| opendevreview | wangzhiguang proposed openstack/tempest master: Test update volume type details with wrong volume name https://review.opendev.org/c/openstack/tempest/+/846242 | 02:15 |
|---|---|---|
| *** pojadhav is now known as pojadhav|ruck | 04:25 | |
| *** elodilles_pto is now known as elodilles | 06:46 | |
| *** jpena|off is now known as jpena | 07:01 | |
| *** arxcruz is now known as arxcruz|rover | 07:08 | |
| opendevreview | zhaoqiuxia proposed openstack/tempest master: create region with description https://review.opendev.org/c/openstack/tempest/+/846281 | 07:51 |
| opendevreview | zhaoqiuxia proposed openstack/tempest master: create update router description https://review.opendev.org/c/openstack/tempest/+/846283 | 07:58 |
| opendevreview | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Deploy with cephadm https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/826484 | 08:08 |
| opendevreview | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Deploy with cephadm https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/826484 | 08:23 |
| opendevreview | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Deploy with cephadm https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/826484 | 08:25 |
| opendevreview | wangzhiguang proposed openstack/tempest master: Test update volume type details with wrong volume name https://review.opendev.org/c/openstack/tempest/+/846242 | 08:36 |
| opendevreview | wangzhiguang proposed openstack/tempest master: Test update volume type details with wrong volume name https://review.opendev.org/c/openstack/tempest/+/846242 | 10:03 |
| opendevreview | wangzhiguang proposed openstack/tempest master: Test update volume type details with wrong is_public https://review.opendev.org/c/openstack/tempest/+/846293 | 10:10 |
| opendevreview | Gorka Eguileor proposed openstack/devstack master: Reduce memory consumption in Cinder services https://review.opendev.org/c/openstack/devstack/+/845805 | 10:10 |
| gibi | hi qa folks I started seeing tempest failures on master gate since yesterday: https://bugs.launchpad.net/tempest/+bug/1979052 | 10:36 |
| gibi | It seems like the validation resource cleanup happens before the server resource is cleand up, but I don't see the reason why | 10:42 |
| frickler | gibi: looks like the failure was on the tempest patch already, but got ignored https://zuul.opendev.org/t/openstack/build/59de845105de4bf8ab16a87811779fd8 I think a revert is in order, do you want to propose it? | 10:54 |
| frickler | but I'm no tempest core anyway | 10:54 |
| frickler | kopecmartin: gmann: ^^ | 10:54 |
| gibi | frickler: good point | 11:01 |
| gibi | I will click revert... | 11:01 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Revert "compute: Move volume attached live migration tests to use SSHABLE" https://review.opendev.org/c/openstack/tempest/+/846251 | 11:03 |
| gibi | kopecmartin, gmann ^^ | 11:03 |
| *** whoami-rajat__ is now known as whoami-rajat | 11:33 | |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Make nova-live-migration voting https://review.opendev.org/c/openstack/tempest/+/846344 | 11:59 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: compute: Move volume attached live migration tests to use SSHABLE https://review.opendev.org/c/openstack/tempest/+/846345 | 12:02 |
| gibi | gmann: do you know if there is a patch open in tempest to fix the multiple network error for test_create_image_from_deleted_server test case? error in https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_7a3/846187/3/check/nova-next/7a30291/testr_results.html | 12:08 |
| kopecmartin | gibi: again? we have dealt with this in the past already | 12:14 |
| kopecmartin | https://bugs.launchpad.net/tempest/+bug/1844568 | 12:14 |
| kopecmartin | i see that there were multiple bug reports on this over the years | 12:15 |
| kopecmartin | https://bugs.launchpad.net/tempest/+bugs?field.searchtext=Multiple+possible+networks+found%2C+use+a+Network+ID+to+be+more+specific&orderby=-importance&search=Search&field.status%3Alist=NEW&field.status%3Alist=EXPIRED&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.status%3Alist=FIXRELEASED&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPL | 12:15 |
| kopecmartin | ETE_WITHOUT_RESPONSE&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_commenter=&field.subscriber=&field.structural_subscriber=&field.tag=&field.tags_combinator=ANY&field.has_cve.used=&field.omit_dupes.used=&field.omit_dupes=on&field.affects_me.used=&field.has_patch.used=&field.has_branches.used=&field.has_branches=on&field.has_no_branches.used=&field.has_no_branches=on&field.has_blueprints.used=&field.has_blueprints=on&fi | 12:15 |
| gibi | yep, it is a totally new run showing it | 12:15 |
| kopecmartin | eld.has_no_blueprints.used=&field.has_no_blueprints=on | 12:15 |
| frickler | gibi: is that also caused by https://review.opendev.org/c/openstack/tempest/+/846251 ? | 12:21 |
| gibi | frickler: I have no idea | 12:22 |
| gibi | a different tc was failing | 12:22 |
| opendevreview | Merged openstack/devstack-tools master: Drop py3.6 and py3.7 from devstack-tools https://review.opendev.org/c/openstack/devstack-tools/+/846138 | 13:17 |
| *** pojadhav|ruck is now known as pojadhav|dinner | 13:18 | |
| *** pojadhav|dinner is now known as pojadhav|ruck | 13:40 | |
| *** pojadhav|ruck is now known as pojadhav|out | 14:26 | |
| gmann | gibi: kopecmartin instead of revert, I think using class level validation resource will fix it? https://review.opendev.org/c/openstack/tempest/+/846251/1/tempest/api/compute/admin/test_live_migration.py#185 | 14:40 |
| gibi | gmann: ooh, I will give it a try | 14:41 |
| gibi | gmann: thanks a lot | 14:41 |
| gmann | gibi: on multiple network error in test_create_image_from_deleted_server : no patch up please propose one by adding create_default_network = True | 14:42 |
| gibi | gmann: I will look at that too | 14:42 |
| gibi | gmann: should we pull the revert from the gate? | 14:42 |
| gmann | gibi: here | 14:42 |
| gmann | https://github.com/openstack/tempest/blob/7e96c8e854386f43604ad098a6ec7606ee676145/tempest/api/compute/images/test_images_negative.py#L46 | 14:42 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Revert "compute: Move volume attached live migration tests to use SSHABLE" https://review.opendev.org/c/openstack/tempest/+/846251 | 14:43 |
| gibi | I pulled the revert out | 14:43 |
| *** whoami-rajat__ is now known as whoami-rajat | 14:43 | |
| gmann | gibi:ohk, sure. i missed gate blocking | 14:44 |
| gmann | gibi: I am fine to merge the revert and on top of it I can try the class level resource fix | 14:45 |
| gibi | I already pulled the revert out and I will propose the class level validation fix | 14:46 |
| gmann | gibi: +1, can you remove -W from 846251 ? | 14:47 |
| gibi | ahh | 14:47 |
| gibi | OK | 14:47 |
| gibi | a bit of a miscommunication :) | 14:47 |
| gibi | I removed | 14:47 |
| gmann | thanks | 14:48 |
| opendevreview | Dan Smith proposed openstack/devstack master: Add perftop.py tool for examining performance https://review.opendev.org/c/openstack/devstack/+/846198 | 14:48 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: compute: Move volume attached live migration tests to use SSHABLE https://review.opendev.org/c/openstack/tempest/+/846345 | 14:53 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Make nova-live-migration voting https://review.opendev.org/c/openstack/tempest/+/846344 | 14:53 |
| gibi | gmann: I think you have to +A the revert again | 14:54 |
| gmann | gibi: done | 14:54 |
| gibi | thanks | 14:54 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Create default net in ImagesNegativeTestJSON https://review.opendev.org/c/openstack/tempest/+/846419 | 15:04 |
| gibi | gmann: the default net patch is also up ^^ | 15:04 |
| gibi | thanks for the help | 15:04 |
| gmann | gibi: sure, thanks I will check after gate result. | 15:04 |
| gibi | cool | 15:04 |
| gmann | gibi: 846345 fixes the live migration job | 16:08 |
| gibi | thanks for the tip | 16:12 |
| gibi | s/tip/suggestion | 16:12 |
| *** jpena is now known as jpena|off | 16:15 | |
| simondodsley | Hi - looking for assistance with my third-party-ci. This is a new build deployed using Software Factory. Whenever a devstack is built the hostname gets `.novalocal` appended to the end which is causing resolution issues. I have tried setting `dhcp_domain: ""` in the setup role tasks `main.yaml` for `nova.conf` in the `[api]` stanza but this doesn't seem to do anything. Any ideas?? | 17:17 |
| clarkb | simondodsley: as mentioned in the zuul room I think the DEVSTACK_HOSTNAME var is being set by `hostname -f` and that is going to look in /etc/hosts ot figure out what the hostname is. The dhcp_domain is for the VMs booted by your nested openstack so configures the wrong level. I'm not sure what the best solution is here, but if foo.novalocal is in /etc/hosts I would've expected | 17:21 |
| clarkb | dns resolution to work generally. | 17:21 |
| clarkb | simondodsley: maybe yo ucan share a paste of the errors you are seeing so that we can work from there? | 17:21 |
| simondodsley | I'm just trying one more thing. | 17:21 |
| clarkb | oh it may not be in /etc/hosts which is what would cause the domain lookups to fail | 17:22 |
| clarkb | ya so on my local machine /etc/hostname is set to foo and /etc/resolv.conf has search localdomain set. hostname -f returns foo.localdomain but there is no foo.localdomain entry in /etc/hosts | 17:24 |
| clarkb | updating /etc/hosts to match hostname -f may be the simplest fix | 17:24 |
| simondodsley | we are also using `dns_domain` as well so the hostname is coming out like `foo.novalocal,dev.acme.com` | 17:25 |
| clarkb | simondodsley: `hostname -f` reports foo.novalocal,dev.acme.com ? | 17:27 |
| simondodsley | yes | 17:27 |
| clarkb | huh I wouldn've have expected it to report multiple fqdns. TIL | 17:28 |
| clarkb | *wouldn't have | 17:28 |
| clarkb | I wonder why novalocal is passed too if dns_domain is set. In any case I'm not sure that devstack needs anything to be changed to make this work. Instead I think the test node needs to be able to resolve its own name (possible that devstack needs to document that requirement) | 17:30 |
| simondodsley | well it seems to me that `.novalocal` is the default for `dhcp_domain` to setting it to "" should stop it being used, but currently it doesn't seem to | 17:31 |
| gmann | gibi: seems like fix is good to merge before than the revert https://review.opendev.org/c/openstack/tempest/+/846345/2 | 17:35 |
| clarkb | simondodsley: you are using openstack to boot your test nodes and you are setting dhcp_domain and dns_domain in the configs for that cloud? | 17:35 |
| frickler | https://docs.openstack.org/devstack/latest/faq.html#what-can-i-do-about-rabbitmq-not-wanting-to-start-on-my-fresh-new-vm | 17:35 |
| gmann | gibi: revert need recheck. If you are online, can you rebase 846345 on master and I can +A on that and we can abandon revert | 17:35 |
| gmann | gibi: if you are done for day, I can do that | 17:36 |
| simondodsley | dns_domain is set in the cloud config. the dhcp_domain clearing is being done in the Zuul configs for the individual instances on which we install devstack | 17:36 |
| simondodsley | i think i described that correctly | 17:36 |
| clarkb | simondodsley: ok, I'm not sure setting dhcp_domain in the zuul configs will affect anything | 17:36 |
| clarkb | simondodsley: reading the docs for dhcp_domain you need toset that value on the nova compute hosts config | 17:36 |
| simondodsley | well that does seem to be the case | 17:36 |
| clarkb | and zuul si consuming the cloud via its APIs and not writing configs to the nova compute hosts | 17:37 |
| clarkb | if you control the cloud deployment you are using you could modify it that way | 17:37 |
| clarkb | Otherwise maybe don't set dns_domain and just use .novalocal with an /etc/hosts entry? | 17:37 |
| simondodsley | i'd have to ask my infra guys if they would allow that | 17:38 |
| simondodsley | it may mess up other stuff in the DMZ | 17:38 |
| clarkb | in that case I think dropping the dns_domain config on the networks and just configuring the hosts to be part of .novalocal for testing should be fine | 17:39 |
| clarkb | that should remove the second entry from hostname -f and yo ucan ensure that the results of hostname -f resolve via /etc/hosts | 17:39 |
| simondodsley | i don't think that will fix it. The first time we an error related to novalocal is when we are forcing `apt` to use IPv4 and we get `udo: unable to resolve host dsvm-focal-manila-aio-0000001756.novalocal: Name or service not known` so at that pint DEVSTACK_HOSTNAME isn't even set | 17:41 |
| simondodsley | damn - someone has created a pre-devstack task to hardcode novalocal into the name... Please ignore all the above as this might be the issue | 17:43 |
| clarkb | that sudo thing shouldn't be fatal iirc | 17:43 |
| clarkb | it will complain but isn't fatal | 17:43 |
| simondodsley | no, not fatal, but its the rest of devstack getting upset | 17:44 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: compute: Move volume attached live migration tests to use SSHABLE https://review.opendev.org/c/openstack/tempest/+/846345 | 17:49 |
| gibi | gmann: moved https://review.opendev.org/c/openstack/tempest/+/846345 to master | 17:49 |
| gibi | but that marks the end of my day | 17:50 |
| gibi | o/ | 17:50 |
| gmann | gibi: thanks +W I will keep eyes on it | 17:51 |
| gmann | o/ | 17:51 |
| opendevreview | Ghanshyam proposed openstack/tempest master: Make nova-live-migration voting https://review.opendev.org/c/openstack/tempest/+/846344 | 19:22 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!