*** btully has joined #openstack-horizon | 00:19 | |
*** btully has quit IRC | 00:24 | |
*** lblanchard has joined #openstack-horizon | 00:25 | |
*** hoangcx has joined #openstack-horizon | 00:41 | |
*** btully has joined #openstack-horizon | 00:56 | |
*** btully has quit IRC | 01:00 | |
*** lblanchard has quit IRC | 01:06 | |
*** btully has joined #openstack-horizon | 01:19 | |
*** btully has quit IRC | 01:23 | |
*** btully has joined #openstack-horizon | 01:24 | |
*** btully has quit IRC | 01:33 | |
*** btully has joined #openstack-horizon | 01:34 | |
*** btully has quit IRC | 01:38 | |
*** btully has joined #openstack-horizon | 01:39 | |
*** rcernin has quit IRC | 01:40 | |
*** rcernin has joined #openstack-horizon | 01:40 | |
*** btully has quit IRC | 01:43 | |
*** btully has joined #openstack-horizon | 01:45 | |
*** btully has quit IRC | 01:49 | |
*** btully has joined #openstack-horizon | 01:50 | |
*** btully has quit IRC | 01:55 | |
*** btully has joined #openstack-horizon | 01:57 | |
*** btully has quit IRC | 02:01 | |
*** btully has joined #openstack-horizon | 02:03 | |
*** Bhujay has joined #openstack-horizon | 02:04 | |
*** btully has quit IRC | 02:08 | |
*** btully has joined #openstack-horizon | 02:09 | |
*** Bhujay has quit IRC | 02:12 | |
*** btully has quit IRC | 02:18 | |
*** btully has joined #openstack-horizon | 02:18 | |
*** btully has quit IRC | 02:23 | |
*** btully has joined #openstack-horizon | 02:30 | |
*** btully has quit IRC | 02:34 | |
*** btully has joined #openstack-horizon | 03:26 | |
*** btully has quit IRC | 03:30 | |
*** btully has joined #openstack-horizon | 03:56 | |
*** btully has quit IRC | 04:08 | |
*** btully has joined #openstack-horizon | 04:09 | |
*** btully has quit IRC | 04:13 | |
*** btully has joined #openstack-horizon | 04:13 | |
*** btully has quit IRC | 04:18 | |
*** btully has joined #openstack-horizon | 04:19 | |
*** btully has quit IRC | 04:37 | |
*** btully has joined #openstack-horizon | 04:45 | |
*** btully has quit IRC | 04:49 | |
*** btully has joined #openstack-horizon | 04:53 | |
*** btully has quit IRC | 04:57 | |
*** btully has joined #openstack-horizon | 04:57 | |
*** Bhujay has joined #openstack-horizon | 04:58 | |
*** btully has quit IRC | 05:06 | |
*** btully has joined #openstack-horizon | 05:09 | |
*** btully has quit IRC | 05:13 | |
*** btully has joined #openstack-horizon | 05:13 | |
*** btully has quit IRC | 05:17 | |
*** btully has joined #openstack-horizon | 05:26 | |
*** btully has quit IRC | 05:34 | |
*** dave-mccowan has joined #openstack-horizon | 05:36 | |
*** jtomasek has joined #openstack-horizon | 05:37 | |
*** btully has joined #openstack-horizon | 05:43 | |
*** dayou has quit IRC | 05:43 | |
*** dayou has joined #openstack-horizon | 05:44 | |
*** btully has quit IRC | 05:51 | |
*** btully has joined #openstack-horizon | 05:51 | |
*** Bhujay has quit IRC | 05:53 | |
*** Bhujay has joined #openstack-horizon | 05:59 | |
*** btully has quit IRC | 05:59 | |
*** pcaruana has joined #openstack-horizon | 06:05 | |
*** btully has joined #openstack-horizon | 06:07 | |
*** btully has quit IRC | 06:12 | |
*** e0ne has joined #openstack-horizon | 06:23 | |
*** ChanServ sets mode: +o e0ne | 06:23 | |
*** btully has joined #openstack-horizon | 06:24 | |
*** btully has quit IRC | 06:29 | |
*** btully has joined #openstack-horizon | 06:36 | |
*** belmoreira has joined #openstack-horizon | 06:36 | |
*** gvrangan has joined #openstack-horizon | 06:38 | |
*** e0ne_ has joined #openstack-horizon | 06:41 | |
*** ChanServ sets mode: +o e0ne_ | 06:41 | |
*** e0ne has quit IRC | 06:44 | |
*** btully has quit IRC | 06:44 | |
*** btully has joined #openstack-horizon | 06:44 | |
*** belmoreira has quit IRC | 06:45 | |
*** belmoreira has joined #openstack-horizon | 06:47 | |
*** btully has quit IRC | 06:49 | |
*** btully has joined #openstack-horizon | 06:52 | |
*** btully has quit IRC | 06:57 | |
*** btully has joined #openstack-horizon | 07:00 | |
*** rdopiera has joined #openstack-horizon | 07:04 | |
*** rcernin has quit IRC | 07:06 | |
*** btully has quit IRC | 07:09 | |
*** btully has joined #openstack-horizon | 07:09 | |
*** btully has quit IRC | 07:14 | |
*** btully has joined #openstack-horizon | 07:17 | |
*** ircuser-1 has joined #openstack-horizon | 07:20 | |
*** btully has quit IRC | 07:22 | |
*** btully has joined #openstack-horizon | 07:25 | |
*** btully has quit IRC | 07:34 | |
*** btully has joined #openstack-horizon | 07:34 | |
*** btully has quit IRC | 07:42 | |
*** gvrangan has quit IRC | 07:42 | |
*** tosky has joined #openstack-horizon | 07:51 | |
*** jpich has joined #openstack-horizon | 07:56 | |
*** btully has joined #openstack-horizon | 07:58 | |
*** btully has quit IRC | 08:02 | |
*** btully has joined #openstack-horizon | 08:05 | |
*** btully has quit IRC | 08:10 | |
*** btully has joined #openstack-horizon | 08:14 | |
*** openstackgerrit has quit IRC | 08:22 | |
*** btully has quit IRC | 08:22 | |
e0ne_ | rdopiera: hi Radomir. are you around? | 08:25 |
---|---|---|
*** btully has joined #openstack-horizon | 08:26 | |
rdopiera | e0ne_: I am, hello | 08:32 |
*** btully has quit IRC | 08:34 | |
rdopiera | e0ne_: what about CVEs for jquery? | 08:35 |
*** btully has joined #openstack-horizon | 08:35 | |
e0ne_ | rdopiera: it's a good question | 08:37 |
e0ne_ | rdopiera: I'm all to fix horizon to use newer JS libs | 08:37 |
rdopiera | there has been a whole bunch of XSS-es for older jquery | 08:38 |
e0ne_ | rdopiera: but let's fix our gates first | 08:38 |
rdopiera | ok, so unblock the gate first, and then miograte to newer jquery | 08:38 |
*** Bhujay has quit IRC | 08:38 | |
rdopiera | I thought that the jquer-migrate was doing that for us | 08:38 |
e0ne_ | rdopiera: +2 | 08:38 |
e0ne_ | rdopiera: I'm looking into the maintainers section here: https://pypi.org/project/XStatic-jQuery/#history | 08:39 |
*** btully has quit IRC | 08:39 | |
e0ne_ | rdopiera: do you know Thomas Waldmann? | 08:39 |
e0ne_ | all xstatic packages from the https://github.com/xstatic-py were updated last week | 08:39 |
rdopiera | yes, we worked on MoinMoin together | 08:39 |
adriant | e0ne_, rdopiera: after you fix the gate, would you be willing to take a look at: https://bugs.launchpad.net/horizon/+bug/1793411 ? | 08:39 |
openstack | Launchpad bug 1793411 in OpenStack Dashboard (Horizon) "Dashboard memory leaks" [Undecided,New] | 08:39 |
rdopiera | yes, in fact he messaged me about it | 08:39 |
adriant | We're hitting that really badly in production | 08:40 |
e0ne_ | adriant: :( | 08:40 |
adriant | we used to solve it with uwsgi max requests setting, but that has started killing workers too quickly and causing weird false logouts | 08:40 |
rdopiera | e0ne_: I plan to look at it, but at the moment I still have some blockers for osp14 and some regression testing | 08:41 |
adriant | e0ne_: but the uwsgi fix we had was only a fix over the underlying problem :( | 08:42 |
*** btully has joined #openstack-horizon | 08:42 | |
e0ne_ | adriant: I didn't take a look on this issue yet | 08:43 |
rdopiera | perhaps it would make sense to replace memoize with a MRU cache | 08:44 |
rdopiera | then we have a guarantee it won't use more memory than what we set | 08:44 |
rdopiera | of course then there is the problem of tweaking that setting | 08:44 |
e0ne_ | rdopiera: or use a memcache as an option | 08:45 |
rdopiera | how does that help? | 08:45 |
adriant | well, memcache then handles the overflow | 08:46 |
adriant | as it evacs any old or unused values | 08:46 |
rdopiera | we mostly use memoize to avoid repeating api calls within the same request | 08:47 |
adriant | which since you already will be using it for session caching isn't too bad | 08:47 |
rdopiera | I wonder if we could explicitly clear it at the end of the request | 08:47 |
adriant | that's what I was playing with | 08:48 |
adriant | since I always assumed it was caching for the duration of just a request | 08:48 |
rdopiera | depend on which api call | 08:48 |
rdopiera | but it's mostly one request | 08:48 |
adriant | can we code it with an explicit expiry on the cached values and clear all expired ones after each request? | 08:50 |
adriant | or might that add too much time to each request | 08:51 |
*** btully has quit IRC | 08:51 | |
rdopiera | I don't think time is the problem here, but it would need a substantial rewrite | 08:51 |
rdopiera | as right now it keeps the cached values in a closure | 08:51 |
adriant | oh that's right | 08:52 |
*** btully has joined #openstack-horizon | 08:55 | |
*** btully has quit IRC | 08:59 | |
*** btully has joined #openstack-horizon | 08:59 | |
adriant | ultimately though it's just a rewrite of the util itself and maybe somewhere in the middleware to to clear it | 09:00 |
adriant | I started playing with ways to solve it but was having issues because lots of javascript seemed broken on master :P | 09:01 |
*** btully has quit IRC | 09:07 | |
*** gvrangan has joined #openstack-horizon | 09:08 | |
e0ne_ | adriant: what exactly is broken on master? | 09:08 |
*** kukacz has quit IRC | 09:10 | |
*** btully has joined #openstack-horizon | 09:11 | |
*** kukacz has joined #openstack-horizon | 09:12 | |
*** btully has quit IRC | 09:19 | |
*** btully has joined #openstack-horizon | 09:20 | |
*** btully has quit IRC | 09:28 | |
rdopiera | looking at the memoized code, it should be easy to add a "clear_memoized" function to the decorated function | 09:29 |
rdopiera | you would still need to explicitly call it for every function you want cleared | 09:30 |
*** ianychoi_ has joined #openstack-horizon | 09:30 | |
*** Bhujay has joined #openstack-horizon | 09:32 | |
*** ianychoi has quit IRC | 09:34 | |
rdopiera | adriant: did you narrow it down to a particular function being decorated? | 09:34 |
*** Bhujay has quit IRC | 09:37 | |
*** e0ne_ has quit IRC | 09:38 | |
*** e0ne has joined #openstack-horizon | 09:38 | |
*** ChanServ sets mode: +o e0ne | 09:38 | |
*** btully has joined #openstack-horizon | 09:40 | |
*** btully has quit IRC | 09:44 | |
adriant | rdopiera: not really, we notice it the most with stuff coming from the network topology panel (mostly coss it refreshes) but i think it's just a lot of small calls | 09:47 |
rdopiera | adriant: it would be interesting to see which kind of function arguments is actually problematic | 09:47 |
rdopiera | adriant: for example, strings and ints can't be weakrefed | 09:48 |
*** btully has joined #openstack-horizon | 09:48 | |
rdopiera | so if a function only takes those, it won't ever trigger the cleanup | 09:48 |
adriant | rdopiera: maybe we can stick some extra debug logging into the function which tells us which function has values we can't weakref | 09:50 |
rdopiera | adriant: all functions have such values, the problem only appears when all the values are like that | 09:52 |
*** btully has quit IRC | 09:53 | |
rdopiera | hmm, we could weakref the request object and add cleanup on that... | 09:54 |
adriant | Oh I see, as long as one arg or param in the key is a weakref, then the delete call will clear the cache | 09:56 |
adriant | i see | 09:56 |
*** btully has joined #openstack-horizon | 09:56 | |
*** btully has quit IRC | 10:00 | |
*** btully has joined #openstack-horizon | 10:01 | |
*** Emine has joined #openstack-horizon | 10:01 | |
*** openstackgerrit has joined #openstack-horizon | 10:02 | |
openstackgerrit | Radomir Dopieralski proposed openstack/horizon master: Add clean_memoized to memorized functions https://review.openstack.org/604721 | 10:02 |
rdopiera | adriant: can you try this? ^^ | 10:02 |
*** e0ne has quit IRC | 10:08 | |
*** btully has quit IRC | 10:09 | |
*** btully has joined #openstack-horizon | 10:09 | |
*** btully has quit IRC | 10:14 | |
*** btully has joined #openstack-horizon | 10:17 | |
*** btully has quit IRC | 10:21 | |
*** btully has joined #openstack-horizon | 10:21 | |
*** e0ne has joined #openstack-horizon | 10:27 | |
*** ChanServ sets mode: +o e0ne | 10:27 | |
*** btully has quit IRC | 10:30 | |
*** btully has joined #openstack-horizon | 10:33 | |
*** btully has quit IRC | 10:38 | |
*** e0ne has quit IRC | 10:40 | |
*** Bhujay has joined #openstack-horizon | 10:40 | |
*** btully has joined #openstack-horizon | 10:42 | |
*** btully has quit IRC | 10:46 | |
*** btully has joined #openstack-horizon | 10:46 | |
*** e0ne has joined #openstack-horizon | 10:49 | |
*** ChanServ sets mode: +o e0ne | 10:49 | |
adriant | rdopiera: will take a look at it tomorrow :) | 10:50 |
*** btully has quit IRC | 10:54 | |
*** btully has joined #openstack-horizon | 10:54 | |
*** e0ne has quit IRC | 10:57 | |
*** btully has quit IRC | 11:02 | |
*** Bhujay has quit IRC | 11:02 | |
*** Bhujay has joined #openstack-horizon | 11:05 | |
*** btully has joined #openstack-horizon | 11:06 | |
*** Bhujay has quit IRC | 11:06 | |
*** Bhujay has joined #openstack-horizon | 11:07 | |
*** Bhujay has quit IRC | 11:08 | |
*** Bhujay has joined #openstack-horizon | 11:08 | |
*** Bhujay has quit IRC | 11:09 | |
*** Bhujay has joined #openstack-horizon | 11:10 | |
*** btully has quit IRC | 11:11 | |
*** Bhujay has quit IRC | 11:11 | |
*** Bhujay has joined #openstack-horizon | 11:11 | |
*** btully has joined #openstack-horizon | 11:16 | |
*** mvkr has quit IRC | 11:17 | |
*** btully has quit IRC | 11:20 | |
*** mvkr has joined #openstack-horizon | 11:29 | |
*** btully has joined #openstack-horizon | 11:32 | |
*** btully has quit IRC | 11:37 | |
*** btully has joined #openstack-horizon | 11:42 | |
*** btully has quit IRC | 11:46 | |
*** btully has joined #openstack-horizon | 11:48 | |
*** btully has quit IRC | 11:53 | |
*** btully has joined #openstack-horizon | 11:53 | |
*** btully has quit IRC | 11:57 | |
*** btully has joined #openstack-horizon | 11:58 | |
*** btully has quit IRC | 12:03 | |
*** btully has joined #openstack-horizon | 12:06 | |
*** e0ne has joined #openstack-horizon | 12:10 | |
*** ChanServ sets mode: +o e0ne | 12:10 | |
*** btully has quit IRC | 12:11 | |
*** btully has joined #openstack-horizon | 12:14 | |
*** Bhujay has quit IRC | 12:15 | |
*** Bhujay has joined #openstack-horizon | 12:16 | |
rdopiera | e0ne: I started wondering... what if instead of having all those special workarounds for memoizing functions that take a request object, we simply made the request object hashable? | 12:20 |
e0ne | rdopiera: sounds good | 12:20 |
rdopiera | e0ne: something like def __hash__(self): return id(self) | 12:20 |
*** btully has quit IRC | 12:22 | |
*** btully has joined #openstack-horizon | 12:26 | |
*** btully has quit IRC | 12:31 | |
*** btully has joined #openstack-horizon | 12:31 | |
*** btully has quit IRC | 12:35 | |
*** btully has joined #openstack-horizon | 12:39 | |
*** Bhujay has quit IRC | 12:39 | |
*** Bhujay has joined #openstack-horizon | 12:40 | |
*** Bhujay has quit IRC | 12:41 | |
*** Bhujay has joined #openstack-horizon | 12:41 | |
*** btully has quit IRC | 12:43 | |
*** btully has joined #openstack-horizon | 12:47 | |
*** btully has quit IRC | 12:52 | |
*** jadajada has joined #openstack-horizon | 12:52 | |
*** jadajada has quit IRC | 12:53 | |
*** jrist has joined #openstack-horizon | 13:00 | |
*** btully has joined #openstack-horizon | 13:00 | |
*** lblanchard has joined #openstack-horizon | 13:00 | |
*** btully has quit IRC | 13:04 | |
*** lbragstad has joined #openstack-horizon | 13:17 | |
*** gvrangan has quit IRC | 13:24 | |
*** btully has joined #openstack-horizon | 13:26 | |
*** belmorei_ has joined #openstack-horizon | 13:30 | |
*** btully has quit IRC | 13:30 | |
*** belmoreira has quit IRC | 13:32 | |
*** e0ne has quit IRC | 13:38 | |
*** btully has joined #openstack-horizon | 13:44 | |
*** e0ne has joined #openstack-horizon | 13:47 | |
*** ChanServ sets mode: +o e0ne | 13:47 | |
*** btully has quit IRC | 13:48 | |
*** btully has joined #openstack-horizon | 13:48 | |
*** btully has quit IRC | 14:01 | |
*** belmorei_ has quit IRC | 14:19 | |
openstackgerrit | Radomir Dopieralski proposed openstack/horizon master: Add clean_memoized to memorized functions https://review.openstack.org/604721 | 14:19 |
*** belmoreira has joined #openstack-horizon | 14:30 | |
*** dave-mccowan has quit IRC | 14:50 | |
*** Bhujay has quit IRC | 15:07 | |
*** btully has joined #openstack-horizon | 15:09 | |
*** dave-mccowan has joined #openstack-horizon | 15:12 | |
*** btully has quit IRC | 15:13 | |
*** dave-mccowan has quit IRC | 15:17 | |
*** dave-mccowan has joined #openstack-horizon | 15:23 | |
*** btully has joined #openstack-horizon | 15:44 | |
*** btully has quit IRC | 15:49 | |
*** e0ne has quit IRC | 16:01 | |
*** lbragstad has quit IRC | 16:10 | |
*** lbragstad has joined #openstack-horizon | 16:18 | |
*** rdopiera has quit IRC | 16:27 | |
*** dave-mccowan has quit IRC | 16:38 | |
*** jpich has quit IRC | 16:50 | |
*** btully has joined #openstack-horizon | 16:57 | |
*** btully has quit IRC | 17:01 | |
*** btully has joined #openstack-horizon | 17:02 | |
*** belmoreira has quit IRC | 17:04 | |
*** btully has quit IRC | 17:06 | |
*** btully has joined #openstack-horizon | 17:09 | |
*** btully has quit IRC | 17:13 | |
*** btully has joined #openstack-horizon | 17:17 | |
*** btully has quit IRC | 17:21 | |
*** btully has joined #openstack-horizon | 17:22 | |
*** mvkr has quit IRC | 17:25 | |
*** btully has quit IRC | 17:26 | |
*** tosky has quit IRC | 17:30 | |
*** btully has joined #openstack-horizon | 17:32 | |
*** dave-mccowan has joined #openstack-horizon | 17:35 | |
*** btully has quit IRC | 17:36 | |
*** e0ne has joined #openstack-horizon | 18:06 | |
*** ChanServ sets mode: +o e0ne | 18:06 | |
*** ediardo has joined #openstack-horizon | 18:07 | |
*** coreycb has joined #openstack-horizon | 18:12 | |
*** btully has joined #openstack-horizon | 18:13 | |
*** mvkr has joined #openstack-horizon | 18:15 | |
*** btully has quit IRC | 18:18 | |
*** btully has joined #openstack-horizon | 18:24 | |
*** btully has quit IRC | 18:28 | |
*** btully has joined #openstack-horizon | 18:28 | |
*** btully has quit IRC | 18:33 | |
*** btully has joined #openstack-horizon | 18:38 | |
*** btully has quit IRC | 18:42 | |
*** ianychoi_ is now known as ianychoi | 18:54 | |
*** tosky has joined #openstack-horizon | 19:11 | |
*** pcaruana has quit IRC | 19:17 | |
*** btully has joined #openstack-horizon | 19:20 | |
*** btully has quit IRC | 19:24 | |
*** btully has joined #openstack-horizon | 20:10 | |
*** btully has quit IRC | 20:15 | |
*** btully has joined #openstack-horizon | 20:24 | |
*** lblanchard has quit IRC | 20:28 | |
*** btully has quit IRC | 20:28 | |
*** btully has joined #openstack-horizon | 20:35 | |
*** e0ne has quit IRC | 20:37 | |
*** btully has quit IRC | 20:43 | |
*** btully has joined #openstack-horizon | 20:57 | |
*** btully has quit IRC | 21:02 | |
*** btully has joined #openstack-horizon | 21:32 | |
*** btully has quit IRC | 21:36 | |
*** jtomasek has quit IRC | 21:37 | |
*** btully has joined #openstack-horizon | 21:50 | |
*** btully has quit IRC | 21:55 | |
*** Emine has quit IRC | 22:03 | |
*** jonmills_nasa has joined #openstack-horizon | 22:09 | |
jonmills_nasa | I am seeing an issue in Queens where images uploaded to Glance via CLI work fine, but images uploaded via Horizon fail to be associated with a project/tenant. The image appears as "Image from Other Project" | 22:12 |
jonmills_nasa | I also see things in horizon error log related to problems with oslo.policy | 22:13 |
jonmills_nasa | "[:error] [pid 5654] DEBUG:oslo_policy.policy:Rule [default] does not exist" | 22:13 |
jonmills_nasa | "[:error] [pid 5654] DEBUG:oslo_policy.policy:Rule [admin_and_matching_domain_id] does not exist" | 22:13 |
jonmills_nasa | All of the policy files exist beneath /etc/openstack-dashboard, and are readable | 22:14 |
*** rcernin has joined #openstack-horizon | 22:45 | |
*** kukacz_ has joined #openstack-horizon | 22:55 | |
*** dims_ has joined #openstack-horizon | 22:59 | |
*** kukacz has quit IRC | 23:00 | |
*** dims has quit IRC | 23:00 | |
*** fbouliane has quit IRC | 23:00 | |
*** tosky has quit IRC | 23:03 | |
*** jonmills_nasa has quit IRC | 23:29 | |
*** btully has joined #openstack-horizon | 23:39 | |
openstackgerrit | Adrian Turjak proposed openstack/horizon master: Ugly fix to clear memo cache every request https://review.openstack.org/604934 | 23:39 |
*** btully has quit IRC | 23:43 | |
openstackgerrit | Adrian Turjak proposed openstack/horizon master: Ugly fix to clear memo cache every request https://review.openstack.org/604934 | 23:55 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!