| *** dsneddon has joined #openstack-ironic | 00:50 | |
| *** dsneddon has quit IRC | 00:55 | |
| *** dsneddon has joined #openstack-ironic | 00:56 | |
| *** igordc has quit IRC | 01:24 | |
| *** dsneddon has quit IRC | 01:42 | |
| *** dsneddon has joined #openstack-ironic | 02:07 | |
| *** dsneddon has quit IRC | 02:33 | |
| *** rcernin has quit IRC | 02:45 | |
| *** rcernin has joined #openstack-ironic | 02:47 | |
| *** dsneddon has joined #openstack-ironic | 02:58 | |
| *** ricolin_ has joined #openstack-ironic | 03:06 | |
| *** dsneddon has quit IRC | 03:21 | |
| *** sabedevops has joined #openstack-ironic | 04:31 | |
| *** sabedevops has quit IRC | 04:37 | |
| openstackgerrit | Merged openstack/ironic-prometheus-exporter master: build universal wheels https://review.opendev.org/672941 | 05:17 |
|---|---|---|
| *** brtknr has quit IRC | 05:17 | |
| *** brtknr has joined #openstack-ironic | 05:19 | |
| *** rachit7 has joined #openstack-ironic | 05:19 | |
| *** whoami-rajat has joined #openstack-ironic | 05:22 | |
| *** zaneb has quit IRC | 05:39 | |
| *** zbitter has joined #openstack-ironic | 05:39 | |
| *** e0ne has joined #openstack-ironic | 05:41 | |
| *** e0ne has quit IRC | 06:11 | |
| *** mkrai has joined #openstack-ironic | 06:36 | |
| *** etingof has joined #openstack-ironic | 06:39 | |
| *** tesseract has joined #openstack-ironic | 07:08 | |
| *** iurygregory has joined #openstack-ironic | 07:11 | |
| *** rcernin has quit IRC | 07:11 | |
| *** rpittau|afk is now known as rpittau | 07:11 | |
| rpittau | good morning ironic! o/ | 07:12 |
| iurygregory | good morning rpittau and Ironic o/ | 07:14 |
| rpittau | hey iurygregory :) | 07:14 |
| *** tssurya has joined #openstack-ironic | 07:14 | |
| *** pcaruana has joined #openstack-ironic | 07:31 | |
| *** lucasagomes has joined #openstack-ironic | 07:41 | |
| *** dsneddon has joined #openstack-ironic | 07:50 | |
| *** dsneddon has quit IRC | 07:55 | |
| *** iurygregory has quit IRC | 08:01 | |
| *** dougsz has joined #openstack-ironic | 08:12 | |
| *** iurygregory has joined #openstack-ironic | 08:31 | |
| *** brtknr has quit IRC | 08:33 | |
| *** brtknr has joined #openstack-ironic | 08:36 | |
| *** yolanda has joined #openstack-ironic | 08:42 | |
| *** hwoarang__ has quit IRC | 08:43 | |
| *** hwoarang has joined #openstack-ironic | 08:43 | |
| *** gkadam has joined #openstack-ironic | 08:44 | |
| *** Goneri has joined #openstack-ironic | 08:45 | |
| *** mkrai has quit IRC | 08:46 | |
| *** derekh has joined #openstack-ironic | 08:53 | |
| *** dtantsur|afk is now known as dtantsur | 08:58 | |
| *** tesseract has quit IRC | 08:58 | |
| dtantsur | morning | 08:58 |
| *** dsneddon has joined #openstack-ironic | 09:00 | |
| *** dsneddon has quit IRC | 09:05 | |
| *** mkrai has joined #openstack-ironic | 09:07 | |
| *** e0ne has joined #openstack-ironic | 09:08 | |
| iurygregory | morning dtantsur | 09:11 |
| *** hemanth_n has joined #openstack-ironic | 09:14 | |
| *** trident has quit IRC | 09:16 | |
| *** trident has joined #openstack-ironic | 09:17 | |
| *** tesseract has joined #openstack-ironic | 09:20 | |
| *** csmart has quit IRC | 09:22 | |
| *** ajya[m] has quit IRC | 09:22 | |
| *** altlogbot_2 has quit IRC | 09:24 | |
| *** altlogbot_2 has joined #openstack-ironic | 09:25 | |
| *** ricolin_ has quit IRC | 09:29 | |
| *** csmart has joined #openstack-ironic | 09:34 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Add an option to abort cleaning and deployment if node is in maintenance https://review.opendev.org/366828 | 09:37 |
| *** S4ren has joined #openstack-ironic | 09:39 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allow configuring global deploy and rescue kernel/ramdisk https://review.opendev.org/672989 | 09:47 |
| *** Goneri has quit IRC | 09:49 | |
| *** Goneri has joined #openstack-ironic | 10:04 | |
| dtantsur | TheJulia: I have some bad news: apparently reloading configuration with SIGHUP doesn't work because oslo.service shutdowns on SIGHUP :( | 10:04 |
| dtantsur | okay, false alert, I may be misinterpreting the logs | 10:07 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allow configuring global deploy and rescue kernel/ramdisk https://review.opendev.org/672989 | 10:08 |
| *** Lucas_Gray has joined #openstack-ironic | 10:21 | |
| *** ajya[m] has joined #openstack-ironic | 10:26 | |
| *** brtknr has quit IRC | 10:30 | |
| *** brtknr has joined #openstack-ironic | 10:32 | |
| *** brtknr has quit IRC | 10:34 | |
| *** Wryhder has joined #openstack-ironic | 10:35 | |
| *** brtknr has joined #openstack-ironic | 10:35 | |
| *** Lucas_Gray has quit IRC | 10:35 | |
| *** brtknr has quit IRC | 10:36 | |
| *** Wryhder is now known as Lucas_Gray | 10:36 | |
| *** jawad_axd has joined #openstack-ironic | 10:38 | |
| *** yaawang has quit IRC | 10:40 | |
| *** brtknr has joined #openstack-ironic | 10:45 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Add an option to abort cleaning and deployment if node is in maintenance https://review.opendev.org/366828 | 10:56 |
| *** mkrai has quit IRC | 10:57 | |
| *** mkrai has joined #openstack-ironic | 10:57 | |
| *** gkadam has quit IRC | 11:02 | |
| *** brtknr has quit IRC | 11:04 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic master: Fixes power-on failure for 'ilo' hardware type https://review.opendev.org/673008 | 11:05 |
| *** brtknr has joined #openstack-ironic | 11:05 | |
| *** brtknr has quit IRC | 11:06 | |
| *** brtknr has joined #openstack-ironic | 11:06 | |
| *** brtknr has quit IRC | 11:07 | |
| *** brtknr has joined #openstack-ironic | 11:08 | |
| *** hemanth_n has quit IRC | 11:10 | |
| *** Lucas_Gray has quit IRC | 11:15 | |
| *** gkadam has joined #openstack-ironic | 11:17 | |
| *** brtknr has quit IRC | 11:20 | |
| *** brtknr has joined #openstack-ironic | 11:21 | |
| *** brtknr has quit IRC | 11:22 | |
| *** brtknr has joined #openstack-ironic | 11:23 | |
| *** brtknr has quit IRC | 11:23 | |
| *** brtknr has joined #openstack-ironic | 11:24 | |
| *** brtknr has quit IRC | 11:24 | |
| *** brtknr has joined #openstack-ironic | 11:25 | |
| *** Lucas_Gray has joined #openstack-ironic | 11:26 | |
| *** jaypipes has joined #openstack-ironic | 11:26 | |
| *** Lucas_Gray has quit IRC | 11:28 | |
| *** Lucas_Gray has joined #openstack-ironic | 11:33 | |
| TheJulia | dtantsur: Oh.... good. I almost didn't get out of bed | 11:36 |
| TheJulia | pillow or sighup... pillow or sighup | 11:36 |
| *** gkadam has quit IRC | 11:42 | |
| *** derekh has quit IRC | 11:43 | |
| *** gkadam has joined #openstack-ironic | 11:44 | |
| *** rh-jelabarre has joined #openstack-ironic | 12:02 | |
| TheJulia | Does anyone have any thoughts on backporting https://review.opendev.org/#/c/670697 ? | 12:07 |
| patchbot | patch 670697 - ironic - IPA does not boot up after cleaning reboot for 're... (MERGED) - 2 patch sets | 12:07 |
| TheJulia | https://review.opendev.org/#/c/654038 would love some reviews | 12:13 |
| patchbot | patch 654038 - ironic - Option to send all portgroup data - 7 patch sets | 12:13 |
| *** mkrai has quit IRC | 12:13 | |
| TheJulia | xXraphXx: I believe https://review.opendev.org/#/c/669906/ needs to be backported? | 12:16 |
| patchbot | patch 669906 - ironic - Ansible module: fix deployment for private and/or ... (MERGED) - 1 patch set | 12:16 |
| TheJulia | https://review.opendev.org/#/c/666296/4 would also love some reviews | 12:17 |
| patchbot | patch 666296 - sushy-tools - Add state sharing and persistence mechanism - 4 patch sets | 12:17 |
| *** jcoufal has joined #openstack-ironic | 12:22 | |
| xXraphXx | TheJulia: yep, we can backport it. Doing. Question, how far in the past should we backport ? Queens or Rocky ? | 12:24 |
| TheJulia | xXraphXx: As far back makes sense as a consumer. So if you want to backport it all the way to ocata, we'll merge it, but it will never be officially released | 12:26 |
| openstackgerrit | raphael.glon proposed openstack/ironic stable/stein: Ansible module: fix deployment for private and/or shared images https://review.opendev.org/673283 | 12:28 |
| TheJulia | NobodyCam: If you review https://review.opendev.org/#/c/662057/, I'll make you guys pizza when I'm back home in socal. | 12:29 |
| patchbot | patch 662057 - bifrost - Fix troubleshooting docs - SSH access to CoreOS - 2 patch sets | 12:29 |
| openstackgerrit | raphael.glon proposed openstack/ironic stable/rocky: Ansible module: fix deployment for private and/or shared images https://review.opendev.org/673284 | 12:29 |
| *** ricolin_ has joined #openstack-ironic | 12:30 | |
| *** rachit7 has quit IRC | 12:31 | |
| TheJulia | xXraphXx: I've added those two patches to the etherpad so they have some review visibility | 12:38 |
| xXraphXx | ok thanks | 12:42 |
| TheJulia | etingof: by chance did you investigate the standalone job failure on https://review.opendev.org/#/c/638453/ ? | 12:46 |
| patchbot | patch 638453 - ironic - Add Redfish Virtual Media Boot support - 40 patch sets | 12:46 |
| etingof | TheJulia, "Unable to stat device /dev/vda1 after attempting to verify 3 times", hmmm | 12:48 |
| TheJulia | etingof: on an iscsi deploy? | 12:48 |
| TheJulia | actually | 12:49 |
| TheJulia | that wouldn't matter | 12:49 |
| etingof | perhaps, it's in ironic.drivers.modules.agent | 12:49 |
| TheJulia | That CI bug is getting annoying | 12:49 |
| TheJulia | Yeah, it is telling the agent to get started | 12:49 |
| TheJulia | and boom | 12:49 |
| etingof | is it a famous bug? | 12:50 |
| TheJulia | Its a bug where the agent doesn't see the freshly partitioned partition | 12:51 |
| etingof | right, why that happens? | 12:52 |
| TheJulia | My theory is disk IO latency because the rescans occur using O_DIRECT, so I bet they have code that only allows it to try for so long... | 12:55 |
| TheJulia | So if a chance is made via iscsi, there is no guarantee it is on the platter yet. | 12:56 |
| TheJulia | Alternatively, the device could be locked. If locked those programs will typically skip the device | 12:56 |
| * TheJulia changes locations | 12:57 | |
| etingof | may be more rescans then? the bread and butter of software engineers | 12:57 |
| rpittau | TheJulia: I think it makes sense to backport https://review.opendev.org/#/c/670697 , just wondering how clean that would be older than Rocky releases | 13:00 |
| patchbot | patch 670697 - ironic - IPA does not boot up after cleaning reboot for 're... (MERGED) - 2 patch sets | 13:00 |
| TheJulia | etingof: more rescans likely won't fix it, I think we already try ten times. Maybe if we backed-off between them. As a crazy idea. What if we had the conductor call for a sync() or fsync() after the iscsi write process completes? | 13:06 |
| TheJulia | It would be performance impacting because that call is blocking and depending on activity level the conductor could be very busy.... but we don't recommend iscsi for high usage environments. | 13:06 |
| TheJulia | s/high/heavy/ | 13:06 |
| *** jroll has quit IRC | 13:07 | |
| *** jawad_axd has quit IRC | 13:08 | |
| *** jroll has joined #openstack-ironic | 13:08 | |
| etingof | would it help of the agent would do sync() instead? is not it its local drive after all? | 13:08 |
| etingof | or it won't help because the data may be still on the conductor side? | 13:09 |
| TheJulia | etingof: not really, because it has no idea what is coming over the iscsi connection | 13:09 |
| TheJulia | if that file system table update or even the io packets are still streaming down, its not going to know it | 13:10 |
| etingof | so that's the essence of the racing here | 13:10 |
| etingof | we do fdisk over iscsi and then proceed locally | 13:11 |
| etingof | what if the agent would mount its own iscsi target? | 13:11 |
| TheJulia | or we write a whole disk image and the device is locked | 13:11 |
| etingof | in the latter case it seems reasonable to back off | 13:12 |
| TheJulia | etingof: won't change the scenario | 13:12 |
| etingof | so why the agent wants to see the partition table? | 13:14 |
| *** S4ren has quit IRC | 13:15 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: CI Documentation https://review.opendev.org/672778 | 13:18 |
| TheJulia | etingof: determine the root/boot partition uuids | 13:24 |
| etingof | may be conductor could communicate that? | 13:25 |
| *** rloo has joined #openstack-ironic | 13:26 | |
| TheJulia | That woud be majorly incompatible retooling | 13:29 |
| etingof | protocol change? or what do you mean - retooling? | 13:30 |
| TheJulia | so the conductor may know nothing, and it may just be streaming a whole disk image | 13:35 |
| TheJulia | it can stop, rescan, but even then until it's buffers are flushed out complete, that data is not completetly on the remote host | 13:35 |
| TheJulia | so that remote host can't proceed, and even the conductor wants that data if a direct to disk write-out is occuring, it has to scan and then orcustrate the needful | 13:36 |
| *** whoami-rajat has quit IRC | 13:42 | |
| etingof | it seems the conductor can read the partition table out of its local cache even if the data is not landed on the plates yet... but, yeah, it's not clear how to determine when the agent could proceed with its things... | 13:52 |
| TheJulia | Well, problem is, even if the conductor tries to refresh and read it, it is direct_io which means cache bypassing in thisc ase | 13:58 |
| TheJulia | case | 13:58 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/stein: IPA does not boot up after cleaning reboot for 'redfish' bios interface https://review.opendev.org/673296 | 14:01 |
| *** S4ren has joined #openstack-ironic | 14:04 | |
| etingof | ah, O_DIRECT it is, yeah | 14:06 |
| rpittau | TheJulia: just verified, https://review.opendev.org/670697 is not a clean merge even from Rocky | 14:08 |
| patchbot | patch 670697 - ironic - IPA does not boot up after cleaning reboot for 're... (MERGED) - 2 patch sets | 14:08 |
| dtantsur | TheJulia: hmm, I have a feeling that using SIGHUP does make ironic unusable: https://review.opendev.org/#/c/672989/ ... | 14:15 |
| patchbot | patch 672989 - ironic - Allow configuring global deploy and rescue kernel/... - 3 patch sets | 14:15 |
| * dtantsur changes for restart_service for now | 14:16 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allow configuring global deploy and rescue kernel/ramdisk https://review.opendev.org/672989 | 14:16 |
| *** Lucas_Gray has quit IRC | 14:17 | |
| *** Lucas_Gray has joined #openstack-ironic | 14:17 | |
| dtantsur | OUCH | 14:20 |
| dtantsur | folks, our CI seems to have gone RED. all IPMI jobs. | 14:20 |
| dtantsur | mmm, not all, bifrost and metalsmith seem unaffected | 14:21 |
| dtantsur | Node did not reach manageable state in 120 seconds | 14:21 |
| *** S4ren has quit IRC | 14:23 | |
| *** S4ren has joined #openstack-ironic | 14:23 | |
| dtantsur | etingof: do you know why we still start vbmc the old way? http://logs.openstack.org/84/673284/1/check/ironic-dsvm-standalone/bd45089/controller/logs/ironic-bm-logs/virtualbmc_log.txt.gz#_2019-07-29_12_57_42_998 | 14:24 |
| dtantsur | also this is suspicious http://logs.openstack.org/84/673284/1/check/ironic-dsvm-standalone/bd45089/controller/logs/ironic-bm-logs/virtualbmc_log.txt.gz#_2019-07-29_12_59_09_304 | 14:24 |
| etingof | dtantsur, that's the fallback, it occurs is the mother-daemon is dead | 14:25 |
| dtantsur | etingof: right, but wasn't it fixed? | 14:25 |
| dtantsur | https://opendev.org/x/pyghmi/commit/eb736283606fa5120f47012f917b945141a722f4 is a suspect for the breakage | 14:25 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: CI Documentation https://review.opendev.org/672778 | 14:25 |
| etingof | dtantsur, I think we changed devstack ironic to invoke the master daemon long ago | 14:26 |
| *** mkrai has joined #openstack-ironic | 14:31 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [DNM] Try blocking pyghmi 1.4.0 https://review.opendev.org/673301 | 14:34 |
| TheJulia | ugh, I had a feeling there were CI issues this morning, but didn't look yet :( | 14:37 |
| dtantsur | TheJulia: https://storyboard.openstack.org/#!/story/2006301 I had a patch above to test it | 14:38 |
| TheJulia | ooh, ouch | 14:39 |
| TheJulia | breaking pyghmi change? | 14:39 |
| dtantsur | I need to check again, u-c has pyghmi 1.3.0.. | 14:40 |
| dtantsur | and we're bypassing u-c, so nice of us! | 14:41 |
| TheJulia | woot | 14:41 |
| *** mrtmr has joined #openstack-ironic | 14:42 | |
| *** sabedevops has joined #openstack-ironic | 14:45 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: CI: stop using pyghmi from git master https://review.opendev.org/673301 | 14:45 |
| dtantsur | TheJulia: this may be the fix for now ^^^ | 14:45 |
| *** jaypipes has quit IRC | 14:46 | |
| dtantsur | holy cow, now we have a storn | 14:46 |
| dtantsur | * storm | 14:46 |
| mrtmr | hi Im trying to do make a RAID disk deployment with bifrost. but ı couldnt find a any implementation in bifrost about RAID so I tried to make with ironic raid configuration after dynamic enrollment step in bifrost I go to openstack cli add raid conf file and then I go on with dynamic deployment step in bifrost but while in deploying step image part | 14:49 |
| mrtmr | ition | 14:49 |
| TheJulia | dtantsur: like.... flying cows? | 14:49 |
| TheJulia | kind of storm? | 14:49 |
| TheJulia | mrtmr: interesting | 14:50 |
| *** Lucas_Gray has quit IRC | 14:50 | |
| mrtmr | couldnt find any partition | 14:50 |
| mrtmr | is there a way to do raid conf with bifrost | 14:51 |
| *** Lucas_Gray has joined #openstack-ironic | 14:51 | |
| TheJulia | you'd have to directly invoke openstack cli commands because those fields are not available to be set via the ansible modules | 14:52 |
| dtantsur | TheJulia: yeah, if we had cows around, they would totally fly | 14:53 |
| TheJulia | mrtmr: mainly because nobody has had a need to extend the ansible module to suppor thte raid configuration fields. As for finding the partition, could you elaborate a little more so we can understand what your encountering? | 14:53 |
| dtantsur | I don't think openstacksdk even supports our RAID.. largely because of the crazy non-restful things we are doing :( | 14:54 |
| TheJulia | I believe the fields should be able to be set... but no explicit extra help | 14:57 |
| etingof | it if's in Germany, I'd rather expect pigs flying by | 15:00 |
| *** sbroderick has joined #openstack-ironic | 15:01 | |
| TheJulia | heh | 15:01 |
| *** sabedevops has quit IRC | 15:01 | |
| iurygregory | pigs flying seems to be cool | 15:01 |
| etingof | pigs flying fast might get hot | 15:02 |
| iurygregory | since its 42C maybe you would get free bacon dtantsur XD | 15:03 |
| dtantsur | lol | 15:03 |
| dtantsur | fortunately it was "just" 33 today, and getting colder | 15:03 |
| iurygregory | awesome I would say XD | 15:03 |
| iurygregory | we are with 28 | 15:04 |
| * rpittau heppy for the wind keeping the temp "low", not happy about the fires | 15:04 | |
| iurygregory | dtantsur, I think they merged a fix in pyghmi | 15:05 |
| iurygregory | oh no its the patch that broke us | 15:06 |
| *** Lucas_Gray has quit IRC | 15:06 | |
| iurygregory | lol the only have jobs for pep-8 and unit tests O.o that would explain a lot hehehe | 15:06 |
| *** Lucas_Gray has joined #openstack-ironic | 15:07 | |
| *** stendulker has joined #openstack-ironic | 15:07 | |
| mrtmr | http://paste.openstack.org/show/755077/ | 15:18 |
| mrtmr | I listed all what did I in this steps | 15:18 |
| *** sbroderick has quit IRC | 15:20 | |
| openstackgerrit | Ilya Etingof proposed openstack/sushy master: Disable HTTP connection pooling https://review.opendev.org/673312 | 15:24 |
| *** Lucas_Gray has quit IRC | 15:26 | |
| *** tssurya has quit IRC | 15:26 | |
| *** mkrai has quit IRC | 15:27 | |
| *** e0ne has quit IRC | 15:30 | |
| mrtmr | dtantsur TheJulia http://paste.openstack.org/show/755077/ i listed all things what did i :blush: | 15:31 |
| *** sabedevops has joined #openstack-ironic | 15:31 | |
| *** dsneddon has joined #openstack-ironic | 15:31 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: [WIP] use openstacksdk for ironic module https://review.opendev.org/672179 | 15:33 |
| NobodyCam | Good Morning Ironic'ers | 15:37 |
| NobodyCam | TheJulia: | 15:38 |
| NobodyCam | I'll take a look! | 15:38 |
| *** jjohnson42 has joined #openstack-ironic | 15:38 | |
| jjohnson42 | dtantsur, sorry about that | 15:38 |
| *** gyee has joined #openstack-ironic | 15:39 | |
| *** Lucas_Gray has joined #openstack-ironic | 15:41 | |
| *** mkrai has joined #openstack-ironic | 15:41 | |
| *** mkrai has quit IRC | 15:43 | |
| TheJulia | NobodyCam: thanks | 15:44 |
| NobodyCam | :) hope you had a good weekend | 15:45 |
| dtantsur | jjohnson42: no worries, things happen | 15:47 |
| *** lucasagomes has quit IRC | 15:47 | |
| jjohnson42 | dtantsur, 1.4.1 will be tagged shortly with the two lines added to restore 1.3.0 capability on bmc side | 15:48 |
| dtantsur | many thanks! | 15:48 |
| jjohnson42 | dtantsur, incidentally, I was driven to do the change due to the emergence of new servers without support for HMAC-SHA1-96 | 15:48 |
| jjohnson42 | so in ipmitool speak, those servers require -C 17 argument | 15:48 |
| jjohnson42 | pyghmi as a client now tries for HMAC-SHA256-128 and then falls back to HMAC-SHA1 to accomodate those servers | 15:49 |
| jjohnson42 | as a BMC, I think 1.4.2 will add properly identifying the requestors integrity/auth algorithm and using SHA256 or SHA1 correctly in response to the requestor | 15:50 |
| jjohnson42 | which would allow tests for default (-C 3) and the ones that require -C 17 | 15:50 |
| dtantsur | cool | 15:51 |
| jjohnson42 | it's all frankly a bit silly, since HMAC transform renders the SHA1 weaknesses moot, but I understand why some circles that's hard to explain why SHA1 is 'dangerous' in digest but not HMAC | 15:52 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: CI: stop using pyghmi from git master https://review.opendev.org/673301 | 15:53 |
| dtantsur | TheJulia: found another place where we clone pyghmi ^^ | 15:53 |
| dtantsur | jjohnson42: yeah, doesn't sound like an easy conversation :D | 15:53 |
| *** derekh has joined #openstack-ironic | 15:54 | |
| jjohnson42 | dtantsur, master and 1.4.1 should be available now/presently with the added two lines to fix the missing lib/len values for cipher suite 3 as a bmc | 15:58 |
| *** rh-jelabarre has quit IRC | 15:58 | |
| dtantsur | thx again! | 15:59 |
| rpittau | bye all! good night! o/ | 16:04 |
| *** rpittau is now known as rpittau|afk | 16:04 | |
| *** sabedevops has quit IRC | 16:10 | |
| *** sabedevops has joined #openstack-ironic | 16:12 | |
| mrtmr | do you have any suggestion about raid configuration? http://paste.openstack.org/show/755077/ | 16:12 |
| *** stendulker has quit IRC | 16:15 | |
| *** Lucas_Gray has quit IRC | 16:20 | |
| *** rh-jelabarre has joined #openstack-ironic | 16:21 | |
| *** Goneri has quit IRC | 16:24 | |
| *** tesseract has quit IRC | 16:28 | |
| dtantsur | mrtmr: ipmi doesn't have RAID on its own, the "agent" RAID interface assumes you have a hardware-specific RAID implementation on your deploy ramdisk. Is it the case? (if unsure, say no) | 16:36 |
| *** sabedevops has quit IRC | 16:37 | |
| *** dtantsur is now known as dtantsur|afk | 16:44 | |
| dtantsur|afk | \o | 16:44 |
| *** S4ren has quit IRC | 16:47 | |
| *** sabedevops has joined #openstack-ironic | 16:50 | |
| *** sthussey has joined #openstack-ironic | 16:51 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Use HTTPProxyToWSGI middleware from oslo https://review.opendev.org/672819 | 16:58 |
| *** igordc has joined #openstack-ironic | 16:59 | |
| *** dougsz has quit IRC | 16:59 | |
| *** derekh has quit IRC | 17:00 | |
| mrtmr | dtantsur|afk what should I do RAID interface option, in ironic master branch document, it say "for Software RAID, this would be agent." on the workflow section | 17:05 |
| mrtmr | https://opendev.org/openstack/ironic/src/branch/master/doc/source/admin/raid.rst | 17:06 |
| mrtmr | ı follow this document | 17:06 |
| *** e0ne has joined #openstack-ironic | 17:14 | |
| *** e0ne has quit IRC | 17:20 | |
| *** sabedevops has quit IRC | 17:22 | |
| *** ricolin_ has quit IRC | 17:24 | |
| *** sabedevops has joined #openstack-ironic | 17:36 | |
| TheJulia | mrtmr: You would use agent, and you'd pass configuration information to tell the agent to build a software raid set | 17:38 |
| *** whoami-rajat has joined #openstack-ironic | 17:39 | |
| TheJulia | mrtmr: oh! so your not invoking the actual raid configuraiton step, your just wanting it to happen upon deploy? | 17:40 |
| *** sabedevops has quit IRC | 17:42 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: [WIP] use openstacksdk for ironic module https://review.opendev.org/672179 | 17:46 |
| *** trident has quit IRC | 17:47 | |
| *** trident has joined #openstack-ironic | 17:51 | |
| mrtmr | TheJulia http://paste.openstack.org/show/755077/ I follow this steps, I set the raid conf with that --target-raid-config param before deployment playbook but while in deploying my node doesdnt know /dev/sda or /dev/md0. Im doing someting wrong or missing steps | 17:55 |
| *** sabedevops has joined #openstack-ironic | 17:57 | |
| mrtmr | I thought if i set root_device param with that /dev/md0 in openstack cli it could work but nothing I got same error (error massage is in the paste) | 17:58 |
| *** jjohnson42 has quit IRC | 18:10 | |
| *** jjohnson42 has joined #openstack-ironic | 18:10 | |
| *** sabedevops has quit IRC | 18:33 | |
| *** sabedevops has joined #openstack-ironic | 18:34 | |
| TheJulia | that won't work if the device is not set yet | 18:36 |
| TheJulia | err, created | 18:36 |
| TheJulia | hmm | 18:36 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Change debug to error for heartbeats https://review.opendev.org/649443 | 18:41 |
| TheJulia | mjturek: o/ looks like powerkvm ci is having issues | 18:52 |
| TheJulia | https://oplab9.parqtec.unicamp.br/pub/ppc64el/openstack/ironic/01/673301/3/check-ironic/tempest-dsvm-ironic-agent_ipmitool/973713a/console.html | 18:52 |
| TheJulia | err | 18:52 |
| * TheJulia blinks at the url | 18:52 | |
| openstackgerrit | Merged openstack/ironic-prometheus-exporter master: Payload Timestamp Metric https://review.opendev.org/671514 | 18:58 |
| openstackgerrit | Merged openstack/ironic-prometheus-exporter master: Voltage Metrics https://review.opendev.org/671609 | 18:58 |
| *** mrtmr has quit IRC | 19:01 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Enable no IP address to be returned https://review.opendev.org/499376 | 19:10 |
| *** e0ne has joined #openstack-ironic | 19:22 | |
| *** jcoufal_ has joined #openstack-ironic | 19:28 | |
| *** jcoufal has quit IRC | 19:31 | |
| *** sabedevops has quit IRC | 19:31 | |
| *** jcoufal_ has quit IRC | 19:34 | |
| TheJulia | hmm... mrtmr timed out :( | 19:50 |
| openstackgerrit | Merged openstack/ironic master: Remove deprecated Neutron authentication options https://review.opendev.org/672971 | 20:17 |
| *** jjohnson2_ has joined #openstack-ironic | 20:32 | |
| openstackgerrit | Merged openstack/ironic master: Fix potential race condition on node power on and reboot https://review.opendev.org/633833 | 20:34 |
| *** jjohnson42 has quit IRC | 20:35 | |
| *** jjohnson2_ has quit IRC | 20:39 | |
| *** whoami-rajat has quit IRC | 20:39 | |
| *** e0ne has quit IRC | 20:46 | |
| *** trident has quit IRC | 20:49 | |
| *** eandersson has joined #openstack-ironic | 20:49 | |
| *** trident has joined #openstack-ironic | 20:52 | |
| *** betherly has joined #openstack-ironic | 21:42 | |
| *** betherly has quit IRC | 21:47 | |
| *** bdodd has quit IRC | 21:47 | |
| *** betherly has joined #openstack-ironic | 21:48 | |
| *** betherly has quit IRC | 21:53 | |
| *** tripul0noob has joined #openstack-ironic | 22:09 | |
| tripul0noob | Hi all. | 22:10 |
| tripul0noob | I am using tripleo and it seems a few machines are in a state of:" clean wait" after introspection. | 22:10 |
| tripul0noob | Seems that it was to clean the disks for me, but likely timed out. | 22:10 |
| tripul0noob | Now i'm wondering how I can reset that state of the few failed physical machines and/or redeploy without clean option. | 22:10 |
| openstackgerrit | Merged openstack/ironic master: CI: stop using pyghmi from git master https://review.opendev.org/673301 | 22:11 |
| tripul0noob | Should I just be able to remove the node, re-add and re-introspect as has been proposed? Or will it likely enter the same state? :-/ | 22:11 |
| tripul0noob | bug #1786109 seems to be a candidate for this. | 22:12 |
| openstack | bug 1786109 in tripleo "did not reach state "available", the state is "clean wait", error: None'" [Medium,Triaged] https://launchpad.net/bugs/1786109 - Assigned to wes hayutin (weshayutin) | 22:12 |
| *** tripul0noob is now known as xarlos | 22:28 | |
| *** xarlos is now known as tripul0noob | 22:28 | |
| tripul0noob | . | 22:28 |
| tonyb | This is kinda an overlap question with TripleO but apart from https://docs.openstack.org/tripleo-docs/latest/install/advanced_deployment/node_discovery.html are there any docs on using auto-discovery? | 22:39 |
| tonyb | I *kinda* thought if I configured tripleo/ironic for it all I really needed to do was boot a node and assuming is came up on the network the 'inspector' boot-target *should* work and have the node "do the right thing" but as I read $link it seems like I have the wrong idea | 22:40 |
| tonyb | tripul0noob: I'd say it'll end up in the same state | 22:41 |
| tonyb | tripul0noob: if this is a dev instance where security matters less you can disablr cleaning IIUC | 22:41 |
| tripul0noob | tonyb: It's a new deployment to go into lab. Looking at the result it seems /dev/sdb2 as a read only filesystem (it's a raid 1 config with just two disks) | 22:43 |
| tripul0noob | tonyb: There's nothing of any securty implications on the hardware at the moment. I would be happy to disable clean if I know how best to do this. | 22:44 |
| tonyb | tripul0noob: https://docs.openstack.org/tripleo-docs/latest/install/advanced_deployment/cleaning.html | 22:45 |
| tripul0noob | (I have re-attempted introspection, but alas the same result) | 22:45 |
| tonyb | tripul0noob: Basically set [DEFAULT]/clean_nodes = False in the undercloud and reinstall/update | 22:45 |
| tripul0noob | I thought that would be the case. Wasn't sure if there was a --no-clean parameter somewhere. | 22:46 |
| tripul0noob | That link you sent looks good (my google-fu seems to have taken me elsewhere!) so i'll see if I Can try a manual clean. If that fails, i'll reinstall the undercloud (seems like a big undertaking for such a small job) | 22:47 |
| tonyb | tripul0noob: Yeah | 22:47 |
| tonyb | tripul0noob: the *right* thing to do would eb to get the cleanign working but I certainly can't hepl with that | 22:47 |
| tonyb | tripul0noob: this is just a bandaid so you can make progress and then come back to this later | 22:48 |
| tonyb | tripul0noob: as you can do manual cleaning at any stage you and iterate on the hardware then until you figure out what the problem is | 22:48 |
| tripul0noob | I am attempting a manual clean now. | 22:50 |
| tripul0noob | Seems weird it's referring to a read only filesystem though. | 22:50 |
| tonyb | tripul0noob: Yeah It may bee your RAID config is b0rked somehow | 22:50 |
| *** rcernin has joined #openstack-ironic | 23:02 | |
| *** dsneddon has quit IRC | 23:03 | |
| tripul0noob | tonyb: Entirely possible. Been swapping stuff about. | 23:05 |
| tripul0noob | Though, not on this one that I know of. | 23:05 |
| tripul0noob | I wonder if previous UEFI deployments are getting in the way here. | 23:15 |
| tripul0noob | Figurative shotgun to the face aught to sort that out. | 23:16 |
| *** dsneddon has joined #openstack-ironic | 23:31 | |
| tripul0noob | well. That didn't seem to work either. | 23:45 |
| *** dsneddon has quit IRC | 23:45 | |
| tripul0noob | I'm going to bed. Will have a fresh look in the morning. | 23:45 |
| tripul0noob | Thanks for help, learned much as always. | 23:45 |
| *** tripul0noob has left #openstack-ironic | 23:46 | |
| mrda | Morning Ironic | 23:48 |
| *** sthussey has quit IRC | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!