February 2020 - openstack-discuss

[heat][tacker] After Stack is Created, will it change nested stack Id?
by Patil, Tushar 10 Feb '20

10 Feb '20

Hi All, In tacker project, we are using heat API to create stack. Consider a case where we want to add OS::Heat::AutoScalingGroup in which there are two servers and the desired capacity is set to 2. So internally heat will create two nested stacks and add following resources to it:- child stack1 VDU1 - OS::Nova::Server CP1 - OS::Neutron::Port VDU2 - OS::Nova::Server CP2- OS::Neutron::Port child stack2 VDU1 - OS::Nova::Server CP1 - OS::Neutron::Port VDU2 - OS::Nova::Server CP2- OS::Neutron::Port Now, as part of tacker heal API, we want to heal VDU1 from child stack2. To do this, we will mark the status of the resources from "child stack2" as unhealthy and then update "child stack2" stack. Since VDU1 resource is present in the two nested stacks, I want to keep the nested stack id information in tacker so that after the stack is updated, I can pull physical resource id of the resources from the nested child stack directly. My question is after the stack is created for the first time, will it ever change the nested child stack id? Thank you. Regards, tpatil Disclaimer: This email and any attachments are sent in strictest confidence for the sole use of the addressee and may contain legally privileged, confidential, and proprietary data. If you are not the intended recipient, please advise the sender by replying promptly to this email and then delete and destroy this email and any attachments without any further use, copying or forwarding.

2 2

[goals][Drop Python 2.7 Support] Week R-14 Update (1 week left to complete)
by Ghanshyam Mann 10 Feb '20

10 Feb '20

Hello Everyone, Below is the progress on "Drop Python 2.7 Support" at end of R-14 week. Schedule: https://governance.openstack.org/tc/goals/selected/ussuri/drop-py27.html#sc… Highlights: ======== * 1 week left to finish the work. * This week was very dramatic and full of failure as expected or more than that :). ** With that, I am expecting more issues to occur, dropping py2 from every repo asap is suggested so that we can stabilize the things well before m-3. * Do not worry about the gate failure, it is better to break it now and fix so that we can smoothly release Ussuri. * Oslo dropped the py2, which caused multiple failures on 1. tempest testing on stable branches 2. projects still running py2 jobs on the master. ** we did not opt to cap the oslo lib for <py3.6 and decided to fix the issue instead. **1. Tempest testing is fixed and stable branches gate is green now. **2. TripleO was one project that failed for this and py2 jobs are dropped now. If any project facing the same issue and waiting for some general fix, need to drop py2 job asap. * Bugs & Issues: ** Stable/ocata|pike|queens|rocky gate is up now. u-c issue on pinned Tempest is resolved now[1] *** http://lists.openstack.org/pipermail/openstack-discuss/2020-February/012462… ** Nova stable branches are still falling for the u-c reason on nova-live-migration. Fixes are up[2] * * Tempest tox 'all-plugin' usage issue[3] *** designate stable branches fixes are up. If any project job faces the same issue, you can apply the same fix.[4] ** neutron-vpass in-tree plugin issue *** slaweq, amotoki and I are discussing the best possible fix. NOTE: I know few py2.7 drop patches are failing on few projects, I will continue debugging those. Help from the failing project will be greatly appreciated to meet the deadline. Project wise status and need reviews: ============================ Phase-1 status: The OpenStack services have not merged the py2 drop patches: NOTE: This was supposed to be completed by milestone-1 (Dec 13th, 19). * Adjutant * Masakari * Qinling Phase-2 status: This is ongoing work and I think most of the repo have patches up to review. Try to review them on priority. If any query and I missed to respond on review, feel free to ping me in irc. * Open review: https://review.opendev.org/#/q/topic:drop-py27-support+status:open How you can help: ============== - Review the patches. Push the patches if I missed any repo. [1] http://lists.openstack.org/pipermail/openstack-discuss/2020-February/012463… [2] https://review.opendev.org/#/q/I8190f93e0a754fa59ed848a3a230d1ef63a06abc [3] https://bugs.launchpad.net/tempest/+bug/1862240 [4] https://review.opendev.org/#/q/topic:bug/1862240+(status:open+OR+status:mer… -gmann

1 0

openstack-dpdk installation
by mesut aygün 10 Feb '20

10 Feb '20

Hey, I am working over dpdk installation with openstack ocata release. How can I use the neutron-openvswitch-agent-dpdk or anything else instead of neutron-openvswitch-agent? Which can I choose the correct packages with os dpdk versions? I couldn't use the devstack versions!! Best regards

2 3

[nova] nova stable pike|queens|rocky is still failing on Tempest u-c: Do not recheck
by Ghanshyam Mann 10 Feb '20

10 Feb '20

After fixing the stable/pike|queens|rocky integrated jobs[1], on using the stable u-c for pinned Tempest run, nova-live-migration fail on these stable branch on nova. I have proposed the fixes[2], wait for recheck till those are merged. [1] https://review.opendev.org/#/q/topic:fix-stable-gate+status:merged [2] https://review.opendev.org/#/q/topic:fix-stable-gate+status:open+projects:o… -gmann

1 0

[qa][stable][tempest-plugins]: Tempest & plugins py2 jobs failure for stable branches (1860033: the EOLing python2 drama)
by Ghanshyam Mann 09 Feb '20

09 Feb '20

Hello Everyone, This is regarding bug: https://bugs.launchpad.net/tempest/+bug/1860033. Using Radosław's fancy statement of 'EOLing python2 drama' in subject :). neutron tempest plugin job on stable/rocky started failing as neutron-lib dropped the py2. neutron-lib 2.0.0 is py3 only and so does u-c on the master has been updated to 2.0.0. All tempest and its plugin uses the master u-c for stable branch testing which is the valid way because of master Tempest & plugin is being used to test the stable branches which need u-c from master itself. These failed jobs also used master u-c[1] which is trying to install the latest neutron-lib and failing. This is not just neutron tempest plugin issue but for all Tempest plugins jobs. Any lib used by Tempest or plugins can drop the py2 now and leads to this failure. Its just neutron-lib raised the flag first before I plan to hack on Tempest & plugins jobs for py2 drop from master and kepe testing py2 on stable bracnhes. We have two way to fix this: 1. Separate out the testing of python2 jobs with python2 supported version of Tempest plugins and with respective u-c. For example, test all python2 job with tempest plugin train version (or any latest version if any which support py2) and use u-c from stable/train. This will cap the Tempest & plugins with respective u-c for stable branches testing. 2. Second option is to install the tempest and plugins in py3 env on py2 jobs also. This should be an easy and preferred way. I am trying this first[2] and testing[3]. [1] https://zuul.opendev.org/t/openstack/build/fb8a928ed3614e09a9a3cf4637f2f6c2… [2] https://review.opendev.org/#/c/703011/ [3] https://review.opendev.org/#/c/703012/ -gmanne

5 17

[all] Gate status: Stable/ocata|pike|queens|rocky is broken: Avoid recheck
by Ghanshyam Mann 09 Feb '20

09 Feb '20

Hello Everyone, Stable/ocata, pike, queens, rocky gate is broken now due to Temepst dependency require >=py3.6. I summarized the situation in ML[1]. Do not recheck on failed patches of those branches until the job is explicitly disabling Tempest. Fixes are in progress, I will update the status here once fixes are merged. [1]http://lists.openstack.org/pipermail/openstack-discuss/2020-February/0123… -gmann

1 1

Regarding OpenStack
by HARISH KUMAR Ivaturi 09 Feb '20

09 Feb '20

Hi I am Harish Kumar, Master Student at BTH, Karlskrona, Sweden. I have started my Master thesis at BTH and my thesis topic is Performance evaluation of OpenStack with HTTP/3. My solutions will take some time (few months) and i would like to request you that could you add me as a contributor to your github repository , so after completing my thesis i could push my codes in that repository. You can contact me for further details. Thanks and Regards Harish Kumar

2 1

[cinder][nova] volume-local-cache meeting day/time set
by Brian Rosmaita 09 Feb '20

09 Feb '20

I picked the day/time with the most votes (9/10, sorry Rajat), but as gibi pointed out in the other thread, it may not have been obvious what you actually voted for. The good news is that the selected day is late in the week, so if you look at this time and say WTF?, let me know and we can scramble to try to find another day/time (though hopefully not--we're trying to find a time that works stretching from Minneapolis to Shanghai). Meeting time: 13:30-14:00 UTC Thursday 13 February 2020 Location: https://bluejeans.com/3228528973 Topic: volume-local-cache specs: cinder: https://review.opendev.org/#/c/684556/ nova: https://review.opendev.org/#/c/689070/ Etherpad: https://etherpad.openstack.org/p/volume-local-cache The meeting will be recorded. Feel free to leave comments on the etherpad if you can't make it but want something addressed. cheers, brian

2 1

[cinder][nova] volume-local-cache proposal cross-project meeting
by Brian Rosmaita 09 Feb '20

09 Feb '20

Liang Fang's volume-local-cache spec has gotten stuck because the Cinder team doesn't want to approve something that won't get approved on the Nova side and vice-versa. We discussed the spec at this week's Cinder meeting and there's sufficient interest to continue with it. In order to keep things moving along, we'd like to have a cross-project video conference next week instead of waiting for the PTG. I think we can get everything covered in a half-hour. I've put together a doodle poll to find a reasonable day and time: https://doodle.com/poll/w8ttitucgyqwe5yc Please take the poll before 17:00 UTC on Saturday 8 February. I'll send out an announcement shortly after that letting you know what time's been selected. I'll send out connection info once the meeting's been set. The meeting will be recorded and interested parties who can't make it can always leave notes on the discussion etherpad. Info: - cinder spec: https://review.opendev.org/#/c/684556/ - nova spec: https://review.opendev.org/#/c/689070/ - etherpad: https://etherpad.openstack.org/p/volume-local-cache cheers, brian

2 2

[Queens][neutron] multiple networks with same vlan id
by Ignazio Cassano 09 Feb '20

09 Feb '20

Hello, I trying to create multiple networks with the same vlan id and different address but it is impossibile because neutron returns vlan id is already present. I tried with success on vcenter and it works. On openstack I can create more subnets with different adresses under the same vlan id, but instances receive static routes from dhcp for all subnets. Any workaround ? I am also trying with segments but they do not seem to fit my case. Regards Ignazio

4 5