Friday, 2021-10-01

*** gmann is now known as gmann_afk00:07
*** gmann_afk is now known as gmann01:11
tabachaHi, I think i found a bug in the flavour cmd,  I reported it here: https://storyboard.openstack.org/#!/story/2009260 and fixed it here: https://review.opendev.org/c/openstack/ansible-collections-openstack/+/811739 what are the next steps to get it back to the git repo?06:14
*** frickler_ is now known as frickler07:25
noonedeadpunktabacha: I guess you need to ask this in #opanstack-ansible-sig if that's about merging that patch to a collection07:49
tabachanoonedeadpunk: okay I asked there.07:56
*** odyssey4me is now known as Guest148308:12
baptistemmhello14:53
noonedeadpunko/14:54
baptistemmI'm trying to make the ansible inventory plugin for openstack returning something, but it gives me a 403 http error, where using openstack-inventory script works14:55
baptistemmansible-inventory does not give me useful input for troubleshooting14:56
baptistemmcan a good soul give me some hints14:56
baptistemmI've modified ansible.cfg to have a section [inventory] which has 'enable_plugins = openstack, auto'14:57
noonedeadpunkum....14:57
baptistemmI've put a clouds.yaml file in ~/.config/openstack/14:58
noonedeadpunkI'm not quite sure I understand what exactly you're trying to do and why14:58
baptistemmhttps://pastebin.com/EsDKB9Jk14:59
noonedeadpunkare you sure you're trying to use OSA?14:59
noonedeadpunkOr it's question regarding openstack collections?14:59
baptistemmI'm an openstack user so the latter15:00
noonedeadpunkyeah, I got this :) this channel is related to specific deployment tooling https://opendev.org/openstack/openstack-ansible15:00
noonedeadpunk* https://docs.openstack.org/openstack-ansible15:00
noonedeadpunkwhich I'm not sure if used in your usecase15:01
baptistemmah that's a bit misleading 15:01
baptistemm:)15:01
noonedeadpunkeventually we have #opanstack-ansible-sig that is related to ansible collections for openstack15:01
spatelFolks, i have very strange issue... with old openstack vs new Wallaby 15:02
spatelwe have same hardware between Stein and Wallaby and looking at disk IO i am seeing bad performance in Wallaby 15:03
spatelis anything changed in Wallaby ?15:03
noonedeadpunkum, you mean storage performance?15:04
spatelhttps://paste.opendev.org/show/809727/15:05
spatelI am using local Disk 15:05
spatelno ceph storage15:05
spatelnoonedeadpunk what do you use for disk_cachemodes = ?15:09
noonedeadpunkand nvmes?15:13
noonedeadpunkBut for ceph we do use writeback15:13
noonedeadpunkI'm not sure dd is valid thing to measure15:14
noonedeadpunkWhat does fio says with randrw?15:15
spatelif i run dd test on host i get very good numbers 15:15
spatelbut in VM its 3x lower 15:15
spatellet me try fio..15:15
spateli think there must be some setting in nova which is bottleneck 15:15
noonedeadpunkwell, if you don't have writeback - that most likely the thing15:18
spatelnoonedeadpunk what do you think of these numbers  - https://paste.opendev.org/show/809730/ 15:22
spatelwriteback ?15:23
spateldisk_cachemodes = block=writeback ?15:23
noonedeadpunk`disk_cachemodes = network=writeback` but it's ceph15:23
spateli have local storage so it should be disk_cachemodes = block=writeback15:24
spatelcorrect?15:24
noonedeadpunkyeah15:24
noonedeadpunkbut your numbers are kind of close15:24
spatelwhy my dd is very different ? 15:24
spatelalso my client complaining performance isn't good 15:25
spatelthey are not complaining about stein release even they have same hardware.. 15:25
spateldo i need to rebuild VM after changing disk_cachemodes = block=writeback 15:25
noonedeadpunkwith same firware versions as well?15:25
spateli think yes..15:26
spateli didn't look at firmware yet.. 15:26
noonedeadpunkI mean - cpu vulnarabilities being patched can lead to 10-20% of performance degradation15:26
spateli am checking now all hardware level details and going from bottom to top 15:26
spateli know what you saying15:28
spatelbut in compute node directly i am getting 225 MB/s write speed 15:29
spateland in VM i am getting 80 MB/s write speed 15:29
noonedeadpunkand dd from /dev/zero can be hitting some entropy issue15:29
noonedeadpunkfio shows 153MB/s for centos and 136MB/s for ubuntu15:29
noonedeadpunkand 37.3kiops vs 33.2k iops15:30
spateli know.. but very odd same time other environment dd working great so kind of not feeling good15:31
spatelalso customer complaining and i trust them.. 15:31
spatelthey had no issue in old environment 15:31
spatelsomething is changed, either ubuntu vs centos compute or wallaby has some extra overhead 15:32
noonedeadpunkWell, it might be because of missing Haveged or smth like that...15:35
spatelhmm15:35
spatellet me try apt install haveged and see 15:36
spatelnoonedeadpunk should i use disk_cachemodes=file=writeback ? 15:52
spatelkvm using file not block right?15:52
spateldisk_cachemodes = file=writethrough,block=none15:52
spatelnoonedeadpunk good news :)16:44
spatelThis solved my issue, now performance amazing, 3x more -  disk_cachemodes = file=writeback,block=none16:45
spatelone interesting thing if i build ubuntu vm then i don't see any issue but if i build centos 7 vm then i see issue.. 16:55
spatellook like its centos and ubuntu combination issue16:55
spatelnoonedeadpunk there?17:56
mgariepyspatel, same FS in the vm ?18:13
spatelFS?18:13
mgariepyfilesystem18:13
spatelext3 ?18:13
spateli have to check that part 18:14
mgariepythe hosts is ubuntu ?18:14
spatelcompute on ubuntu and vm using centos7 18:15
-opendevstatus- NOTICE: The review.opendev.org Gerrit server has become unreachable as of approximately 19:10 UTC due to a networking issue in the provider, but should be reachable again shortly19:43

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!