January 2020 - openstack-discuss - lists.openstack.org

[ops] live-migration progress
by Martin Gehrke 16 Jan '20

16 Jan '20

Hi, Last week at the Openstack Operators meetup in London, someone mentioned that there was an issue with the progress updates during a live migration and that by turning it off you could increase your success rate. Does anyone know more TIA Martin Gehrke DevOps Manager & OpenStack Tech Lead Two Sigma Investments, LP New York, NY

1 0

[cinder] consistency group not working
by Francois Scheurer 16 Jan '20

16 Jan '20

Dear Cinder Experts We are running the rocky release. |We can create a consistency group: openstack consistency group create --volume-type b9f67298-cf68-4cb2-bed2-c806c5f83487 fsc-consgroup Bug 1: but adding volumes is not working: openstack consistency group add volume c3f49ef0-601e-4558-a75a-9b758304ce3b b48752e3-641f-4a49-a892-6cb54ab6b74d c0022411-59a4-4c7c-9474-c7ea8ccc7691 0f4c6493-dbe2-4f75-8e37-5541a267e3f2 => Invalid volume: Volume is not local to this node. (HTTP 400) (Request-ID: req-7f67934a-5835-40ef-b25c-12591fd79f85) Bug 2: deleting consistency group is also not working (silently failing): openstack consistency group delete c3f49ef0-601e-4558-a75a-9b758304ce3b |||=> AttributeError: 'RBDDriver' object has no attribute 'delete_consistencygroup'| See details below. Using the --force option makes no difference and the consistency group is not deleted. Do you think this is a bug or a configuration issue? Thank you in advance. | Cheers Francois |Details: ==> /var/lib/docker/volumes/kolla_logs/_data/cinder/cinder-api-access.log <== 10.0.129.17 - - [27/Sep/2019:12:16:24 +0200] "POST /v3/f099965b37ac41489e9cac8c9d208711/consistencygroups/3706bbab-e2df-4507-9168-08ef811e452c/delete HTTP/1.1" 202 - 109720 "-" "python-cinderclient" ==> /var/lib/docker/volumes/kolla_logs/_data/cinder/cinder-volume.log <== 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server [req-9010336e-d569-47ad-84e2-8dd8b729939c b141574ee71f49a0b53a05ae968576c5 f099965b37ac41489e9cac8c9d208711 - default default] Exception during message handling: AttributeError: 'RBDDriver' object has no attribute 'delete_consistencygroup' 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server Traceback (most recent call last): 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/oslo_messaging/rpc/server.py", line 163, in _process_incoming 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server res = self.dispatcher.dispatch(message) 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 265, in dispatch 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server return self._do_dispatch(endpoint, method, ctxt, args) 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 194, in _do_dispatch 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server result = func(ctxt, **new_args) 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/osprofiler/profiler.py", line 159, in wrapper 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server result = f(*args, **kwargs) 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/cinder/volume/manager.py", line 3397, in delete_group 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server vol_obj.save() 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/oslo_utils/excutils.py", line 220, in __exit__ 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server self.force_reraise() 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/oslo_utils/excutils.py", line 196, in force_reraise 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server six.reraise(self.type_, self.value, self.tb) 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server File "/var/lib/kolla/venv/local/lib/python2.7/site-packages/cinder/volume/manager.py", line 3362, in delete_group 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server self.driver.delete_consistencygroup(context, cg, 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server AttributeError: 'RBDDriver' object has no attribute 'delete_consistencygroup' 2019-09-27 12:16:24.491 30 ERROR oslo_messaging.rpc.server| -- EveryWare AG François Scheurer Senior Systems Engineer Zurlindenstrasse 52a CH-8003 Zürich tel: +41 44 466 60 00 fax: +41 44 466 60 10 mail: francois.scheurer(a)everyware.ch web: http://www.everyware.ch

2 2

[blazar] No IRC meeting today
by Pierre Riteau 16 Jan '20

16 Jan '20

Hello, Similarly to Tuesday, I have to cancel today's Blazar IRC meeting. Sorry for the late notice. Thanks, Pierre Riteau (priteau)

1 0

[tc][all] Updates on Ussuri cycle community-wide goals
by Ghanshyam Mann 16 Jan '20

16 Jan '20

Hello Everyone, We have two goals with their champions ready for review. Please review and provide your feedback on Gerrit. 1. Add goal for project specific PTL and contributor guides - Kendall Nelson - https://review.opendev.org/#/c/691737/ 2. Propose a new goal to migrate all legacy zuul jobs - Luigi Toscano - https://review.opendev.org/#/c/691278/ We are still looking for the Champion volunteer for RBAC goal[1]. If you have any new ideas for goal, do not hesitate to add in etherpad[2] [1] http://lists.openstack.org/pipermail/openstack-discuss/2019-October/010291.… [2] https://etherpad.openstack.org/p/PVG-u-series-goals -gmann & diablo_rojo

1 2

Re: [magnum]: K8s cluster creation times out. OpenStack Train : [ERROR]: Unable to render networking. Network config is likely broken
by Feilong Wang 16 Jan '20

16 Jan '20

Hi Donny, Do you mean Fedore CoreOS or just CoreOS? The current CoreOS driver is not actively maintained, I would suggest migrating to Fedora CoreOS and I'm happy to help if you have any question. Thanks. On 14/01/20 9:57 AM, Donny Davis wrote: > FWIW I was only able to get the coreos image working with magnum oob.. > the rest just didn't work. > > On Mon, Jan 13, 2020 at 2:31 PM feilong <feilong(a)catalyst.net.nz > <mailto:feilong@catalyst.net.nz>> wrote: > > Hi Bhupathi, > > Firstly, I would suggest setting the use_podman=False when using > fedora atomic image. And it would be nice to set the "kube_tag", > e.g. v1.15.6 explicitly. Then please trigger a new cluster > creation. Then if you still run into error. Here is the debug steps: > > 1. ssh into the master node, check log /var/log/cloud-init-output.log > > 2. If there is no error in above log file, then run journalctl -u > heat-container-agent to check the heat-container-agent log. If > above step is correct, then you must be able to see something > useful here. > > > On 11/01/20 12:15 AM, Bhupathi, Ramakrishna wrote: >> >> Wang, >> >> Here it is . I added the labels subsequently. My nova and >> neutron are working all right as I installed various systems >> there working with no issues.. >> >> >> >> >> >> *From:*Feilong Wang [mailto:feilong@catalyst.net.nz] >> *Sent:* Thursday, January 9, 2020 6:12 PM >> *To:* openstack-discuss(a)lists.openstack.org >> <mailto:openstack-discuss@lists.openstack.org> >> *Subject:* Re: [magnum]: K8s cluster creation times out. >> OpenStack Train : [ERROR]: Unable to render networking. Network >> config is likely broken >> >> >> >> Hi Bhupathi, >> >> Could you please share your cluster template? And please make >> sure your Nova/Neutron works. >> >> >> >> On 10/01/20 2:45 AM, Bhupathi, Ramakrishna wrote: >> >> Folks, >> >> I am building a Kubernetes Cluster( Openstack Train) and >> using fedora atomic-29 image . The nodes come up fine ( I >> have a simple 1 master and 1 node) , but the cluster creation >> times out, and when I access the cloud-init logs I see this >> error . Wondering what I am missing as this used to work >> before. I wonder if this is image related . >> >> >> >> [ERROR]: Unable to render networking. Network config is >> likely broken: No available network renderers found. Searched >> through list: ['eni', 'sysconfig', 'netplan'] >> >> >> >> Essentially the stack creation fails in “kube_cluster_deploy” >> >> >> >> Can somebody help me debug this ? Any help is appreciated. >> >> >> >> --RamaK >> >> The contents of this e-mail message and >> any attachments are intended solely for the >> addressee(s) and may contain confidential >> and/or legally privileged information. If you >> are not the intended recipient of this message >> or if this message has been addressed to you >> in error, please immediately alert the sender >> by reply e-mail and then delete this message >> and any attachments. If you are not the >> intended recipient, you are notified that >> any use, dissemination, distribution, copying, >> or storage of this message or any attachment >> is strictly prohibited. >> >> -- >> Cheers & Best regards, >> Feilong Wang (王飞龙) >> Head of R&D >> Catalyst Cloud - Cloud Native New Zealand >> -------------------------------------------------------------------------- >> Tel: +64-48032246 >> Email: flwang(a)catalyst.net.nz <mailto:flwang@catalyst.net.nz> >> Level 6, Catalyst House, 150 Willis Street, Wellington >> -------------------------------------------------------------------------- >> The contents of this e-mail message and >> any attachments are intended solely for the >> addressee(s) and may contain confidential >> and/or legally privileged information. If you >> are not the intended recipient of this message >> or if this message has been addressed to you >> in error, please immediately alert the sender >> by reply e-mail and then delete this message >> and any attachments. If you are not the >> intended recipient, you are notified that >> any use, dissemination, distribution, copying, >> or storage of this message or any attachment >> is strictly prohibited. > > -- > Cheers & Best regards, > Feilong Wang (王飞龙) > ------------------------------------------------------ > Senior Cloud Software Engineer > Tel: +64-48032246 > Email: flwang(a)catalyst.net.nz <mailto:flwang@catalyst.net.nz> > Catalyst IT Limited > Level 6, Catalyst House, 150 Willis Street, Wellington > ------------------------------------------------------ > > > > -- > ~/DonnyD > C: 805 814 6800 > "No mission too difficult. No sacrifice too great. Duty First" -- Cheers & Best regards, Feilong Wang (王飞龙) Head of R&D Catalyst Cloud - Cloud Native New Zealand -------------------------------------------------------------------------- Tel: +64-48032246 Email: flwang(a)catalyst.net.nz Level 6, Catalyst House, 150 Willis Street, Wellington --------------------------------------------------------------------------

5 7

[infra] Retire x/zmq-event-publisher
by Andreas Jaeger 16 Jan '20

16 Jan '20

This repo is not used anymore, it was forked and is maintained for Jenkins now elsewhere. I'll retire the repo now with topic retire-zmq-event-publisher, Andreas -- Andreas Jaeger aj(a)suse.com Twitter: jaegerandi SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, D 90409 Nürnberg (HRB 36809, AG Nürnberg) GF: Felix Imendörffer GPG fingerprint = EF18 1673 38C4 A372 86B1 E699 5294 24A3 FF91 2ACB

1 0

[tripleo] Switch to tripleo-container-manage by default (replacing Paunch)
by Emilien Macchi 15 Jan '20

15 Jan '20

Hi folks, Some work has been done to replace Paunch and use the new "podman_container" Ansible module; It's possible thanks to a role that is now documented here: https://docs.openstack.org/tripleo-ansible/latest/roles/role-tripleo-contai… While some efforts are still ongoing to replace container-puppet.py (which uses Paunch to execute a podman run); the tripleo-container-manage role has reached stability and enough maturity to be the default now. I would like to give it a try and for that I have 2 patches that would need to be landed: https://review.opendev.org/#/c/700737 https://review.opendev.org/#/c/700738 The most popular question that has been asked about $topic so far is: how can I run paunch debug to print the podman commands. Answer: https://docs.openstack.org/tripleo-ansible/latest/roles/role-tripleo-contai… Please raise any concern here and we'll address it. Hopefully we can make the default on time before U cycle ends. On a side note: I prepared all the backports to stable/train so the feature will be available in that branch as well. Thanks, -- Emilien Macchi

3 4

[deployment][kolla][tripleo][osa][docs] Intro to OpenStack
by Radosław Piliszek 15 Jan '20

15 Jan '20

Hiya, Folks! I had some thought after talking with people new to OpenStack - our deployment tools are too well-hidden in the ecosystem. People try deploying devstack for fun (and some still use packstack!) and then give up on manual installation for production due to its complexity and fear of upgrades. I also got voices that Kolla/TripleO/OSA (order random) is an "unofficial" way to deploy OpenStack and the installation guide is the only "official" one (whatever that may mean in this very context). So I decided I go the "newbie" route and inspected the website. https://www.openstack.org invites us to browse: https://www.openstack.org/software/start/ which is nice and dandy, presents options of enterprise-grade solutions etc. but fails to really mention OpenStack has deployment tools (if one does not look at the submenu bar) and instead points the "newbie" to the installation guide: https://docs.openstack.org/install-guide/overview.html which kind-of negates that OpenStack ecosystem has any ready tools of deployment by saying: "After becoming familiar with basic installation, configuration, operation, and troubleshooting of these OpenStack services, you should consider the following steps toward deployment using a production architecture: ... Implement a deployment tool such as Ansible, Chef, Puppet, or Salt to automate deployment and management of the production environment." Just some food for thought. Extra for Kolla and OSA: https://docs.openstack.org/train/deploy/ seems we no longer deploy OpenStack since Train. -yoctozepto

3 5

[nova] Removal of 'deallocate_networks_on_reschedule' virt driver API method
by Stephen Finucane 15 Jan '20

15 Jan '20

Just FYI, the 'deallocate_networks_on_reschedule' method of the nova virt driver API has been removed in [1]. It was only used for nova- network based flows and is therefore surplus to requirements. Third party drivers that implement this function can now remove it. Stephen [1] https://review.opendev.org/#/c/696516/

1 0

[scientific-sig] IRC Meeting today 1100UTC - large scale & planning for 2020
by Stig Telfer 15 Jan '20

15 Jan '20

Hi All - We have a Scientific SIG meeting today at 1100UTC (about 30 minutes time) in channel #openstack-meeting. Everyone is welcome. Today’s agenda is here: https://wiki.openstack.org/wiki/Scientific_SIG#IRC_Meeting_January_15th_2020 <https://wiki.openstack.org/wiki/Scientific_SIG#IRC_Meeting_January_15th_2020> We’d like to gather some datapoints for the Large Scale SIG, and talk CFPs and conferences for 2020. Cheers, Stig

1 0