity, do you see any immediate problems with this idea ? | 20:06 |
| rosmaita | thinking | 20:07 |
| imacdonn | Just did a quick prototype, with Pike glance, and Queens "everything else" ... seems to be working | 20:07 |
| rosmaita | rocky will have some new database columns in the images table | 20:10 |
| imacdonn | but the APIs should still work, right? unless I try to consume some new Rocky feature | 20:11 |
| rosmaita | i think so | 20:12 |
| imacdonn | I guess cinder could try to consume some new feature - that wouldn't be within my control | 20:12 |
| jokke_ | Rocky API will not work with queens db | 20:12 |
| jokke_ | but queens API should work just fine with rocky db | 20:12 |
| rosmaita | yeah, it's like a live upgrade, except lasting a long time | 20:13 |
| imacdonn | I'm suggesting not using any rocky bits of glance, so it'd be queens glance, with a queens glance DB ... but rocky keystone, cinder, nova, etc. | 20:13 |
| rosmaita | probably don't want it to last into stein though, that is definitely unsupported | 20:13 |
| rosmaita | ok, i misunderstood Plan C | 20:14 |
| jokke_ | imacdonn: ok, that should work, but why not going to rocky glance? | 20:14 |
| * jokke_ clearly has missed some part of the conversation | 20:14 | |
| rosmaita | imacdonn is the "web reference" blueprint guy | 20:15 |
| imacdonn | jokke_: because I need to register images by URL using the HTTP backend, and you guys took the v1 API away, and didn't implement an equivalent in v2 | 20:15 |
| jokke_ | ahh | 20:15 |
| jokke_ | that makes sense | 20:15 |
| imacdonn | I'm "that guy" :) | 20:15 |
| jokke_ | imacdonn: was it that your users needs to do that or is it admin only adding those http backend images | 20:16 |
| jokke_ | ? | 20:16 |
| imacdonn | admin only | 20:17 |
| imacdonn | It's a private cloud, where images are maintained by IT .. and published on a central web service on the intranet | 20:17 |
| imacdonn | I should say "a collection of private clouds" | 20:17 |
| imacdonn | copying the (rather large) images to glance stores for each cloud would be a huge waste of time, bandwidth and storage | 20:18 |
| imacdonn | they get used only to create cinder volumes, which are then snapshotted and cloned to make boot volumes | 20:18 |
| jokke_ | ok, for future proofing your workflow (btw I really love the fact that you guys are planning already moving to Rocky!!!) if you can afford admin VM you can keep registering the images via exposing the locations | 20:18 |
| jokke_ | imacdonn: makes sense | 20:18 |
| imacdonn | jokke_: see https://bugs.launchpad.net/glance/+bug/1750892 | 20:19 |
| openstack | Launchpad bug 1750892 in Glance "Image remains in queued status after location set via PATCH" [Undecided,Triaged] | 20:19 |
| imacdonn | a) doesn't work b) no way to set checksum | 20:19 |
| jokke_ | imacdonn: yes, at least the point a) needs to be fixed | 20:19 |
| jokke_ | we need to figure out what to do with the checksums in those cases | 20:20 |
| imacdonn | I'd like to understand the reasoning for checksum being read-only better | 20:20 |
| imacdonn | my comment from yesterday: <imacdonn> that seems kinda weird to me ... if I have an image prepared, I may already have a checksum that I know is good .. what if there's some corruption in the transfer to glance? You'd end up with a bad image and a bad checksum and not catch it...... | 20:20 |
| jokke_ | so would I ... like now at this point of time when we have intrduced for example the image conversion it makes more sense, but in historical perspective it really doesn't | 20:21 |
| jokke_ | imacdonn: yeah you would need to compare the checksums which really haven't made any sense | 20:21 |
| rosmaita | you could use the image signature verification, nova understands it | 20:23 |
| imacdonn | OK, so I guess I'm tentatively going with my "plan C" for rocky, and hope we can figure out a proper solution for stein ... I tried to get ahead of this, but I was told that nothing could be done until some policy-related stuff was figured out, so felt I couldn't push it, but feared we'd end up here :/ | 20:33 |
| jokke_ | imacdonn: yes unfortunately the person who was set up for doing the policy work disappeared and we really didn't have the manpower to tackle it in this cycle | 20:34 |
| jokke_ | was hoping we would have had decent solution to fix these issues with the policy refactoring | 20:35 |
| jokke_ | but lets get that bug you linked fixed and provide you at least some way to plan your work ahead | 20:36 |
| imacdonn | that'd be good | 20:40 |
| jokke_ | and by fixed I also mean backported so you can start testing and migrating your workflow in that queens glance you're lanning running | 20:40 |
| jokke_ | planning | 20:41 |
| imacdonn | yeah, if we can get a long-term-sustainable solution that will back-port to Queens, we could start developing new process around that ... I guess we need to decide if the PATCH method is the long-term solution or a stop-gap | 20:42 |
| imacdonn | don't really want to develop new process around a stop-gap | 20:43 |
| jokke_ | imacdonn: yeah, but really can't tell that either before it works on the level you can see how that workflow would work for you | 20:43 |
| jokke_ | and it's clearly a bug | 20:43 |
| imacdonn | ok... so who can take on the bug? I can work on it a bit, but would need guidance from someone who knows the internal workings of glance.... | 20:45 |
| jokke_ | imacdonn: we need to jump on that after about a weeks time | 20:46 |
| jokke_ | we need to get this feature/feature freeze work done by next Thu | 20:46 |
| jokke_ | so next weeks glance weekly meeting is good place to lift that cat on to the table | 20:46 |
| imacdonn | fair enough ... I'll bug you guys again the week after next | 20:46 |
| rosmaita | man, i hate being in the integrated queue | 20:48 |
| jokke_ | yup | 20:48 |
| jokke_ | ~40 zuul minutes left | 20:55 |
| rosmaita | at least abhishek's patch is out of queued status | 20:56 |
| jokke_ | oh yeah that's still queued | 20:57 |
| rosmaita | at least yours is at the head of the queue, should really finish this time | 20:58 |
| *** r-daneel_ has joined #openstack-glance | 20:59 | |
| jokke_ | yeah, that was the one I was looking for 40 zuul mins | 20:59 |
| *** r-daneel has quit IRC | 21:00 | |
| *** r-daneel_ is now known as r-daneel | 21:00 | |
| clarkb | I think the qa team would appreciate help making things run more smoothly :) mriedem has been poking around that recently fwiw | 21:02 |
| jokke_ | clarkb: yeah, I'm sure you guys are on the same boat with us not being the sexy projects in the ecosystem ;) | 21:04 |
| clarkb | thats some of it, I just think the attitude should be more of "lets make the integrated testing work well" rather than "I hate we are tested with this openstack thing" :) | 21:06 |
| rosmaita | integrated testing works great when you're first in the queue :P | 21:07 |
| jokke_ | or like discussed in Dublin when you can have majority of the changes behaving | 21:08 |
| clarkb | ya, unfortunately it isn't the easiest or most rewarding work to sort out why some race between whatever systems tempest is touching causes tests to fail 1% of the time | 21:11 |
| jokke_ | clarkb: indeed and it really hurts only when everyone are super busy | 21:12 |
| *** cfriesen_ has joined #openstack-glance | 21:13 | |
| *** cfriesen_ has left #openstack-glance | 21:13 | |
| *** r-daneel has quit IRC | 21:17 | |
| *** edmondsw has quit IRC | 21:43 | |
| *** edmondsw has joined #openstack-glance | 21:45 | |
| *** itlinux has quit IRC | 21:47 | |
| *** edmondsw has quit IRC | 21:49 | |
| jokke_ | rosmaita: you're gonna love this ... tempest failed | 21:57 |
| rosmaita | arrrgh ... guess it's not because of your patch, though, since it passed for the check | 22:01 |
| rosmaita | Timeout while verifying metadata on server. | 22:03 |
| *** Sukhdev has quit IRC | 22:08 | |
| *** mvk_ has joined #openstack-glance | 22:12 | |
| *** rcernin has joined #openstack-glance | 22:19 | |
| *** itlinux has joined #openstack-glance | 22:41 | |
| *** lbragstad_ has joined #openstack-glance | 23:01 | |
| *** lbragstad has quit IRC | 23:01 | |
| *** knikolla has quit IRC | 23:01 | |
| *** btully has joined #openstack-glance | 23:02 | |
| jokke_ | rosmaita: I need to get some sleep. I've been staring these screens for over 14hrs now | 23:02 |
| jokke_ | I'll put early alarm and if we do not have Rocky-3 I will put the release patch up first thing at the morn | 23:03 |
| *** btully has quit IRC | 23:07 | |
| *** itlinux has quit IRC | 23:16 | |
| *** arvindn05 has joined #openstack-glance | 23:17 | |
| *** arvindn05 has quit IRC | 23:18 | |
| *** abhishekk-mbl has quit IRC | 23:18 | |
| *** itlinux has joined #openstack-glance | 23:43 | |
| *** itlinux has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!