*** tosky has quit IRC | 01:19 | |
*** rcernin has quit IRC | 07:56 | |
*** pcaruana has joined #openstack-sahara | 07:58 | |
*** pcaruana is now known as muttley | 07:58 | |
*** rcernin has joined #openstack-sahara | 08:03 | |
*** rcernin has quit IRC | 08:33 | |
*** Gaasmann has quit IRC | 08:56 | |
*** tosky has joined #openstack-sahara | 09:00 | |
tellesnobrega | tosky, are you here already? | 11:15 |
---|---|---|
tosky | yes | 11:17 |
tellesnobrega | I have a question related to the 500 issue | 11:17 |
tellesnobrega | https://storyboard.openstack.org/#!/story/2004506 | 11:17 |
tellesnobrega | do you know when exactly that happens? | 11:18 |
tosky | uh, no, I didn't personally hit it | 11:19 |
tosky | didn't Jeremy provide few details during last meeting? | 11:19 |
tosky | maybe there are also few details on the etherpads of the past PTGs | 11:19 |
tellesnobrega | I will take a look | 11:19 |
tosky | "Any unexpected value in the query string besides the ones used for pagination" sounds like "inject random values in the pagination parameters used in the query" | 11:20 |
tosky | s/random/invalid/ | 11:20 |
tellesnobrega | sure | 11:21 |
openstackgerrit | Luigi Toscano proposed openstack/sahara-tests master: DNM Test without event log https://review.openstack.org/623193 | 11:45 |
tellesnobrega | tosky, I have to go out to get my driver's license, if I'm not back before the meeting starts, can you start the meeting for me? | 12:41 |
tellesnobrega | and I will join as soon as I get home | 12:41 |
*** tmckay has joined #openstack-sahara | 12:56 | |
tosky | sure | 13:04 |
*** muttley has quit IRC | 13:08 | |
*** tmckay has quit IRC | 13:12 | |
*** tmckay has joined #openstack-sahara | 13:13 | |
*** rcernin has joined #openstack-sahara | 13:17 | |
tellesnobrega | tosky, back :) | 13:20 |
tosky | yay :) | 13:20 |
*** muttley has joined #openstack-sahara | 13:21 | |
*** muttley has quit IRC | 13:25 | |
*** muttley has joined #openstack-sahara | 13:26 | |
*** rcernin has quit IRC | 13:29 | |
*** muttley has quit IRC | 13:29 | |
*** pcaruana has joined #openstack-sahara | 13:34 | |
*** pcaruana has quit IRC | 13:39 | |
*** pcaruana has joined #openstack-sahara | 13:43 | |
*** pcaruana has quit IRC | 13:47 | |
*** jeremyfreudberg has joined #openstack-sahara | 13:59 | |
*** tmckay has quit IRC | 14:19 | |
*** jeremyfreudberg has quit IRC | 15:00 | |
tosky | tellesnobrega: with split plugins, if sahara-api and sahara-engine are deployed on different machines, do the plugins need to be installed on both, or just for -engine? | 16:00 |
*** Gaasmann has joined #openstack-sahara | 16:08 | |
Gaasmann | tosky: the other day you were saying it wasn't that long when you build images using the new way on a VM. How long does it takes for you? | 16:14 |
tosky | uh, it depends on the image | 16:14 |
tosky | let's do an interactive test - which image should I build? | 16:14 |
Gaasmann | well honestly it's more about an order of magnitude than a precise timing, but if you want to test anyway, I'm on Ubuntu/cdh and Ubuntu/Ambari (but the latter fails) | 16:16 |
tellesnobrega | tosky, I would say just the engine | 16:18 |
tellesnobrega | api calls go to the engine before it goes to the plugins | 16:18 |
tosky | Gaasmann: I checked my internal an internal test builder, which uses instances from an internal openstack instance | 16:18 |
tosky | the last run of RHEL/CDH 5.13 took 55 minutes; RHEL/Ambari 2.6 took 36 minutes | 16:19 |
tosky | tellesnobrega: ok, it's important in case packagers asks; for example, this means that plugins need to live only on the sahara-engine containers on kolla "packaging" | 16:20 |
tellesnobrega | I see | 16:20 |
Gaasmann | tosky: thanks. Ubuntu/cdh took 2h32 and Ubuntu/ambari(incomplete) took 1h24. So it's way longer on my side but it may be because of my setup. | 16:23 |
tosky | uh | 16:23 |
tellesnobrega | that is a long time | 16:23 |
tosky | the speed of the connection helps too, but usually the CDH mirrors (and ambari mirrors) are not so fast | 16:23 |
Gaasmann | It's obviously the "apt-get install" commands that take the most time but I don't have details. | 16:30 |
tosky | uhm, you may try to wrap it around eatmydata | 16:32 |
tosky | it's dangerous and not safe and whatnot, but it delays the fsync, so it makes it fast (if everything works) | 16:34 |
Gaasmann | Is not safe like "I could loose data" type of risk? | 16:40 |
tosky | technically yes, it's possible | 16:41 |
Gaasmann | ok, then it shouldn't be a problem in this case I guess | 16:42 |
tosky | Gaasmann: https://wiki.debian.org/Teams/Dpkg/FAQ#Q:_Why_is_dpkg_so_slow_when_using_new_filesystems_such_as_btrfs_or_ext4.3F | 16:42 |
Gaasmann | interresting | 16:44 |
*** dave-mccowan has joined #openstack-sahara | 18:29 | |
openstackgerrit | Nicolas Haller proposed openstack/sahara stable/rocky: Fix image generation for ambari plugin on Ubuntu https://review.openstack.org/623289 | 19:18 |
tosky | Gaasmann: it should go to master first | 19:19 |
tosky | libtirpc-devel require an os_case, as it's needed on centos too | 19:21 |
Gaasmann | Ho I wonder how I pushed that on rocky ^^ | 19:44 |
openstackgerrit | Telles Mota Vidal Nóbrega proposed openstack/sahara master: APIv2 - Fix 500 on malformed query string https://review.openstack.org/623300 | 19:53 |
*** rcernin has joined #openstack-sahara | 20:12 | |
openstackgerrit | Telles Mota Vidal Nóbrega proposed openstack/sahara master: APIv2 - Fix 500 on malformed query string https://review.openstack.org/623300 | 21:34 |
openstackgerrit | Telles Mota Vidal Nóbrega proposed openstack/sahara master: APIv2 - Fix 500 on malformed query string https://review.openstack.org/623300 | 21:37 |
*** tmckay has joined #openstack-sahara | 22:05 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!