*** bfournie has quit IRC | 00:03 | |
TheJulia | tonyb: please go ahead and submit a patch | 00:09 |
---|---|---|
TheJulia | anupn: if it detects yaml, it does a direct conversion to json for ansible's loader. | 00:09 |
TheJulia | anupn: I would guess your missing --- at the beginning of your json file | 00:10 |
TheJulia | err | 00:10 |
TheJulia | yaml file | 00:10 |
anupn | TheJulia: No it is there | 00:10 |
anupn | And when bifrost_inventory.py ran alone it is finding the inventory file and converting it to JSON | 00:11 |
anupn | so not sure when ran with ansible-playbook why it is unable to parse | 00:12 |
anupn | TheJulia: But yes it is saying that "home/ironic/bifrost/playbooks/inventory/bifrost_inventory.py: Syntax Error while loading YAML. did not | 00:13 |
anupn | find expected <document start>" | 00:13 |
TheJulia | I'd run it through a yaml validator | 00:14 |
anupn | TheJulia: I ran it, it says valid YAML | 00:15 |
anupn | Also there is a warning " Unable to parse /home/ironic/bifrost/playbooks/inventory/bifrost_inventory.py as an inventory | 00:18 |
anupn | source | 00:18 |
anupn | " | 00:18 |
TheJulia | oh.... | 00:38 |
TheJulia | hmm | 00:38 |
TheJulia | is it not executable? | 00:38 |
anupn | it is | 00:39 |
TheJulia | anupn: seems it is trying to read the file instead of execute it :\ | 00:39 |
TheJulia | that would do it :\ | 00:39 |
TheJulia | ironic-cores, I've submitted two forum sessions based upon the discussions on https://etherpad.openstack.org/p/DEN-train-ironic-brainstorming | 00:40 |
TheJulia | ironic-cores, largely they seemed to be the most "we need requirements, wants, and desires" items where as most of the other items seemed to be more technical deep-dive items. | 00:40 |
anupn | TheJulia: this is the output from the bifrost_inventory.py https://pastebin.com/cNcmZ1Jm | 00:40 |
TheJulia | maybe ansible changed its behavior for reading executable output at some point? | 00:41 |
TheJulia | ironic-cores and anyone else, if there are other topics that are necessary as requirements gathering/larger community setting items, please let me know ASAP.. | 00:42 |
*** bfournie has joined #openstack-ironic | 01:04 | |
anupn | TheJulia: Hmm may be good to check that how ansible executes an executable. But still I am not yet clear if it is getting rejected because of not finding a particular barmetal group or something else. Has anyone brought this up recently about dynamic enrollment? | 01:15 |
*** gyee has quit IRC | 01:15 | |
*** andrein has quit IRC | 01:15 | |
anupn | TheJulia: Can you think of anyone who has deployed using Bifrost? | 01:16 |
openstackgerrit | Merged openstack/ironic master: Document using a URL for image_checksum https://review.openstack.org/641614 | 01:22 |
*** whoami-rajat has joined #openstack-ironic | 01:23 | |
*** sthussey has quit IRC | 01:24 | |
*** anupn has quit IRC | 02:01 | |
*** yajushiSri has joined #openstack-ironic | 02:03 | |
openstackgerrit | Harald Jensås proposed openstack/ironic master: Initial processing of network port events https://review.openstack.org/633729 | 02:04 |
openstackgerrit | Harald Jensås proposed openstack/ironic master: WiP - Implement Event Handler in driver interfaces https://review.openstack.org/637840 | 02:04 |
openstackgerrit | Harald Jensås proposed openstack/ironic master: WIP - Cleaning network - events https://review.openstack.org/637841 | 02:04 |
openstackgerrit | Harald Jensås proposed openstack/networking-baremetal master: Enable ironic notifications https://review.openstack.org/576873 | 02:06 |
openstackgerrit | Harald Jensås proposed openstack/networking-baremetal master: DNM - Testing Neutron Events https://review.openstack.org/633610 | 02:07 |
openstackgerrit | Harald Jensås proposed openstack/networking-baremetal master: DNM - Testing Neutron Events https://review.openstack.org/633610 | 02:07 |
yajushiSri | I am an Outreachy applicant. Which task should I choose to begin working upon if I haven't got any experience with Ironic? https://storyboard.openstack.org/#!/story/2005114 | 02:09 |
*** hwoarang has quit IRC | 02:28 | |
*** hwoarang has joined #openstack-ironic | 02:30 | |
openstackgerrit | Merged openstack/ironic-tempest-plugin master: Deploy templates: API test nits https://review.openstack.org/641278 | 02:39 |
*** jhesketh has quit IRC | 02:45 | |
*** jhesketh has joined #openstack-ironic | 02:47 | |
*** rloo has quit IRC | 03:07 | |
*** yajushiSri has quit IRC | 03:11 | |
*** yajushiSri has joined #openstack-ironic | 03:18 | |
*** MattMan_ has joined #openstack-ironic | 03:23 | |
*** ianw_pto has quit IRC | 03:23 | |
*** ianw has joined #openstack-ironic | 03:23 | |
*** MattMan_1 has quit IRC | 03:24 | |
larsks | yajushiSri: the folks you need to ask will be around mostly during US business hours. | 03:27 |
larsks | Or european, actually. | 03:28 |
*** yaawang has quit IRC | 03:31 | |
yajushiSri | larsks, Okay, thanks! :D | 03:32 |
*** yaawang has joined #openstack-ironic | 03:36 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Implement basic interfaces for GraphicalConsole Interface https://review.openstack.org/547356 | 04:12 |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: [WIP] RPC changes for Graphical Console https://review.openstack.org/640827 | 04:12 |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: [WIP] Adds graphical console implementation for ilo drivers https://review.openstack.org/640973 | 04:20 |
*** whoami-rajat has quit IRC | 04:33 | |
*** whoami-rajat has joined #openstack-ironic | 04:53 | |
*** hwoarang has quit IRC | 05:02 | |
*** hwoarang has joined #openstack-ironic | 05:08 | |
*** Qianbiao has joined #openstack-ironic | 05:21 | |
Qianbiao | Hello | 05:21 |
Qianbiao | The code on master seems has issues for ironic-rootwrap feature | 05:22 |
Qianbiao | I try to deploy node with patch https://review.openstack.org/#/c/639288/. | 05:22 |
patchbot | patch 639288 - ironic - Add Huawei iBMC driver support - 7 patch sets | 05:22 |
Qianbiao | Result in error: Deploy failed for instance None. Error: Unexpected error while running command.\nCommand: sudo ironic-rootwrap /etc/ironic/rootwrap.conf iscsiadm -m discovery -t st -p 112.93.129.196:3260\nExit code: 1\nStdout: u''\nStderr: u'sudo: no tty present and no askpass program specified\\n | 05:23 |
Qianbiao | The command is ok in rocky branch. | 05:23 |
*** Dinesh_Bhor has quit IRC | 05:29 | |
*** stendulker has joined #openstack-ironic | 05:30 | |
openstackgerrit | Harald Jensås proposed openstack/networking-baremetal master: Enable ironic notifications https://review.openstack.org/576873 | 06:16 |
openstackgerrit | Harald Jensås proposed openstack/networking-baremetal master: DNM - Testing Neutron Events https://review.openstack.org/633610 | 06:17 |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Implement basic interfaces for GraphicalConsole Interface https://review.openstack.org/547356 | 06:19 |
*** Bhujay has joined #openstack-ironic | 06:52 | |
*** Qianbiao has quit IRC | 07:11 | |
openstackgerrit | Ilya Etingof proposed openstack/ironic master: Set boot_mode in node properties during Redfish introspection https://review.openstack.org/641357 | 07:12 |
openstackgerrit | Ilya Etingof proposed openstack/ironic master: Set boot_mode in node properties during Redfish introspection https://review.openstack.org/641357 | 07:19 |
*** cjloader has quit IRC | 07:22 | |
*** hwoarang has quit IRC | 07:25 | |
*** pcaruana has joined #openstack-ironic | 07:26 | |
*** Qianbiao has joined #openstack-ironic | 07:29 | |
*** hwoarang has joined #openstack-ironic | 07:30 | |
arne_wiebalck | good morning, ironic! | 07:36 |
*** cjloader has joined #openstack-ironic | 07:39 | |
etingof | good Friday morning arne_wiebalck o/ | 07:45 |
arne_wiebalck | hey etingof o/ | 07:45 |
*** gkadam__ has joined #openstack-ironic | 07:50 | |
*** Bhujay has quit IRC | 07:50 | |
arne_wiebalck | Qianbiao: I think this error indicates that the executing user does not have password less access to the command. | 07:55 |
*** Bhujay has joined #openstack-ironic | 07:57 | |
*** Qianbiao has quit IRC | 08:01 | |
rpittau|afk | good morning ironic! o/ | 08:06 |
*** rpittau|afk is now known as rpittau | 08:06 | |
arne_wiebalck | good morning rpittau o/ | 08:08 |
rpittau | hey arne_wiebalck :) | 08:09 |
openstackgerrit | Merged openstack/ironic-inspector master: introspection data backend: migration tool https://review.openstack.org/625769 | 08:09 |
*** Qianbiao has joined #openstack-ironic | 08:15 | |
Qianbiao | arne_wiebalck I am not sure about why. Everything works on rocky branch. | 08:15 |
Qianbiao | And if i delete ironic commands under /usr/local/bin, everything works again. | 08:16 |
Qianbiao | There are ironic commands created by old ironic version under /usr/bin | 08:16 |
Qianbiao | ironic-rootwrap is created under /usr/local/bin i think this is the reason | 08:17 |
Qianbiao | arne_wiebalck https://review.openstack.org/#/c/639288/ this patch is waiting for review. | 08:21 |
patchbot | patch 639288 - ironic - Add Huawei iBMC driver support - 7 patch sets | 08:21 |
arne_wiebalck | When you say rocky, you use the same machine (same installation/config of ironic), you really only switch branches? | 08:23 |
* arne_wiebalck checking the pacth | 08:23 | |
*** amoralej|off is now known as amoralej | 08:24 | |
* arne_wiebalck was not expecting a 3k line patch :) | 08:24 | |
rpittau | there's an svg a bit big in that patch | 08:25 |
arne_wiebalck | I can review that! ;) | 08:25 |
rpittau | lol ok that's all your! :D | 08:25 |
arne_wiebalck | lol | 08:25 |
rpittau | down to only 1900 :D | 08:27 |
arne_wiebalck | Qianbiao: just for my understanding, is the patch related to the problem you see or was this an independent remark? | 08:27 |
arne_wiebalck | So, when you say rocky works for you, but master does not: is that vanilla ironic (or is that vanilla rocky vs master + patch)? | 08:28 |
*** tssurya has joined #openstack-ironic | 08:28 | |
Qianbiao | yes, rocky vs master+patch | 08:29 |
arne_wiebalck | does it work for you when you compare rocky vs master ? | 08:29 |
Qianbiao | Rocky will generate ironic-rootwrap under /usr/bin | 08:29 |
Qianbiao | master /usr/local/bin | 08:30 |
Qianbiao | i think this is the reason | 08:30 |
Qianbiao | rootwrap.conf has a line: exec_dirs=/sbin,/usr/sbin,/bin,/usr/bin | 08:30 |
*** Bhujay has quit IRC | 08:31 | |
*** sburette has joined #openstack-ironic | 08:34 | |
rpittau | Qianbiao, are you using 2 separate environments, one for master and one for rocky ? | 08:38 |
iurygregory|home | morning everyone o/ | 08:39 |
rpittau | hey iurygregory|home :) | 08:40 |
*** iurygregory|home is now known as iurygregory | 08:40 | |
*** andrein has joined #openstack-ironic | 08:41 | |
*** andrein has quit IRC | 08:55 | |
openstackgerrit | Riccardo Pittau proposed openstack/ironic-tempest-plugin master: Marking ironic-standalone job voting again https://review.openstack.org/641967 | 08:57 |
iurygregory | yay | 08:58 |
arne_wiebalck | Qianbiao: Sorry, got distracted. | 09:00 |
rpittau | that's ready as soon the other patch is merged :) | 09:01 |
arne_wiebalck | Qianbiao: Where is that configuration generated? | 09:01 |
arne_wiebalck | Qianbiao: Also, it'd still be good to compare vanilla rocky and master (to determine whether the issue is coming with your patch). | 09:02 |
Qianbiao | <arne_wiebalck> /etc/ironic/rootwrap.conf | 09:04 |
arne_wiebalck | Also, rpittau touches a good point: if rocky is installed via some packages, while master is done in another way, the difference may come from this (rather than the code base). | 09:04 |
arne_wiebalck | This file is the same in rocky and master, no? | 09:05 |
Qianbiao | it's same | 09:05 |
Qianbiao | but the locate ironic-rootwrap is generated in different folder | 09:06 |
arne_wiebalck | Qianbiao: Right, but why does it end up in different places? | 09:06 |
Qianbiao | Not sure. ;( | 09:07 |
arne_wiebalck | Qianbiao: That's what I meant by "where is the configuration generated?" | 09:07 |
arne_wiebalck | Qianbiao: Sorry if that was ambiguous. | 09:07 |
Qianbiao | I did not look into the code. Just desc the situation. I thought maybe u knows the reason | 09:08 |
*** dougsz has joined #openstack-ironic | 09:09 | |
arne_wiebalck | Qianbiao: My initial thought was that this is done by the package/installer (not ironic itself). | 09:09 |
Qianbiao | i setup master with `python setup.py install` | 09:09 |
*** pcaruana has quit IRC | 09:12 | |
arne_wiebalck | and rocky? | 09:13 |
Qianbiao | rocky is setup with an ansible playbook, will check now | 09:15 |
openstackgerrit | Digambar proposed openstack/ironic stable/ocata: Fix OOB introspection to use pxe_enabled flag in idrac driver https://review.openstack.org/641683 | 09:16 |
arne_wiebalck | I'd guess that the playbook "does the right thing"(TM) | 09:16 |
arne_wiebalck | Qianbiao: Unless someone with more knowledge in this area jumps in, I think the best way forward is to reduce the differences between the rocky and master install, and take it from there. | 09:17 |
Qianbiao | thanks, i can fix it by delete scripts under /usr/local/bin for now. | 09:19 |
rpittau | Qianbiao, that doesn't make sense to me though, removing the script from /usr/local/bin shouldn't influence the rest, if it's true that it can't be found | 09:20 |
Qianbiao | It does fix my issue. | 09:21 |
Qianbiao | because there are some cmd under /usr/bin created by rocky version. | 09:21 |
Qianbiao | I thought maybe i install master with `sudo python setup.py install` cause the issue. | 09:22 |
Qianbiao | sudo will install commands to /usr/local/bin | 09:23 |
Qianbiao | but /usr/local/bin is not config in rootwrap.conf ? | 09:23 |
rpittau | as arne_wiebalck suggested already, you should separate as much as possible the 2 branch installations | 09:23 |
rpittau | Qianbiao, that's my point, it should not change the behavior removing the script from /usr/local/bin | 09:24 |
Qianbiao | chajian@chajian:/usr/local/bin$ cat /etc/ironic/rootwrap.conf | 09:25 |
Qianbiao | # Configuration for ironic-rootwrap | 09:25 |
Qianbiao | # This file should be owned by (and only writable by) the root user | 09:25 |
Qianbiao | [DEFAULT] | 09:25 |
Qianbiao | # List of directories to load filter definitions from (separated by ','). | 09:25 |
Qianbiao | # These directories MUST all be only writable by root ! | 09:25 |
Qianbiao | filters_path=/etc/ironic/rootwrap.d,/usr/share/ironic/rootwrap | 09:25 |
Qianbiao | # List of directories to search executables in, in case filters do not | 09:25 |
Qianbiao | # explicitly specify a full path (separated by ',') | 09:25 |
Qianbiao | # If not specified, defaults to system PATH environment variable. | 09:25 |
Qianbiao | # These directories MUST all be only writable by root ! | 09:25 |
Qianbiao | exec_dirs=/sbin,/usr/sbin,/bin,/usr/bin,/usr/local/bin | 09:25 |
Qianbiao | check the last line, i add /usr/local/bin | 09:25 |
rpittau | Qianbiao, can you please use a paste service for that ? :) like --> http://paste.openstack.org/ | 09:26 |
*** pcaruana has joined #openstack-ironic | 09:27 | |
Qianbiao | sorry | 09:27 |
Qianbiao | http://paste.openstack.org/show/747441/ | 09:27 |
Qianbiao | this is the rootwrap conf file | 09:27 |
Qianbiao | it will not wrap /usr/local/bin by default. | 09:28 |
Qianbiao | by the script created by `sudo python setup.py install` will be in /usr/local/bin | 09:28 |
rpittau | Qianbiao, the exec_dirs entry is overridden by the filters_path, as far as you have /usr/local/bin in your path is enough to execute the ironic-rootwrap script from there | 09:28 |
Qianbiao | I add it just a minute ago. | 09:29 |
Qianbiao | Not the time issue happened | 09:29 |
Qianbiao | I will create a new clean env to reproduce it. | 09:30 |
rpittau | Qianbiao, it doesn't matter if /usr/local/bin is not in rootwrap.conf | 09:31 |
rpittau | Qianbiao, the problem you're facing now is a conflict between versions in the same environment | 09:32 |
rpittau | Qianbiao, you need to do 2 separate installations | 09:32 |
Qianbiao | Ok will test in a new env | 09:34 |
*** fdegir has quit IRC | 09:34 | |
*** ijpascual has quit IRC | 09:34 | |
*** hrw has joined #openstack-ironic | 09:36 | |
hrw | morning | 09:36 |
csmart | is there any way to insert metadata into an instance providing the baremetal details, for example rack location? | 09:42 |
iurygregory | csmart, you can update the extra field for the baremetal node i think =) | 09:44 |
openstackgerrit | Marcin Juszkiewicz proposed openstack/ironic master: driver-requirements: mark UcsSdk as Python 2 only https://review.openstack.org/641972 | 09:44 |
hrw | this change allows me to build ironic-conductor image in Kolla using only Python 3 in image | 09:45 |
iurygregory | csmart, https://developer.openstack.org/api-ref/baremetal/?expanded=update-node-detail | 09:46 |
csmart | iurygregory: ahh with `openstack baremetal node set --extra key=pair` ? sweet I'll check it out, thanks | 09:46 |
iurygregory | csmart, yeah =) | 09:47 |
csmart | cool, thanks.. my google-fu was failing me | 09:47 |
iurygregory | you can add any information for the node | 09:47 |
iurygregory | it happens =) | 09:47 |
arne_wiebalck | csmart: that's for the node, not the instance, though | 09:47 |
*** rcernin has quit IRC | 09:47 | |
*** blinkiz has joined #openstack-ironic | 09:47 | |
*** ijpascual has joined #openstack-ironic | 09:48 | |
*** fdegir has joined #openstack-ironic | 09:48 | |
csmart | arne_wiebalck: ahh k.. yeah I was kinda hoping I could provide some information to the users of the instance about the physical node | 09:49 |
blinkiz | Hello. Am trying to understand introspection rules/matchning. I would like to find baremetal servers with nvme disk which will indicate for me that this server should be Ceph server. | 09:49 |
*** andrein has joined #openstack-ironic | 09:49 | |
blinkiz | Under ".inventory.disks" in the json tree I can see all nvme disks. But I can not understand how I should match my rules against it | 09:50 |
iurygregory | csmart, you can also update the server metadata i think | 09:50 |
csmart | iurygregory: yeah I was thinking I could do something like that with a script post instance deploy... won't get it on config drive but that's OK so long as it's there with a curl against metadata server.. I'll play with that, thanks | 09:51 |
iurygregory | csmart, good luck o/ | 09:51 |
csmart | cheers | 09:52 |
rpittau | blinkiz, I suggest to check conditions here: https://docs.openstack.org/ironic-inspector/latest/user/usage.html#introspection-rules | 09:54 |
*** jistr is now known as jistr|biab | 10:21 | |
openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: DNM - Test legacy with Bionic https://review.openstack.org/641671 | 10:41 |
*** jistr|biab has quit IRC | 10:51 | |
*** jistr has joined #openstack-ironic | 10:54 | |
*** dtantsur|afk is now known as dtantsur | 11:08 | |
dtantsur | morning ironic | 11:08 |
*** hrw has left #openstack-ironic | 11:11 | |
rpittau | hi dtantsur :) | 11:12 |
arne_wiebalck | dtantsur: o/ | 11:12 |
iurygregory | morning dtantsur | 11:18 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: Add python3 unit test with drivers installed https://review.openstack.org/641984 | 11:20 |
*** stendulker has quit IRC | 11:31 | |
dtantsur | mgoddard: won't work because of ucs.. | 11:33 |
mgoddard | dtantsur: it's based on a patch that excludes UCS for py3 | 11:33 |
dtantsur | ah! | 11:33 |
dtantsur | mgoddard: but I suspect IMC has the same issue | 11:33 |
mgoddard | seems to work | 11:33 |
mgoddard | locally at least | 11:33 |
dtantsur | okay, I'll wait for the CI before approving the first one | 11:34 |
rpittau | mgoddard, dtantsur, I would mark imc also python2 | 11:44 |
openstackgerrit | Rachit Kapadia proposed openstack/ironic master: Set boot_mode in node properties during OOB Introspection https://review.openstack.org/639698 | 11:44 |
rpittau | the version in the reqs is legacy and doesn't support py3 | 11:44 |
rpittau | not officially | 11:44 |
rpittau | there was an effort to use a more recent version but it's not moving forward since a while now https://review.openstack.org/#/c/598194/ | 11:50 |
patchbot | patch 598194 - ironic - Upgrade to IMCSDK v0.9.3.1 - 1 patch set | 11:50 |
*** gkadam__ has quit IRC | 12:02 | |
*** gkadam has joined #openstack-ironic | 12:04 | |
*** Qianbiao has quit IRC | 12:07 | |
openstackgerrit | Ilya Etingof proposed openstack/sushy master: Fix exposed UpdateService constants https://review.openstack.org/641989 | 12:09 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Reset cached extension managers in the unit tests https://review.openstack.org/641990 | 12:19 |
dtantsur | this ^^ will make RDO people a bit happier | 12:20 |
*** yaawang has quit IRC | 12:21 | |
*** yaawang has joined #openstack-ironic | 12:22 | |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Stop requiring root_gb for whole-disk images https://review.openstack.org/641993 | 12:29 |
dtantsur | and this will make me a bit happier ^^^ | 12:29 |
openstackgerrit | Ilya Etingof proposed openstack/sushy master: Make UpdateService.simple_update() operational https://review.openstack.org/642002 | 12:48 |
jroll | morning ironic, and TGIF! | 12:48 |
dtantsur | morning jroll \o/ | 12:49 |
jroll | \o/ | 12:49 |
dtantsur | jroll: do you remember why we're requiring root_gb for whole disk images? I could not find any traces of usage. | 12:49 |
jroll | dtantsur: I would guess so that we can make sure the disk is large enough, no? | 12:50 |
jroll | or is there a size property we use for that | 12:50 |
jroll | oh instance_inf | 12:50 |
dtantsur | jroll: we don't do these checks for whole-disk images | 12:50 |
dtantsur | yep, you're thinking about properties/local_gb | 12:51 |
jroll | yeah I have no clue why we'd require that | 12:51 |
dtantsur | (which we don't check either for whole-disk images) | 12:51 |
dtantsur | okay, good. just wanted to double-check with someone :) | 12:51 |
jroll | the only reason I can think of is because it would be in the nova flavor anyway, so make sure it's passed over | 12:51 |
jroll | but that doesn't make sense anymore :) | 12:51 |
*** iurygregory is now known as iurygregory_mtg | 12:53 | |
openstackgerrit | Ilya Etingof proposed openstack/sushy master: Make UpdateService.simple_update() operational https://review.openstack.org/642002 | 12:55 |
iurygregory_mtg | morning jroll o/ | 12:56 |
dtantsur | jroll: another sanity check please: does preserve_ephemeral make any sense for whole-disk images? | 12:57 |
jroll | erm | 12:58 |
jroll | so I would think there's a use case for putting "ephemeral" on disks other than the root | 12:59 |
jroll | but I see your patch and we don't allow setting ephemeral_mb for whole disk | 12:59 |
jroll | dtantsur: ^ | 12:59 |
dtantsur | jroll: right. we also do not support "disks other than the root" | 12:59 |
dtantsur | but maybe somebody could create an ephemeral partition via e.g. cloud-init | 13:00 |
jroll | dtantsur: right, though it's a valid use case imo | 13:00 |
dtantsur | so maybe it makes sense to support it | 13:00 |
jroll | we actually have downstream crap that does something like "preserve ephemeral" | 13:00 |
jroll | but it's really just "don't wipe non-root disks during rebuild" | 13:00 |
jroll | or something | 13:00 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Stop requiring root_gb for whole-disk images https://review.openstack.org/641993 | 13:01 |
dtantsur | okay, I'll keep it in ^^^ | 13:01 |
* dtantsur tests locally again | 13:01 | |
dtantsur | I need to mention that our instance_info handling is an ungodly mess >_< | 13:03 |
*** rh-jelabarre has joined #openstack-ironic | 13:04 | |
jroll | yes. | 13:04 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: Add python3 unit test with drivers installed https://review.openstack.org/641984 | 13:10 |
kandi | dtantsur: I think I'm finished running the script for building tinyipa ramdisk and got `tinyipa.vmlinuz` and `tinyipa.gz` | 13:19 |
kandi | during the process I got some Permission denied msgs like this | 13:19 |
kandi | Connecting to repo.tinycorelinux.net (89.22.99.37:80) | 13:20 |
kandi | liblvm2.tcz 100% |***********************************************************************************************************************************************************| 1252k 0:00:00 ETA | 13:20 |
kandi | liblvm2.tcz: OK | 13:20 |
kandi | touch: /usr/local/tce.installed/liblvm2: Permission denied | 13:20 |
kandi | but it still finished running the script, is this a message I can ignore? | 13:20 |
* etingof would look into the script to see what is happening there | 13:22 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic master: WIP: move all testing to bionic https://review.openstack.org/642015 | 13:22 |
* jroll hopes he did that right but isn't 100% sure | 13:22 | |
jroll | iurygregory_mtg: ^ fyi, I'm working on ironic there (I know you've touched some of this) | 13:23 |
*** iurygregory_mtg is now known as iurygregory | 13:24 | |
iurygregory | jroll, ack | 13:24 |
*** pcaruana has quit IRC | 13:30 | |
*** jcoufal has joined #openstack-ironic | 13:30 | |
openstackgerrit | Dmitry Tantsur proposed openstack/metalsmith master: Ensure instance_info is clean before deploy and after failure https://review.openstack.org/642016 | 13:32 |
dtantsur | This ^^ is another part of "everything around instance_info is a mess" story | 13:34 |
dtantsur | kandi: yes, I've seen these messages, and I've ignored them safely | 13:37 |
iurygregory | dtantsur, do we have a book with all storys? XD | 13:37 |
dtantsur | we should, but it's going to be a sad and depressing book | 13:37 |
kandi | dtantsur: I've found that there is no `make clean_tinyipa` in the Makefile, should I take that off the documentation? | 13:37 |
iurygregory | XD | 13:38 |
dtantsur | kandi: probably. maybe there is a typo in the name? please check the Makefile | 13:38 |
* dtantsur goes to lunch | 13:38 | |
kandi | I checked it here: https://github.com/openstack/ironic-python-agent/blob/master/imagebuild/tinyipa/Makefile | 13:39 |
*** derekh has joined #openstack-ironic | 13:53 | |
* etingof would look back in git time to see the context - why this option appeared/disappeared | 13:53 | |
kandi | git time XD | 13:55 |
larsks | etingof: did you see the question from yajushiSri last night re: outreachy and https://storyboard.openstack.org/#!/story/2005114? | 13:58 |
TheJulia | good morning | 13:59 |
larsks | morning! | 13:59 |
iurygregory | morning TheJulia o/ | 14:02 |
*** amoralej is now known as amoralej|lunch | 14:03 | |
openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Update python-hardware requirement for py3 https://review.openstack.org/642022 | 14:04 |
bfournie | good morning, now that https://review.openstack.org/#/c/626552/ has merged, is this OK to backport to Rocky/Queens? | 14:05 |
patchbot | patch 626552 - ironic-inspector - Use getaddrinfo instead of gethostbyname while res... (MERGED) - 10 patch sets | 14:05 |
dtantsur | etingof: we should leave a note on our projects to ask people to write emails.. | 14:07 |
dtantsur | at least for the initial introduction | 14:07 |
etingof | larsks, sorry, I did not and I do still do not see it there | 14:10 |
etingof | dtantsur, I will update the projects with a note | 14:11 |
larsks | etingof: yajushiSri asked, "I am an Outreachy applicant. Which task should I choose to begin working upon if I haven't got any experience with Ironic?" | 14:12 |
larsks | I told them to come back during us/european work hours :) | 14:12 |
kandi | etingof: I looked back in git time and it was in the documentation from the first version but never implemented | 14:13 |
kandi | should I try implement it or just remove it from the documentation? | 14:14 |
yajushiSri | etingof: I also messaged this on #openstack-outreachy - I have a question regarding Project: Improve Ironic Agent Image Builder. Are we expected to implement one of the features described here: https://storyboard.openstack.org/#!/story/2005114 , as part of the application process? As they are listed under the 'Internship Tasks' section on project page, as well as the link for contributions corresponds to the same tasks. Project | 14:16 |
yajushiSri | description: https://www.outreachy.org/may-2019-august-2019-outreachy-internships/communities/openstack/#improve-ironic-agent-image-builder | 14:16 |
etingof | larsks, we try to tag suitable tasks with the 'outreachy' tag on the storyboard | 14:16 |
etingof | larsks, but the best thing would be for yajushiSri to e-mail us | 14:17 |
larsks | etingof: well, there he is right now ^^^^ | 14:17 |
yajushiSri | she** :D | 14:17 |
larsks | Yeah, sorry, I spotted that after hitting enter. | 14:17 |
yajushiSri | etingof, Noted! | 14:18 |
yajushiSri | larsks, No worries. | 14:18 |
etingof | yajushiSri, larsks I think the goal of the application process is not so much to deliver a feature, but to 1) get a feeling if this project is a good fit for you and 2) prove yourself as a capable contributor | 14:18 |
etingof | so I think you could start working on one of the listed features and see where it comes to | 14:19 |
yajushiSri | Got it! Thanks. :D | 14:20 |
etingof | also we maintain a list of relatively easy fixes to ironic, likely irrelevant to the listed projects -- https://storyboard.openstack.org/#!/story/list?status=active&tags=low-hanging-fruit&project_group_id=75 | 14:20 |
yajushiSri | Will mail you an introduction. xD | 14:21 |
dtantsur | TheJulia, jroll, new crazy idea: https://storyboard.openstack.org/#!/story/2005167 | 14:22 |
TheJulia | ask again in like 15 minutes | 14:22 |
yajushiSri | etingof, Yea, but if the tag:outreachy ones will be considered for the application process, my approach should be to understand Ironic and its workflow and work on those features, right? | 14:22 |
etingof | retry/reboot is the cornerstone of IT | 14:24 |
etingof | yajushiSri, tagged tasks could be started and partially finished during the application process. it depends on the task - they differ in man-months | 14:26 |
dtantsur | bfournie: I have a feeling that this patch is wrong and should be reverted | 14:26 |
yajushiSri | etingof, okay, thanks. :D | 14:27 |
bfournie | dtantsur: oh | 14:27 |
TheJulia | dtantsur: so not pxe or ipxe, but still retry pxe? | 14:27 |
dtantsur | TheJulia: well, it all starts as pxe boot anyway | 14:27 |
TheJulia | so filter nodes with those interfaces | 14:27 |
TheJulia | okay | 14:28 |
TheJulia | as in include them | 14:28 |
dtantsur | yeah (bad wording) | 14:29 |
kandi | etingof: Can I try to push a patch and also setup my OpenStack development environment through this process? | 14:32 |
*** pcaruana has joined #openstack-ironic | 14:37 | |
kandi | yajushiSri: I'm also an applicant, hi!!!!!!! | 14:39 |
rpioso | mgoddard: Are you available to chat today? | 14:39 |
mgoddard | rpioso: yes, now? | 14:40 |
rpioso | mgoddard: Sounds good :-) | 14:40 |
yajushiSri | kandi, Hey! :D | 14:40 |
mgoddard | rpioso: have you had a look at the patch I proposed recently? I added you as a reviewer | 14:41 |
rpioso | mgoddard: It's towards the top of my list. I'll try to get to it today. | 14:41 |
rpioso | Thank you for proposing it. | 14:41 |
mgoddard | rpioso: I'm going to pick out the redfish stuff into a separate patch, you can ignore that part | 14:42 |
mgoddard | rpioso: the idea is that we add a new deploy step to the base RAID interface that subclasses can advertise support for | 14:42 |
mgoddard | rpioso: to support it, their create_configuration method needs to return states.DEPLOYWAIT for a deploy step, and the driver needs to handle resuming deployment when the configuration has been applied | 14:43 |
mgoddard | I'm using the drac module's periodic task for that, the same as the clean step | 14:43 |
rpioso | mgoddard: Got it. | 14:47 |
rpioso | mgoddard: I see that the idrac's RAID delete_configuration() method has been annotated, but its create_configuration() hasn't. Is that an oversight or am I missing something? | 14:47 |
arne_wiebalck | etingof: For the IPA builder: nice to see this, wasn't aware! We currently build our deploy images from inside a docker container which has all the required dependencies (images are DIB-built, based on Centos7). | 14:48 |
arne_wiebalck | etingof: We've also added a mechanism to git clone a branch of our hw manager from inside the image, so we do not need to rebuild the image when doing changes. | 14:49 |
mgoddard | rpioso: for applying configuration, I've added apply_configuration() to the base class | 14:49 |
mgoddard | it allows you to set target_raid_configuration and call create_configuration in a single step | 14:50 |
arne_wiebalck | etingof: This way, we build per use case / development images which pull in the corresponding branch. | 14:50 |
*** efried is now known as fried_rice | 14:51 | |
rpioso | mgoddard: I see. Doesn't it need the annotation? | 14:51 |
mgoddard | rpioso: which one? apply_configuration? | 14:51 |
rpioso | mgoddard: y | 14:52 |
mgoddard | rpioso: I haven't annotated it because not all drivers will support it | 14:52 |
mgoddard | if it's supported the driver sets SUPPORTS_APPLY_CONFIGURATION_DEPLOY_STEP, which causes the base class to return apply_configuration in get_deploy_steps() | 14:53 |
rpioso | mgoddard: Tricky ;-) Seems a bit roundabout to me, but it's a detail. | 14:55 |
openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Fix lookup when ipmi_address is a hostname https://review.openstack.org/642039 | 14:55 |
dtantsur | TheJulia, bfournie ^^^ | 14:55 |
bfournie | dtantsur: great! | 14:56 |
rpioso | mgoddard: I suppose a new apply_configuration abstract method could be defined and implemented by those drivers that support it. They could leverage a base class helper method implementation. | 14:57 |
mgoddard | rpioso: I guess, but drivers won't automatically support that step. I guess an alternative would be to provide a common method then let drivers define their own deploy step | 14:57 |
rpioso | mgoddard: Jinx! | 14:57 |
mgoddard | rpioso: I quite liked that we only need to define argsinfo in one place with this method. I guess we could make it common somehow | 14:58 |
jroll | dtantsur: re: retrying pxe, rather than checking all of these things in the periodic task, why not make use of node.fault? | 14:59 |
dtantsur | jroll: how would node.fault help here? | 14:59 |
rpioso | mgoddard: By argsinfo, are you referring to the docstring for apply_configuration's params? | 14:59 |
jroll | dtantsur: mark it as a pxe failure, instead of checking all nodes for timeouts and boot interface | 15:00 |
mgoddard | rpioso: yes | 15:00 |
dtantsur | jroll: but how do we know if a node is in PXE failure without checking timeouts? | 15:00 |
* rpioso feels the need for more caffeine | 15:00 | |
etingof | kandi, I think that would be useful | 15:00 |
jroll | dtantsur: set node.fault to "pxe failure", I should say | 15:00 |
jroll | dtantsur: we already check these timeouts, right? | 15:00 |
dtantsur | jroll: right, but in response to what? | 15:00 |
dtantsur | well, we have a generic deploy timeout | 15:01 |
dtantsur | we may try calling boot.prepare_ramdisk again, but it may be more complex | 15:01 |
jroll | dtantsur: right, when we hit whatever produces "timeout waiting for callback", if it's using pxe/ipxe, mark it as a pxe failure | 15:02 |
jroll | and then use that as the key in the new task | 15:02 |
rpioso | mgoddard: That's what the base create_configuration() and delete_configuration() do. I misspoke earlier. apply_configuration could not be an abstract method, because not all drivers will support it. It could have a base class implementation that raises a not supported exception. | 15:02 |
dtantsur | jroll: that will make conductor aware of the PXE boot interface, right? | 15:02 |
jroll | dtantsur: doesn't safe much effort, but it's worth doing IMO as operators might appreciate that filter | 15:02 |
dtantsur | in a sense of hardcoding it? | 15:02 |
jroll | I guess? | 15:03 |
jroll | you could always say node.driver.boot.powered_by_pxe or something | 15:03 |
dtantsur | mmmmmm.. sounds like a leaky abstraction | 15:03 |
dtantsur | even though I think fault = "pxe failure" is a great idea | 15:03 |
dtantsur | I need to think more about it | 15:04 |
rpioso | mgoddard: Anyways, thank you for the overview. Very helpful. I will try to review it today. | 15:04 |
dtantsur | jroll: mind putting your thoughts as a comment to the story? | 15:04 |
jroll | dtantsur: thanks, worth a thought. we've considered adding a retry as well | 15:05 |
rpioso | mgoddard: If memory serves, you previously asked about the idrac's support of the BIOS interface. | 15:05 |
jroll | sure | 15:05 |
rpioso | mgoddard: The direction is suggested by https://storyboard.openstack.org/#!/story/2004592 | 15:05 |
rpioso | mgoddard: In the meantime, idrac does offer a vendor_passthru interface for applying BIOS setting config. It may be possible to create a deploy step that leverages that. | 15:07 |
mgoddard | rpioso: I think we'd need something to poll the jobs in the LC, much like we do for RAID | 15:11 |
rpioso | mgoddard: That's available through the idrac's vendor_passthru interface. | 15:11 |
rpioso | https://github.com/openstack/ironic/blob/77ee49f21ce703ed4235a0799241f0b8b283b68b/ironic/drivers/modules/drac/vendor_passthru.py#L150 | 15:12 |
TheJulia | rpioso: I'd highly prefer a proper raid interface... fwiw. | 15:14 |
dtantsur | folks, can any core please check https://review.openstack.org/#/c/641376/ ? it recovers our standalone job | 15:14 |
patchbot | patch 641376 - ironic - Add option to set python interpreter for ansible - 15 patch sets | 15:14 |
*** amoralej|lunch is now known as amoralej | 15:14 | |
mgoddard | TheJulia: this is for BIOS - RAID exists for drac | 15:14 |
mgoddard | TheJulia: but yes, agreed | 15:15 |
TheJulia | \o/ | 15:15 |
rpioso | mgoddard: AFAIU, the present Redfish BIOS interface will not work against the iDRAC. The iDRAC's Redfish service has required an OEM action to create a configuration job in order to apply the requested settings. | 15:15 |
mgoddard | rpioso: I see, makes sense | 15:15 |
mgoddard | rpioso: I saw some extension points on the Redfish BIOS driver - post_configure, post_reboot | 15:16 |
rpioso | mgoddard: However, that may change. There's a standard Redfish annotation that can be used to specify when to apply the settings, @Redfish.SettingsApplyTime. | 15:16 |
rpioso | mgoddard: However, neither sushy nor the Redfish BIOS interface presently support that annotation. | 15:17 |
TheJulia | pxe boot interfaces do have capabilities tags.... | 15:17 |
rpioso | mgoddard, TheJulia: That would be a proper BIOS interface :-) | 15:17 |
rpioso | mgoddard: Must the functionality exist in Stein for you to demo it at the Summit? | 15:18 |
mgoddard | rpioso: I'd hoped it would, but I can't see us getting BIOS support in | 15:19 |
mgoddard | rpioso: RAID might be possible | 15:19 |
mgoddard | rpioso: I'd be happy enough with that | 15:19 |
rpioso | mgoddard: Yeah, BIOS won't land in Stein. | 15:19 |
TheJulia | proper bios interface for idrac would be <3 | 15:20 |
dtantsur | ++ | 15:20 |
mgoddard | rpioso: I'd consider using master or a custom driver if necessary | 15:20 |
mgoddard | +1. Seems like all the pieces are in place | 15:21 |
dtantsur | I think using master will send a better message | 15:21 |
dtantsur | "in only a few months you'll be able to: <..>" | 15:21 |
rpioso | mgoddard: Sound like plausible options. | 15:22 |
rpioso | etingof: ^^^ sushy needs @Redfish.SettingsApplyTime | 15:22 |
dtantsur | etingof: look, a new outreach task ^^^! | 15:23 |
dtantsur | :) | 15:23 |
larsks | Is ironic ci stable? I see failures for https://review.openstack.org/#/c/640930/ that look like timeouts unrelated to the change. | 15:23 |
patchbot | patch 640930 - ironic - honor ipmi_port in serial console drivers - 6 patch sets | 15:23 |
rpioso | mgoddard: I'll be focused on that story -- https://storyboard.openstack.org/#!/story/2004592 | 15:23 |
dtantsur | larsks: relatively. sometimes it fails. | 15:23 |
dtantsur | larsks: the standalone job has been temporary disabled, you can recheck | 15:23 |
* etingof needs to dig into the email thread on @Redfish.SettingsApplyTime | 15:23 | |
rpioso | mgoddard: It could offer what you need if Redfish BIOS with @Redfish.SettingsApplyTime is available, too. | 15:24 |
larsks | dtantsur: it also failed ironic-tempest-functional-python2 with "Conflict with state of target resource". | 15:24 |
larsks | I'll recheck and cross my fingers. | 15:24 |
mgoddard | I'm going to focus on RAID for now. If I have time to build BIOS support I'll include that also | 15:24 |
dtantsur | larsks: yes, this one is also transient. I haven't been able to figure it out yet.. | 15:25 |
*** baha has joined #openstack-ironic | 15:25 | |
rpioso | mgoddard: I'll focus on that story and twisting etingof's arm ;-) | 15:25 |
mgoddard | rpioso: in terms of roadmap, do you expect to keep both the wsman drivers and redfish around? | 15:26 |
mgoddard | rpioso: as in, if I glued together a wsman-based BIOS interface from existing pieces, would it be useful in the long term? | 15:26 |
rpioso | mgoddard: There will be a single idrac h/w type composed of both WS-Man and Redfish interface implementations. The operator gets to choose. | 15:27 |
rpioso | mgoddard: Absolutely useful! | 15:27 |
mgoddard | rpioso: ok. I'll consider what would be required, and how much of the raid job monitoring could be reused | 15:27 |
rpioso | mgoddard: Cool! | 15:28 |
rpioso | Do you have anything else regarding the demo? | 15:28 |
mgoddard | rpioso: I don't think so at this point. Thanks for the help | 15:29 |
rpioso | mgoddard: You're most welcome. Have a great weekend :-) | 15:29 |
mgoddard | rpioso: Thanks, you too | 15:29 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: WIP: Add iDRAC RAID deploy steps https://review.openstack.org/641731 | 15:31 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: Deploy Templates: factor out ironic.conductor.steps https://review.openstack.org/642059 | 15:31 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: WIP: Add deploy steps for Redfish BIOS interface https://review.openstack.org/642060 | 15:31 |
mgoddard | dtantsur: ^^^ I put together that conductor.steps refactor you requested | 15:31 |
dtantsur | \o/ | 15:31 |
mgoddard | seems cleaner | 15:32 |
TheJulia | woo a +2 on fast tracked deployments | 15:32 |
TheJulia | a follow-up needed, but easy :) | 15:32 |
TheJulia | once I have braincells and time in my life_of_meetings | 15:32 |
TheJulia | #meetinglife ? | 15:33 |
dtantsur | heh | 15:34 |
TheJulia | I looked at my schedule for next week and cried some | 15:35 |
TheJulia | like 16 hours of meetings or something stupid | 15:35 |
TheJulia | (if anyone wonders why I'm always so async.... this is why) | 15:35 |
dtantsur | :( | 15:36 |
openstackgerrit | Vlad Sorokin proposed openstack/networking-generic-switch master: Add Mellanox MLNX-OS Switch support https://review.openstack.org/642062 | 15:38 |
*** ankita-2798 has joined #openstack-ironic | 15:39 | |
larsks | Why is ironic_lib separate? Is it used by anything other than ironic? | 15:39 |
larsks | (I am adding a utility function and trying to figure out where it should live) | 15:39 |
dtantsur | larsks: IPA and ironic-inspector | 15:40 |
jroll | dtantsur is too fast :( | 15:40 |
dtantsur | if it's only used by ironic, leave it in ironic | 15:40 |
larsks | got it, thanks. | 15:40 |
dtantsur | jroll: it's Friday, I'm hurrying towards its evening! | 15:40 |
* jroll passes dtantsur an ale | 15:41 | |
dtantsur | an ale \o/ | 15:41 |
* rpittau hears "ale" and looks up | 15:41 | |
openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: DNM - Test Functional in a cleaner env https://review.openstack.org/641671 | 15:42 |
TheJulia | ale ++++++++ | 15:42 |
TheJulia | Coffee stouts all around? | 15:42 |
dtantsur | daaaaaamn, now I really want one | 15:43 |
TheJulia | success() | 15:43 |
jroll | mmm coffee stout. | 15:43 |
* dtantsur wonders if the craft beer place works on the 8th of March.. | 15:43 | |
rpittau | no coffee stout in Europe I think :/ | 15:43 |
* TheJulia wonders if the craft beer place in palm springs has coffee stouts | 15:43 | |
jroll | dtantsur: the mikeller bar has some excellent beer, if your usual place isn't open | 15:44 |
dtantsur | jroll: yeah, I've been there. but my place is open, sooo ;) | 15:44 |
jroll | though I only remember drinking lambic there, I think they had some stouts | 15:44 |
jroll | heh, fair :) | 15:44 |
TheJulia | brewdog seemed to have a coffee stout | 15:44 |
dtantsur | TheJulia: that's the place I'm talking about ;) | 15:45 |
jroll | aha | 15:45 |
*** rloo has joined #openstack-ironic | 15:45 | |
jroll | I wouldn't complain about going to brewdog :) | 15:45 |
rpittau | of course no brewdog in Paris :( | 15:47 |
dtantsur | rpittau: the closed to you may be Brussels. But I don't know if all of them are equally good. | 15:47 |
dtantsur | * closest | 15:47 |
rpittau | wow there's 3 in Italy | 15:48 |
*** fried_rice is now known as efried_afk | 15:48 | |
rloo | so much for fried_rice... | 15:49 |
*** mjturek has joined #openstack-ironic | 15:49 | |
dtantsur | lemme distract you from the sweet thoughts with https://review.openstack.org/641376 to fix the standalone job | 15:50 |
patchbot | patch 641376 - ironic - Add option to set python interpreter for ansible - 15 patch sets | 15:50 |
dtantsur | :) | 15:50 |
dtantsur | morning rloo | 15:50 |
*** TheJulia is now known as needssleep | 15:50 | |
rpittau | lol | 15:50 |
*** rloo is now known as waterloo | 15:52 | |
dtantsur | needssleep: have you had a chance to submit forum session? I can submit something around right now, if you need me to | 15:52 |
dtantsur | oh, right | 15:52 |
*** dtantsur is now known as creepy_owlet | 15:52 | |
needssleep | dtantsur: submitted two yesterday | 15:52 |
needssleep | creepy_owlet: tagged you as one | 15:52 |
needssleep | s/as/for/ | 15:52 |
creepy_owlet | hmm, I haven't got any emails | 15:52 |
creepy_owlet | anyway, thanks! | 15:52 |
needssleep | I got it though | 15:52 |
needssleep | so I know your on the hook *muahahhahahaa* | 15:53 |
needssleep | :) | 15:53 |
* creepy_owlet flies to the woods | 15:53 | |
waterloo | thx needssleep. lucky creepy_owlet :) | 15:53 |
needssleep | lol | 15:53 |
* etingof thinks that the update service is non-functional in sushy, reviews appreciated -- https://review.openstack.org/#/c/642002/ | 15:54 | |
patchbot | patch 642002 - sushy - Make UpdateService.simple_update() operational - 2 patch sets | 15:54 |
*** mjturek has quit IRC | 15:54 | |
jroll | looks like bionic is fine for all jobs except the UEFI job. is anyone working on that or have bandwidth to work on fixing it? else we need to make it non-voting by march 13 | 15:55 |
*** mjturek has joined #openstack-ironic | 15:56 | |
waterloo | creepy_owlet: i assume we can approve this rfe? https://storyboard.openstack.org/#!/story/2005159 | 15:56 |
needssleep | afaik it is a different path, should be an easy fix but Id on't have the bandwidth for it at the moment | 15:56 |
needssleep | waterloo: it is a bug fix afaik | 15:57 |
needssleep | waterloo: since that was the path the ansible community decided due to the conundrum of how they work | 15:57 |
waterloo | needssleep: adds a new config option. i'll update the story. doesn't matter if rfe or not. | 15:57 |
needssleep | rpittau: ^^^^ | 15:57 |
needssleep | waterloo: thanks | 15:57 |
*** andrein has quit IRC | 15:58 | |
rpittau | waterloo, thanks, that is primarily to fix an issue with ironic-standalone | 15:59 |
waterloo | rpittau: right, but it adds a config option. i just did the 'paper work' :) | 15:59 |
rpittau | but since it's a new option I added the RFE | 15:59 |
creepy_owlet | waterloo: we need it to fix the gate | 16:00 |
creepy_owlet | so, I suggest yes :) | 16:00 |
rpittau | oh yeah, I think we're saying the same thing :D | 16:00 |
*** ankita-2798 has quit IRC | 16:00 | |
waterloo | creepy_owlet: yeah, i know. all done. +A'd and rfe-approved :) | 16:00 |
rpittau | waterloo, thanks | 16:01 |
creepy_owlet | thnx! | 16:01 |
*** mjturek has quit IRC | 16:05 | |
*** mmethot has quit IRC | 16:06 | |
*** andrein has joined #openstack-ironic | 16:07 | |
*** andrein has quit IRC | 16:10 | |
*** mjturek has joined #openstack-ironic | 16:10 | |
*** mjturek has quit IRC | 16:11 | |
*** mjturek has joined #openstack-ironic | 16:12 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic master: Move all testing to Bionic https://review.openstack.org/642015 | 16:15 |
jroll | ^ I think this patch makes sense now | 16:16 |
*** efried_afk is now known as fried_rice | 16:16 | |
needssleep | \o/ | 16:18 |
*** sthussey has joined #openstack-ironic | 16:19 | |
jroll | needssleep: I won't be here next week, added status on this to the whiteboard at lines 258-262 | 16:19 |
jroll | someone can take that patch over if it needs to change | 16:19 |
needssleep | jroll: thanks | 16:19 |
needssleep | ack, have a wonderful week next week | 16:19 |
jroll | needssleep: it's for travel and meetings, unfortunately :P | 16:20 |
jroll | I'll pop in but mostly AFK | 16:20 |
needssleep | Sounds like my last week | 16:22 |
*** mmethot has joined #openstack-ironic | 16:26 | |
*** andrein has joined #openstack-ironic | 16:32 | |
*** andrein has quit IRC | 16:32 | |
*** gkadam has quit IRC | 16:33 | |
creepy_owlet | needssleep: do we need a pre-RC release of ironic and others early next week? or should we wait for the RC? | 16:41 |
needssleep | we're cycle-with-intermedicy, so I don't think we need to do any sort of pre releases or rc releases | 16:45 |
needssleep | I just figure unless we have a good reason to hold off, we consider stein releases for most things in the next week or two, create our stable branch, and then push forward | 16:45 |
creepy_owlet | right | 16:50 |
openstackgerrit | Mark Goddard proposed openstack/ironic master: Deploy templates: conductor and API nits https://review.openstack.org/640446 | 16:52 |
*** iurygregory has quit IRC | 16:55 | |
*** e0ne has joined #openstack-ironic | 17:03 | |
*** e0ne has quit IRC | 17:08 | |
*** sburette has quit IRC | 17:10 | |
openstackgerrit | Harald Jensås proposed openstack/ironic master: Fix typo and docstring in pxe/ipxe https://review.openstack.org/642081 | 17:12 |
*** pcaruana has quit IRC | 17:12 | |
*** gyee has joined #openstack-ironic | 17:22 | |
*** andrein has joined #openstack-ironic | 17:25 | |
*** creepy_owlet is now known as dtantsur|afk | 17:29 | |
dtantsur|afk | have a good weekend | 17:29 |
rpittau | bon weekend! o/ | 17:31 |
*** rpittau is now known as rpittau|afk | 17:31 | |
*** fried_rice is now known as fried_rolls | 17:43 | |
needssleep | https://review.openstack.org/#/c/635996/ <-- has a +2 | 17:44 |
patchbot | patch 635996 - ironic - fast tracked deployment support - 21 patch sets | 17:44 |
*** derekh has quit IRC | 17:44 | |
*** asettle is now known as away | 17:46 | |
*** away is now known as asettle | 17:46 | |
openstackgerrit | Merged openstack/ironic master: Add option to set python interpreter for ansible https://review.openstack.org/641376 | 17:48 |
arne_wiebalck | bye, have a good week-end o/ | 17:52 |
needssleep | o/ | 17:53 |
*** tssurya has quit IRC | 17:55 | |
openstackgerrit | Harald Jensås proposed openstack/ironic master: Fix typo and docstring in pxe/ipxe https://review.openstack.org/642081 | 17:58 |
*** dougsz has quit IRC | 18:01 | |
NobodyCam | Good Morning Ironic'ers | 18:13 |
NobodyCam | and ofc TGIF! | 18:13 |
*** anupn has joined #openstack-ironic | 18:31 | |
*** betherly has joined #openstack-ironic | 18:34 | |
needssleep | It is friday \o/ | 18:38 |
*** betherly has quit IRC | 18:39 | |
*** amoralej is now known as amoralej|off | 18:44 | |
*** anupn has quit IRC | 19:21 | |
*** dustinc has quit IRC | 19:43 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: RPC changes for Graphical Console https://review.openstack.org/640827 | 19:49 |
*** mjturek has quit IRC | 19:50 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: [WIP] Adds graphical console implementation for ilo drivers https://review.openstack.org/640973 | 19:53 |
openstackgerrit | Nisha Agarwal proposed openstack/ironic master: [WIP] Adds graphical console implementation for ilo drivers https://review.openstack.org/640973 | 19:59 |
*** mjturek has joined #openstack-ironic | 20:02 | |
*** irclogbot_1 has quit IRC | 20:13 | |
*** e0ne has joined #openstack-ironic | 20:24 | |
openstackgerrit | Vlad Sorokin proposed openstack/networking-generic-switch master: Add Mellanox MLNX-OS Switch support https://review.openstack.org/642062 | 20:53 |
*** andrein has quit IRC | 20:54 | |
*** mbeierl has quit IRC | 20:54 | |
*** fried_rolls is now known as fried_rice | 20:56 | |
*** Joysong has joined #openstack-ironic | 20:57 | |
*** mkarray has joined #openstack-ironic | 20:59 | |
*** mkarray has left #openstack-ironic | 20:59 | |
openstackgerrit | Merged openstack/ironic-python-agent master: Update python-hardware requirement for py3 https://review.openstack.org/642022 | 21:01 |
*** mjturek has quit IRC | 21:01 | |
*** e0ne has quit IRC | 21:03 | |
*** Joysong has quit IRC | 21:08 | |
*** dustinc has joined #openstack-ironic | 21:10 | |
*** yajushiSri has quit IRC | 21:11 | |
*** yajushiSri has joined #openstack-ironic | 21:13 | |
*** bnemec is now known as beekneemech | 21:18 | |
openstackgerrit | Vlad Sorokin proposed openstack/networking-generic-switch master: Add Mellanox MLNX-OS Switch support https://review.openstack.org/642160 | 21:25 |
*** betherly has joined #openstack-ironic | 21:26 | |
*** betherly has quit IRC | 21:31 | |
*** baha has quit IRC | 21:42 | |
*** betherly has joined #openstack-ironic | 21:47 | |
*** betherly has quit IRC | 21:52 | |
*** rh-jelabarre has quit IRC | 22:00 | |
*** rh-jelabarre has joined #openstack-ironic | 22:03 | |
*** betherly has joined #openstack-ironic | 22:08 | |
*** betherly has quit IRC | 22:13 | |
*** andrein has joined #openstack-ironic | 22:20 | |
*** jtomasek has quit IRC | 22:34 | |
*** rh-jelabarre has quit IRC | 22:35 | |
*** betherly has joined #openstack-ironic | 22:50 | |
*** betherly has quit IRC | 22:55 | |
*** irclogbot_1 has joined #openstack-ironic | 23:01 | |
*** jcoufal has quit IRC | 23:15 | |
*** mmethot has quit IRC | 23:27 | |
*** betherly has joined #openstack-ironic | 23:32 | |
*** whoami-rajat has quit IRC | 23:33 | |
*** betherly has quit IRC | 23:36 | |
*** andrein has quit IRC | 23:43 | |
*** Joysong has joined #openstack-ironic | 23:50 | |
*** fried_rice is now known as efried | 23:56 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!