Friday, 2020-07-03

« Thursday, 2020-07-02 Index Saturday, 2020-07-04 »
*** paras333 has quit IRC00:09
*** ccamposr has quit IRC00:15
*** ccamposr has joined #openstack-qa00:15
*** brinzhang0 has joined #openstack-qa00:51
*** brinzhang_ has quit IRC00:54
*** rf0lc0 has quit IRC01:14
*** ricolin has joined #openstack-qa01:15
*** yamamoto has joined #openstack-qa01:15
*** Liang__ has joined #openstack-qa01:19
*** lbragstad has quit IRC01:24
*** Liang__ has quit IRC01:28
*** Liang__ has joined #openstack-qa01:29
*** markvoelker has joined #openstack-qa01:38
*** yamamoto has quit IRC01:42
*** markvoelker has quit IRC01:43
*** rcernin has quit IRC02:26
*** rcernin has joined #openstack-qa02:35
openstackgerritKevin Zhao proposed openstack/devstack master: arm64 devstack run  https://review.opendev.org/70831702:36
*** ysirndjuro has left #openstack-qa02:41
*** ysirndjuro has quit IRC02:41
*** rf0lc0 has joined #openstack-qa02:47
*** rf0lc0 has quit IRC02:56
*** rcernin has quit IRC03:04
*** rh-jlabarre has quit IRC03:15
*** rcernin has joined #openstack-qa03:15
*** rcernin has quit IRC03:19
*** rcernin has joined #openstack-qa03:19
*** brinzhang0 has quit IRC03:29
*** brinzhang0 has joined #openstack-qa03:30
*** markvoelker has joined #openstack-qa03:39
*** yamamoto has joined #openstack-qa03:40
*** markvoelker has quit IRC03:43
*** brinzhang_ has joined #openstack-qa03:44
*** brinzhang0 has quit IRC03:47
*** brinzhang0 has joined #openstack-qa03:55
*** rcernin has quit IRC03:55
*** rcernin has joined #openstack-qa03:56
*** brinzhang_ has quit IRC03:57
*** brinzhang_ has joined #openstack-qa04:16
*** vishakha has joined #openstack-qa04:17
*** brinzhang0 has quit IRC04:19
*** viks____ has joined #openstack-qa04:25
*** evrardjp has quit IRC04:33
*** evrardjp has joined #openstack-qa04:33
*** udesale has joined #openstack-qa04:39
*** rcernin has quit IRC04:45
*** rcernin has joined #openstack-qa04:47
*** brinzhang0 has joined #openstack-qa04:52
*** brinzhang_ has quit IRC04:55
*** brinzhang_ has joined #openstack-qa04:57
*** brinzhang0 has quit IRC05:00
*** gcheresh_ has joined #openstack-qa05:06
*** bhagyashris|afk is now known as bhagyashris05:08
*** akahat|pto is now known as akahat05:10
*** markvoelker has joined #openstack-qa05:23
*** markvoelker has quit IRC05:27
*** gcheresh_ has quit IRC05:36
*** gcheresh_ has joined #openstack-qa05:53
*** gcheresh_ has quit IRC06:10
*** ccamposr has quit IRC06:17
*** mnasiadka_ is now known as mnasiadka06:42
*** ccamposr has joined #openstack-qa06:57
*** ccamposr__ has joined #openstack-qa06:59
*** ccamposr has quit IRC07:02
*** markvoelker has joined #openstack-qa07:18
*** slaweq has joined #openstack-qa07:18
*** markvoelker has quit IRC07:23
*** rcernin has quit IRC07:24
*** tosky has joined #openstack-qa07:26
*** jpena|off is now known as jpena07:26
*** ralonsoh has joined #openstack-qa07:31
*** vishakha has quit IRC07:35
*** gfidente|afk is now known as gfidente07:58
*** lucasagomes has joined #openstack-qa08:09
*** dtantsur|afk is now known as dtantsur08:11
*** tosky has quit IRC08:32
*** nightmare_unreal has joined #openstack-qa08:58
*** gcheresh_ has joined #openstack-qa09:01
*** yamamoto has quit IRC09:04
*** yamamoto has joined #openstack-qa09:04
*** yamamoto has quit IRC09:05
*** gcheresh_ has quit IRC09:10
*** markvoelker has joined #openstack-qa09:19
*** yamamoto has joined #openstack-qa09:21
*** tosky has joined #openstack-qa09:21
*** markvoelker has quit IRC09:24
*** markvoelker has joined #openstack-qa09:33
*** ricolin has quit IRC09:33
toskyiurygregory: still troubles with that multinode job?09:34
iurygregorytosky, yeah =(09:36
iurygregorywe found out that for some reason the networking-generic-plugin didnt load any devices09:36
iurygregorybut we use it in another multinode job and it works fine .-.09:37
*** markvoelker has quit IRC09:38
iurygregoryhttps://zuul.opendev.org/t/openstack/build/d39f74f5bdf34b9a94bd13e02aa70954/log/controller/logs/screen-q-svc.txt#353509:38
iurygregorythat is probably the main reason why the tempest tests are failling09:38
toskyweird indeed, but there must be a difference between the two09:38
iurygregorywell the base job is different09:38
iurygregoryone is grenade-multinode and the other is tempest-multinode-full-py309:39
toskyand does the device loading fail before or after the upgrade?09:39
iurygregorybut their configuration in neutron looks the same09:40
iurygregoryhumm let me double check09:40
toskythe configuration looks the same because the first part of a grenade job is basically a devstack job09:40
iurygregorytosky, it fails after the upgrade09:43
iurygregoryhttps://zuul.opendev.org/t/openstack/build/d39f74f5bdf34b9a94bd13e02aa70954/log/controller/logs/screen-q-svc.txt#3258-325909:43
toskyand noting else in the logs shows whether it didn't load the devices?09:44
iurygregorywe dont have much log on ngs09:45
iurygregoryI'm adding a DNM on ngs to see09:45
iurygregoryI added*09:45
*** tkajinam has quit IRC09:46
iurygregorynewbie question, the folder grenade_save should have etc.neutron right?09:48
iurygregoryor maybe it didn't copy, but the service was running since there is the screen showing that09:48
*** rcernin has joined #openstack-qa09:51
*** rcernin has quit IRC09:51
openstackgerritLucas Alvares Gomes proposed openstack/devstack master: DO NOT REVIEW: Change Neutron's default ML2 driver to OVN  https://review.opendev.org/73509709:52
iurygregorylucasagomes, don't break ironic pls =P09:53
iurygregorytosky, one thing that is different is that in our multinode job that is working we have tls enable09:54
iurygregoryhttps://zuul.opendev.org/t/openstack/build/6031592a7ce84fb4af47c2bb753dffb0/log/controller/logs/local_conf.txt#1209:55
toskyiurygregory: it seems etc.neutron should be there, but in fact I don't see it09:55
toskytls, I thought it's easier to break *with* tls :)09:56
iurygregoryhehehe09:56
iurygregorywould be a problem if ironic sets tls config? (because it does)09:57
iurygregorybut the normal grenade works fine09:57
iurygregoryXD09:57
iurygregorywe have enable_proxy_headers_parsing = True09:57
iurygregoryhttps://zuul.opendev.org/t/openstack/build/d39f74f5bdf34b9a94bd13e02aa70954/log/controller/logs/etc/ironic/ironic_conf.txt#18509:57
iurygregorywe have a config set but we don't have enable_service tls-proxy in our local.conf09:58
*** vishakha has joined #openstack-qa10:00
lucasagomesiurygregory, :D I will try not to heh just testing waters here10:02
iurygregorylucasagomes, awesome =D10:03
*** gcheresh_ has joined #openstack-qa10:05
*** gcheresh_ has quit IRC10:18
*** ramishra has quit IRC10:26
*** ramishra has joined #openstack-qa10:27
*** Liang__ has quit IRC10:32
*** ramishra has quit IRC10:48
*** udesale_ has joined #openstack-qa11:09
*** udesale has quit IRC11:11
*** yamamoto has quit IRC11:17
*** bhagyashris is now known as bhagyashris|brb11:30
*** jpena is now known as jpena|lunch11:38
*** ramishra has joined #openstack-qa11:46
*** rf0lc0 has joined #openstack-qa11:56
*** yamamoto has joined #openstack-qa11:58
*** yamamoto has quit IRC12:03
*** yamamoto has joined #openstack-qa12:04
*** yamamoto has quit IRC12:21
*** bhagyashris|brb is now known as bhagyashris12:29
*** lyarwood has joined #openstack-qa12:36
lyarwoodhttps://review.opendev.org/#/c/739072/ - looking for a final core review on this stable/rocky backport to unblock multinode jobs and open the branch back up for openstack/nova etc.12:36
*** jpena|lunch is now known as jpena12:43
openstackgerritLucas Alvares Gomes proposed openstack/devstack master: DO NOT REVIEW: Change Neutron's default ML2 driver to OVN  https://review.opendev.org/73509712:45
*** yamamoto has joined #openstack-qa12:52
*** brinzhang_ has quit IRC13:00
*** yamamoto has quit IRC13:02
*** bhagyashris is now known as bhagyashris|afk13:49
*** markvoelker has joined #openstack-qa13:53
*** markvoelker has quit IRC13:58
*** yamamoto has joined #openstack-qa14:06
*** jparker has joined #openstack-qa14:32
*** raildo_ is now known as raildo14:36
gmannfrickler: yoctozepto if you are around, please check this to unblock stable/rocky - https://review.opendev.org/#/c/739072/14:38
*** nightmare_unreal has quit IRC14:59
*** gfidente has quit IRC15:09
*** gfidente has joined #openstack-qa15:13
*** dtantsur is now known as dtantsur|afk15:20
*** yamamoto has quit IRC15:36
*** markvoelker has joined #openstack-qa15:54
*** markvoelker has quit IRC15:59
*** lucasagomes has quit IRC16:00
*** yamamoto has joined #openstack-qa16:08
*** dosaboy has quit IRC16:12
*** dosaboy has joined #openstack-qa16:17
*** yamamoto has quit IRC16:20
*** udesale has joined #openstack-qa16:23
*** udesale_ has quit IRC16:25
openstackgerritMerged openstack/devstack stable/rocky: Use uwsgi binary from path  https://review.opendev.org/73907216:25
openstackgerritGhanshyam Mann proposed openstack/tempest master: Fix pdf-doc build by skiping the sample file  https://review.opendev.org/73926216:28
gmannkopecmartin: can you check this gate fix for pdf-doc build - https://review.opendev.org/#/c/739262/16:29
*** jpena is now known as jpena|off16:30
kopecmartingmann: looks ok, waiting for the gates16:32
gmannkopecmartin: thanks. I was trying to add dynamic generated config reference in place of sample file but that need more work so fixing gate first and then we can work on reference part16:34
*** udesale has quit IRC16:49
kopecmartingmann: makes sense .. gates passed, i voted +216:50
kopecmartingmann: btw, what do you think about this? https://review.opendev.org/#/c/710996/16:50
gmannkopecmartin:  I will check after lunch16:54
*** vishakha has quit IRC17:17
fricklerdevstack is broken (again) due to new setuptools, see #opendev17:29
*** gfidente has quit IRC17:36
openstackgerritColleen Murphy proposed openstack/tempest master: Add system scope for admin auth  https://review.opendev.org/68607317:44
openstackgerritColleen Murphy proposed openstack/tempest master: Add default RBAC personas to dynamic credentials  https://review.opendev.org/68630617:44
openstackgerritColleen Murphy proposed openstack/tempest master: Add default RBAC personas to pre-provisioned creds  https://review.opendev.org/69839717:44
*** ralonsoh has quit IRC17:54
*** markvoelker has joined #openstack-qa17:55
gmannfrickler: humm, any solution yet?17:55
*** markvoelker has quit IRC18:00
clarkbgmann: my guess is the path changed to /usr/bin instead of /usr/local/bin but someone will need to test that with global install on ubuntu18:02
openstackgerritGhanshyam Mann proposed openstack/devstack master: Set full path in Tempest's img_file option  https://review.opendev.org/73927018:06
*** mvkr has quit IRC18:06
gmannkopecmartin: done, few comments. also testing in ^^18:06
gmannclarkb: ohk.18:06
kopecmartingmann: i already have a patch for that in devstack: https://review.opendev.org/#/c/712634/418:09
kopecmartinit worked .. i'm gonna implement your comments, just few minutes18:09
gmannkopecmartin: ah, sorry, did not see. i abandon mine.18:10
*** gcheresh_ has joined #openstack-qa18:10
*** markvoelker has joined #openstack-qa18:13
*** ajitha has joined #openstack-qa18:15
*** markvoelker has quit IRC18:17
*** gcheresh_ has quit IRC18:19
openstackgerritMartin Kopec proposed openstack/tempest master: Remove deprecated scenario image options  https://review.opendev.org/71099618:37
*** markvoelker has joined #openstack-qa18:43
openstackgerritMerged openstack/tempest master: Fix pdf-doc build by skiping the sample file  https://review.opendev.org/73926218:46
*** markvoelker has quit IRC18:48
clarkbgmann: frickler: if I start a ubuntu:18.04 container and apt-get update && apt-get install git python3-pip && pip3 install -U setuptools pip && git clone https://opendev.org/openstack/keystone && pip3 install ./keystone it installs to /usr/local/bin/keystone-manage19:00
clarkblooks like we use pip install -e in devstack though so trying that now19:02
clarkbaha19:03
clarkbpip install -e produces /usr/bin/keystone-manage19:03
clarkbone workaround here may be to stop using editable installs19:03
*** ajitha has quit IRC19:14
*** ajitha has joined #openstack-qa19:17
ajithaHi all, we are running a third party cinder CI. We are trying to include multiattach feature in our driver. For that, we need to add extra-spec property to the volume type. Can it be provided from local.conf file?19:19
ajithaopenstack volume type set --property property_name=true type_name19:21
ajithaIs there any option to include this extra-spec value from local.conf.. so that cinder volume type will be created automatically with that property19:27
*** andrebeltrami has joined #openstack-qa19:27
clarkbajitha: looking in devstack/lib/cinder openstack --os-region-name="$REGION_NAME" volume type create --property volume_backend_name="${be_name}" ${be_name} is the only line that sets properties on volumes19:28
*** danielarthurt has joined #openstack-qa19:29
clarkbyou may awant to set the property in a followup script or with a devstack plugin (or possibly update devstack to take that input and do it)19:29
toskyclarkb: it seems all jobs are failing because they can't find /usr/local/bin/keystone-admin19:29
clarkbtosky: yup19:29
toskyI guess it's time to call a week and see what happens on Monday :D19:29
clarkbsent email about what I found19:29
toskyoh, I see now (just back from dinner)19:29
clarkbI'm not intending to fix anything, but hopefully give people who want ot fix it a jumpstart19:29
* clarkb is technically on vacation now or something until end of next week19:30
toskyyep19:30
toskyI wonder why setuptools people did such change19:31
toskybut that's a different discussion19:31
ajithaclarkb: Currently, we are creating the volume type using the following configuration in local.conf,  [[post-config|$CINDER_CONF]]19:32
lsekiajitha: in tempest.conf, you'll need to set `[compute_feature_enabled]` `volume_multiattach=True`19:33
clarkbah its set from the nova side19:33
lsekihttps://opendev.org/openstack/tempest/src/branch/master/tempest/api/compute/volumes/test_attach_volume.py#L30019:34
lsekiso it won't skip multiattach tests19:34
ajithaclarkb: so i cant add another property along with backend_name19:34
clarkbajitha: see lseki's comments19:34
lsekiseems that tempest creates volume types dynamically19:34
lsekihttps://opendev.org/openstack/tempest/src/branch/master/tempest/api/compute/volumes/test_attach_volume.py#L34319:34
ajithaIseki: Thank you. yes. but we have a specific extra-spec.19:36
ajithahttps://docs.openstack.org/mitaka/config-reference/block-storage/drivers/nimble-volume-driver.html#nimble-driver-extra-spec-options19:36
ajithai need to add ‘nimble:multi-initiator’=’true’ in volume type19:36
lsekioh19:38
ajithawhile testing i did install devstack and then source the credentials and updated the volume type manually using "openstack volume type set --property nimble:multi-initiator=true nimble19:38
ajithathen executed tempest test cases19:38
ajithai need to automate that. Dont know where i should include that19:38
ajithaIseki: Is there any way19:42
lsekiafaik there's no way to set extra-specs to the dynamic volume types that tempest creates...19:44
clarkbif they are dynamically created by tempest then my suggestion to do it via devstack plugin or after devstack runs won't help19:45
ajithaIseki: Ok thank you19:48
ajithaclarkb: Thank you, If you are able to suggest a link for using devstack plugin. It would be a great help for me19:49
clarkbajitha: I don't think it will work19:50
clarkbbecause tempest is creating the volume types dynamically19:50
ajithaok19:51
openstackgerritBernard Cafarelli proposed openstack/devstack stable/queens: Fix gate, squashing multiple fixes  https://review.opendev.org/73794019:53
ajithatempest will create a  volume type specified by us right? in local.conf we do have a parameter as TEMPEST_VOLUME_DRIVER19:56
ajithaclarkb:19:56
clarkbI don't know sorry. I was just looking at the devstack side of things19:56
ajithaok thanks20:00
lsekitosky: any ideas to help ajitha ^?20:04
lsekiin summary, his driver needs a vendor specific extra-spec to use multiattach feature20:04
lsekiso multiattach tests will fail, as the volume-types are created dynamically during the tests20:05
lsekimy guess is to modify his driver, to make multiattach work without the extra-spec...20:07
toskyuhm, apart from changing the test code, no20:09
toskyand yes, I agree that the driver shouldn't need the extra-spec for that to work20:09
toskyI don't think other drivers need it20:10
ajithaIseki: In my understanding,all the volume will become multiattach enabled if there is not extra-spec. volume A will be normal one with volume type type-A and Volume B can be  multiattach enabled with volume type type-B and type-B will be having multiattach=True20:11
ajithais it not?20:11
*** ajitha has quit IRC20:25
gmannTempest cannot support driving specific extra-specs. cinder allow to create multi attach volume with multiattach=True20:28
*** ajitha has joined #openstack-qa20:36
*** gcheresh has joined #openstack-qa20:38
gmannajitha: resposting on multiattach volume-  Tempest cannot support driving specific extra-specs. cinder allow to create multi attach volume with multiattach=True20:41
gmannclarkb: frickler not sure if setuptools will fix it or not as this seems incompatible change. I think we should use old setuptools till we figured out the final behavior from setuptools20:42
ajithagmann: Thanks. So at the backend, i should enable the all volumes to be multiattach enabled ?? and its the responsibility of the cinder to allow the multiattach or not. Is it that way20:44
ajitha?20:44
*** markvoelker has joined #openstack-qa20:44
gmannajitha: I am not sure how driver are expected to enable it, may be cinder team can confirm please check on #openstack-cinder but from API perspective we/tempest pass it in request body multiattach=true20:46
ajithagmann: yes thank you20:48
*** markvoelker has quit IRC20:48
*** raildo has quit IRC20:54
*** gcheresh has quit IRC20:58
*** rf0lc0 has quit IRC21:03
*** markvoelker has joined #openstack-qa21:09
openstackgerritGhanshyam Mann proposed openstack/devstack stable/pike: Fix gate, squashing multiple fixes  https://review.opendev.org/73561621:16
*** ajitha has quit IRC21:34
*** rf0lc0 has joined #openstack-qa21:45
*** markvoelker has quit IRC21:48
*** rf0lc0 has quit IRC21:50
openstackgerritMartin Kopec proposed openstack/tempest master: Remove deprecated scenario image options  https://review.opendev.org/71099622:24
*** tosky has quit IRC22:50
*** andrebeltrami has quit IRC23:36
*** markvoelker has joined #openstack-qa23:49
*** markvoelker has quit IRC23:53
« Thursday, 2020-07-02 Index Saturday, 2020-07-04 »

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