August 2019 - openstack-discuss - lists.openstack.org

[Mistral] PTL candidacy
by Renat Akhmerov 29 Aug '19

29 Aug '19

Hi, I'd like to announce my PTL candidacy for Mistral in U cycle. In Train, we again made a big step towards having better performance, which turned to be very important for real production use cases. Mistral “join” mechanism is now much more CPU and RAM efficient (60-70% faster for same workflows). Mistral now also provides a mechanism to plug in an alternative scheduler implementation. The new implementation of scheduler, which designed to be more scalable, is now available and being tested. For U cycle I'd like to focus on making Mistral more usable: * Consistent and well-structured documentation * Additional tools that simplify creating custom actions and functions for Mistral * Easier debugging of workflow errors * Easier and smoother installation process Some of the mentioned activities are already going on. For example, we added the workflow execution report that, when requested, shows the entire execution tree (including sub workflows). That allows to see error paths (and hence root causes) w/o having to do many copy-paste actions. It's not a secret that in the last few cycles there was a struggle with attracting more active contributors. Now, we finally have several very talanted engineers fully allocated to work on Mistral. So it makes me believe that we'll be able to solve a number of big tasks (including cool user facing goodies) that we had to put on a shelve just for the lack of resources. As usually, anybody is welcome to join our team! It's a fun project to work on. The best way to get in touch with us is IRC channel #openstack-mistral or the openstack-discuss mailing list (with [mistral] tag in email subject). Renat [1] https://review.opendev.org/#/c/679191

1 0

[neutron][ovn] networking-ovn plugin support for multiple remote ovsdb-server
by Zufar Dhiyaulhaq 29 Aug '19

29 Aug '19

[neutron][ovn] networking-ovn plugin support for multiple remote ovsdb-server Hi All, I have testing OpenStack (Queens Version) with OVN and works, but when I want to cluster the ovsdb-server, OpenStack stuck to create neutron object. based on my discussion in openvswitch mailing list ( https://mail.openvswitch.org/pipermail/ovs-discuss/2019-August/049175.html) my ovsdb-server clustering databases are working. Are networking-ovn support to connect to multiple ovsdb-server? something like this in ml2_conf.ini [ovn] ovn_nb_connection = tcp:10.101.101.100:6641,tcp:10.101.101.101:6641,tcp: 10.101.101.102:6641 ovn_sb_connection = tcp:10.101.101.100:6642,tcp:10.101.101.101:6642,tcp: 10.101.101.102:6642 networking-ovn-metadata-agent.ini [ovn] ovn_sb_connection = tcp:10.101.101.100:6642,tcp:10.101.101.101:6642,tcp: 10.101.101.102:6642 Below is the full step to create ovsdb-server and neutron services. - step 1: bootstrapping ovsdb-server cluster : http://paste.openstack.org/show/766812/ - step 2: creating neutron service in controller : http://paste.openstack.org/show/766464/ - step 3: creating neutron service in compute : http://paste.openstack.org/show/766465/ But when I try to create neutron resource, Its always hang (only neutron resource). This is the full logs of all nodes, contain: http://paste.openstack.org/show/766461/ - all openvswitch logs - port (via netstat) - step bootstraping ovsdb-server Neutron logs in controller0: paste.openstack.org/show/766462/ Are networking-ovn support to connect to multiple ovsdb-server? Best Regards, Zufar Dhiyaulhaq

1 0

[ptl][zaqar][election]WangHao's candidacy for Zaqar PTL in U cycle
by hao wang 29 Aug '19

29 Aug '19

Hi all, I would like to announce my candidacy for Zaqar PTL for U I've been PTL for Zaqar for two cycles(Rocky and Stein), I still want to serve this project in the new cycle, so there is my self-nomination: In U release, we will continue our works in: 1) Keep the Zaqar health and stable 2) Continue to support more backends for topic resource 3) Consider removing the old API code and let it more clearly. 4) Support sms subscription in Zaqar Thanks for your consideration!

1 0

[ptl][searchlight][election] Trinh Nguyen's candidacy for Searchlight PTL in U cycle
by Trinh Nguyen 29 Aug '19

29 Aug '19

Hello. I would like to announce my candidacy for Searchlight PTL for U. I've been the PTL for Searchlight for the last two cycles (Stein and Train) which I believe have been giving me enough understanding and confidence to help to maintain it for at least one more release. Even though there was not much activity of Searchlight in the Train cycle, as I surveyed people and companies around my circle, there're still needs for a centralized resource indexing service as well as multi-cloud support as we laid out [1] in the use cases and vision documents [2]. Moreover, with a simple architecture and well-defined contributing process, I strongly think Searchlight would be a good starting point for future OpenStack contributors. Therefore, Searchlight has good reasons to live one. My jobs as a PTL for Searchlight in the U cycle as I envisioned are: * Welcome and help new contributors * Help to make Searchlight work with ElasticSearch 6.x, 7.x * Finish on-going blueprints and fix bugs Thank you for your consideration. Trinh Nguyen (dangtrinhnt) [1] https://docs.openstack.org/searchlight/latest/user/usecases.html [2] https://docs.openstack.org/searchlight/latest/contributor/searchlight-visio… -- *Trinh Nguyen* *www.edlab.xyz <https://www.edlab.xyz>*

1 0

[election][ptl][magnum] PTL Candidacy
by feilong 29 Aug '19

29 Aug '19

Greeting, Over the last release, I have had the privilege and honor to serve as the PTL of Magnum. I would like to take this opportunity to express my appreciation for all your support and contribution to Magnum. And I look forward to working with you all over the next release. Therefore, I would like announce my candidacy to serve as the Magnum project team leader for the U cycle, if you will have me. In case if you don't know me, I've been working on and contributing to Magnum since Queens release as a core developer. And I have done a lot work for Magnum to help it getting production ready, e.g. the Calico network driver, nodes anti affinity, k8s health status check, Keystone auth, coreDNS autoscale, rolling upgrade, auto scaling, auto healing, private cluster, getting Magnum certified by CNCF and etc, etc. That said, I really understand the ups and downs which Magnum went through and I know the blueprints, direction and pains of this project, for interests from both private cloud and public cloud sides, which make me a good candidate for the PTL position. For U release (including the rest of Train cycle), things I'd like to do: 1. Fedora CoreOS The most matured driver for Kubernetes in Magnum is Fedora Atomic, however, the Fedora community is moving to Fedora CoreOS, so we should start to think about the migration. We have started the work but haven't finished it due to the limited resource among the team. This need to be done in Train or early of U cycle. 2. Node groups Node groups provide users with the ability to specify groups of nodes with different properties for different purposes. Within the scope of a group users are able to define labels, used image, flavor, etc depending on the purpose. We have merged part of this huge feature, but we do need to finish the rest of the work. 3. Security hardening Security is always the most important area we would like to improve for Magnum, there are still parts we need to improve to make Magnum more securer. 4. Containerize Kubernetes master This could be one of the most interesting features for Magnum and it's definitely hard work. Running Kubernetes master nodes in a seed/root Kubernetes could bring us a lot of benefits. 5. Rolling upgrade for node operating system With the big support from CERN, We have done the k8s version rolling upgrade, but we still need the rolling upgrade work for node operating system. I have already got a patch for that being reviewed now. It has been a fantastic experience working with this amazing team and I know without the dedication and hard work of everyone who has contributed to Magnum we can't get those good work done. So the PTL of this cool team is most like a facilitator, coordinator, and mentor. I would be pleased to serve as PTL for Magnum for the U cycle and I'd appreciate your vote. Thanks for your consideration! -- Cheers & Best regards, Feilong Wang (王飞龙) ------------------------------------------------------ Senior Cloud Software Engineer Tel: +64-48032246 Email: flwang(a)catalyst.net.nz Catalyst IT Limited Level 6, Catalyst House, 150 Willis Street, Wellington ------------------------------------------------------

1 0

[all] [tc] [osc] [glance] [train] [ptls] Legacy client CLI to OSC review 639376
by Alexandra Settle 29 Aug '19

29 Aug '19

Hi all, For the Train cycle Artem Goncharov proposed moving the legacy client CLIs to OSC. This goal did not move forward due to a broad range of concerns and was eventually -1'd by Erno Kuvaja (Glance PTL) as he was unable to support it from a Glance point of view. Moving forward, I'd like to end the discussion in the review [1] by abandoning the patch and move this to the mailing list. The following looks like it needs to happen: 1. This PR should be abandoned. It is not going to be accepted as a commmunity goal in this format and the debate within the comments is circular. Let's step out of here, and start having conversation elsewhere. 2. To start those conversations, a pop-up team would be a suitable alternative to begin driving that work. Someone needs to step forward to lead the pop-up team. The TC recommends two individuals step forwards as indicated in the pop-up team Governance document [2]. 3. I recommend the pop-up team include Erno Kuvaja or a representative from the Glance team and including Matt Reidemann (who has been working to close the compute API gaps). 4. The leaders must then identify a clear objective, and a clear disband criteria. If in case the pop-up team decide that this could be better drafted as a community goal for U, then that is a suitable alternative that can be defined using the new goal selection criteria that is being defined here [3] But this should not be decided before clear objectives are defined. Thanks, Alex IRC: asettle [1] https://review.opendev.org/#/c/639376/ [2] https://governance.openstack.org/tc/reference/popup-teams.html [3] https://review.opendev.org/#/c/667932/

8 10

[ptl][tripleo][election] PTL Candidacy
by Wesley Hayutin 29 Aug '19

29 Aug '19

Greetings, I would like to nominate myself for the TripleO PTL role for the U cycle. I have been working closely with several previous PTL's like Alex, Emilien and Juan Osorio and I am hoping to contribute back to the community as a PTL while following their excellent example. I would like to see Tripleo continue focusing on simplification and improving the user experience. Specifically continue the path set in previous TripleO cycles. - Push forward with tripleo-ansible and the transformation squad. - Work with the newly formed ansible sig group. - Help continue to simplify the stack of components where possible. - Help with new features like the undercloud minion node. - Continue to support the wider adoption of the Tripleo Single Node deployment as a development and test tool. - Continue the great work being done to clean up logging and error handling to improve operators ability to troubleshoot their installations. Thank you for the opportunity! Thanks, Wes Hayutin irc: weshay

1 0

[tripleo] TripleO CI Summary: Sprint 35
by Rafael Folco 29 Aug '19

29 Aug '19

Greetings, The TripleO CI team has just completed Sprint 35 / Unified Sprint 14 (Aug 08 thru Aug 28). The following is a summary of completed work during this sprint cycle: - Completed testing on the RHEL8 OVB featureset 001 job and added it to the periodic pipeline. - Added RHEL8 scenario{1-4} jobs to the periodic pipeline as non-voting. - Completed the design for staging environment as a zuul job for testing changes in the promoter server and also prepare for the multi-arch builds. - Created a staging promoter job and started implementing functional tests on molecule for the staging environment setup. - Completed a PoC for multi-arch container builds with tagging and manifests approach. The change [3] on build-containers role that introduces arch aware tagging is still under review. - Promotion status: green on all branches for most of the sprint cycle. - Removed Fedora and Pike jobs everywhere. - Bumped ansible to version 2.8 across tripleo ci jobs. The planned work for the next sprint [1] are: - Address issues for scenario{1-4} jobs for RHEL8 in the periodic pipeline and get jobs to green status. - Merge multi-arch container tagging support and change the promoter code to push manifests with annotated metadata for architecture in addition to the arch tagged containers strategy. - Implement the staging promoter job for testing changes in the promotion workflow. - Create functional test cases for the staging promoter setup using molecule. - Design a staging promotion workflow to emulate promotions using mock containers and test servers. The Ruck and Rover for this sprint are Gabriele Cerami (panda) and Sagi Shnaidman (sshnaidm). Please direct questions or queries to them regarding CI status or issues in #tripleo, ideally to whomever has the ‘|ruck’ suffix on their nick. Ruck/rover notes are being tracked in etherpad [2]. Thanks, rfolco [1] https://tree.taiga.io/project/tripleo-ci-board/taskboard/unified-sprint-15 [2] https://etherpad.openstack.org/p/ruckroversprint15 [3] https://review.opendev.org/#/c/663977/

1 0

[oslo] Reminder: Feature Freeze is this Friday
by Ben Nemec 29 Aug '19

29 Aug '19

This is your friendly neighborhood PTL reminding you that if you need a feature to merge in Oslo this cycle, it should be in by this Friday. We feature freeze earlier than other projects so they have a chance to make use of any features we add before their feature freeze. After Friday, any new features in Oslo will require an exception from me. I don't see much in the way of feature reviews open against Oslo right now (at least none that are mergeable), but if I missed anything let me know. Thanks. -Ben

1 0

[requirements][placement] orjson instead of stdlib json
by Chris Dent 29 Aug '19

29 Aug '19

Requirements people: For the past few months we've been doing profiling of placement, modeling large clouds where many results (7000 resource providers) will be returned in response to common requests. One of the last remaining chunks of slowness is serializing the very large dicts representing those providers (and associated allocation requests) to JSON (greater than 100k lines when pretty printed). Using the oslo jsonutils (which is a light wrapper over the stdlib 'json' module) approximately 25% of time is consumed by dumps(). Using orjson [1] instead, time consumption drops to 1%, so this is something it would be good to make real in placement (and perhaps other projects that have large result sets). There's a WIP that demonstrates its use at [2]. I'm posting about it because JSON serialization seems like it might be an area that the requirements team has greater interest than some other areas. Also, orjson is only Python 3, so use needs to be adapted for Python 2, as shown in the WIP [2]. Otherwise it's fine: active, Apache2 or MIT license. Thoughts? [1] https://pypi.org/project/orjson/ [2] https://review.opendev.org/#/c/674661/ -- Chris Dent ٩◔̯◔۶ https://anticdent.org/ freenode: cdent

7 9