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

[qa] Proposing Radosław Piliszek to devstack core
by Ghanshyam Mann 03 Feb '20

03 Feb '20

Hello Everyone, Radosław Piliszek (yoctozepto) has been doing nice work in devstack from code as well as review perspective. He has been helping for many bugs fixes nowadays and having him as Core will help us to speed up the things. I would like to propose him for Devstack Core. You can vote/feedback on this email. If no objection by end of this week, I will add him to the list. -gmann

4 4

[Watcher] confused about meeting schedule
by info＠dantalion.nl 03 Feb '20

03 Feb '20

Hello everyone, The documentation for Watcher states that meetings will be held on a bi-weekly basis on odd-weeks, however, last meeting was held on the 8th of January which is not an odd-week. Today I was expecting a meeting as the meetings are held bi-weekly and the last one was held on the 8th of January, however, there was none. Can someone clarify when the next meeting will be held and the subsequent one after that? If these are on even weeks we should also update Watcher's documentation. Kind regards, Corne lukken

3 4

Kayobe Openstack deployment
by Tony Pearce 03 Feb '20

03 Feb '20

Hi all - I wanted to ask if there was such a reference architecture / step-by-step deployment guide for Openstack / Kayobe that I could follow to get a better understanding of the components and how to go about deploying it? The documentation is not that great so I'm hitting various issues when trying to follow what is there on the Openstack site. There's a lot of technical things like information on variables - which is fantastic, but there's no context about them. For example, the architecture page is pretty small, when you get further on in the guide it's difficult to contextually link detail back to the architecture. I tried to do the all-in-one deployment as well as the "universe from nothing approach" but hit some issues there as well. Plus it's kind of like trying to learn how to drive a bus by riding a micro-scooter :) Also, the "report bug" bug link on the top of all the pages is going to an error "page does not exist" - not sure that had been realised yet. Regards, *Tony Pearce* | *Senior Network Engineer / Infrastructure Lead**Cinglevue International <https://www.cinglevue.com>* Email: tony.pearce(a)cinglevue.com Web: http://www.cinglevue.com *Australia* 1 Walsh Loop, Joondalup, WA 6027 Australia. Direct: +61 8 6202 0036 | Main: +61 8 6202 0024 Note: This email and all attachments are the sole property of Cinglevue International Pty Ltd. (or any of its subsidiary entities), and the information contained herein must be considered confidential, unless specified otherwise. If you are not the intended recipient, you must not use or forward the information contained in these documents. If you have received this message in error, please delete the email and notify the sender.

2 3

[kolla] ujson issue affected to few containers.
by Eddie Yen 03 Feb '20

03 Feb '20

Hi everyone, I'm not sure it should be bug report or not. So I email out about this issue. In these days, I found the Rocky source deployment always failed at Ceilometer bootstrapping. Then I found it failed at ceilometer-upgrade. So I tried to looking at ceilometer-upgrade.log and the error shows it failed to import ujson. https://pastebin.com/nGqsM0uf Then I googled it and found this issue is already happened and released fixes. https://github.com/esnme/ultrajson/issues/346 But it seems like the container still using the questionable one, even today (Jan 30 UTC+8). And this not only affected to Ceilometer, but may also Gnocchi. I think we have to patch it, but not sure about the workaround. Does anyone have good idea? Many thanks, Eddie.

3 9

[nova] I would like to add another option for cross_az_attach
by Kim KS 02 Feb '20

02 Feb '20

Hello all, In nova with setting [cinder]/ cross_az_attach option to false, nova creates instance and volume in same AZ. but some of usecase (in my case), we need to attach new volume in different AZ to the instance. so I need two options. one is for nova block device mapping and attaching volume and another is for attaching volume in specified AZ. [cinder] cross_az_attach = False enable_az_attach_list = AZ1,AZ2 how do you all think of it? Best, Kiseok

4 4

[cinder] last call for ussuri spec comments
by Brian Rosmaita 01 Feb '20

01 Feb '20

The following specs have two +2s. I believe that all expressed concerns have been addressed. I intend to merge them at 22:00 UTC today unless a serious issue is raised before then. https://review.opendev.org/#/c/684556/ - support volume-local-cache https://review.opendev.org/#/c/700977 - add backup id to volume metadata cheers, brian

3 3

"encoding error when doing console log show" in Rocky
by Albert Braden 01 Feb '20

01 Feb '20

We're running openstack-ansible Rocky and seeing the "encoding error when doing console log show." I see here that there is a patch: https://bugs.launchpad.net/python-openstackclient/+bug/1747862 How can I apply this patch to our Rocky install? Do I need to just copy the file over the existing /usr/lib/python2.7/dist-packages/openstackclient/compute/v2/console.py on controllers and hypervisors? Or is there a better way?

2 13

[tripleo] no rechecks
by Wesley Hayutin 01 Feb '20

01 Feb '20

Greetings, Hey folks we need a few patches to land before it's safe to blindly recheck. A few minor patches need to land before you can expect to get +1 from zuul in a lot of cases. We appreciate your patience while we resolve the issues. Thanks * TRIPLEO GATES ARE BROKEN* * need two patches to fix - * https://review.opendev.org/#/c/705051/ - * https://review.opendev.org/#/c/704885/ - * Also need - * https://review.rdoproject.org/r/#/c/24754/

1 1

[all][requirements][qa] zipp failing py3.5 again
by Radosław Piliszek 01 Feb '20

01 Feb '20

Folks, we are running circles here. [1] merged breaking zipp again for py3.5 The script which generates these proposals needs to learn python version constraints. Reproposed [2]. [1] https://review.opendev.org/705128 [2] https://review.opendev.org/705184 -yoctozepto

3 4

[ironic] An approach to removing WSME
by Steve Baker 01 Feb '20

01 Feb '20

I've put together a set of changes for removing WSME which involves copying just enough of it into ironic, and I think we need to have the conversation about whether this approach is desirable :) This git branch[1] finishes with WSME removed and existing tests passing. Here are some stats about lines of code (not including unit tests, calculated with cloc): 4500 wsme/wsme 6000 ironic/ironic/api master 7000 ironic/ironic/api story/1651346 In words, we need 1000 out of 4500 lines of WSME source in order to support 6000 lines of ironic specific API code. Switching to a replacement for WSME would likely touch a good proportion of that 6000 lines of ironic specific API code. If we eventually replace it with a new library I think it would be easier if the thing being replaced is inside the ironic source tree to allow for a gradual transition. So this approach could be an end in itself, or it could be a step towards the final goal. My strategy for copying in code was to pull in chunks of required logic while removing some unused features (like request pass-through and date & time type serialization). I also replaced py2/3 patterns with pure py3. There is likely further things which can be removed or refactored for simplicity, but what exists currently works. If there is enough positive feedback for this approach I'll start on unit test coverage for the new code. [1] https://review.opendev.org/#/q/topic:story/1651346

3 2