December 2019 - openstack-discuss

[tripleo] Gate is unstable, do not recheck/approve anything now
by Emilien Macchi 18 Dec '19

18 Dec '19

We're encountering a bug in TripleO, see https://bugs.launchpad.net/tripleo/+bug/1856864 In the meantime please do not recheck or approve a patch, the gate is already fully loaded; we'll purge it, fix the bug and recheck what needs to be done. You can follow the progress of that work on our IRC channel #tripleo. Thanks for your patience. -- Emilien Macchi

1 0

[neutron] Cancle neutron L3 meeting
by LIU Yulong 18 Dec '19

18 Dec '19

Hi all, Due to some local reasons, I'm not be able to host the L3 meeting today. So let's cancel it. And also because contributors from most countries will have the Christmas and new year's holidays in the next two weeks, let's also cancel the following two L3 meetings. So the next L3 meeting will be raised on 8th Jan, 2020. OK, see you guys then.  And happy new year! Regards, LIU Yulong

1 0

[kolla][horizon] Get 503 Service unavailable after set RETRIEVE_PASSWORD.
by Eddie Yen 18 Dec '19

18 Dec '19

Hi, Below is the information about this environment. Openstack release: Rocky Kolla container source type: binary OS distro: Ubuntu 18.04 We tried to set OPENSTACK_ENABLE_PASSWORD_RETRIEVE to True in local_settings, but the dashboard got "503 Service Unavailable" after restart the Horizon container. And there's no error log or something in horizon.log Is there anyway to see what's going on? - Eddie

2 2

[openstack-dev][kuryr] Unable to connect to kube-apiserver
by Veera.reddy B 18 Dec '19

18 Dec '19

Hi , I install Kubernetes using "kubeadm init". Followed below link to install "kuryr-kubernetes". https://docs.openstack.org/kuryr-kubernetes/latest/installation/manual.html… Kube api configuration : /etc/kubernetes/manifests/kube-apiserver.yaml http://paste.openstack.org/show/787704/ kuryr config file : /etc/kuryr/kuryr.conf http://paste.openstack.org/show/787705/ Error while starting kuryr : #> kuryr-k8s-controller --config-file /etc/kuryr/kuryr.conf http://paste.openstack.org/show/787706/ Let me what I am missing in "/etc/kuryr/kuryr.conf" Thanks, Veera.

2 1

[openstack-dev][kuryr] Unable to connect to kube-apiserver
by VeeraReddy 18 Dec '19

18 Dec '19

Hi , I install Kubernetes using “kubeadm init”. Followed below link to install “kuryr-kubernetes”. https://docs.openstack.org/kuryr-kubernetes/latest/installation/manual.html… Kube api configuration : /etc/kubernetes/manifests/kube-apiserver.yaml http://paste.openstack.org/show/787704/ kuryr config file : /etc/kuryr/kuryr.conf http://paste.openstack.org/show/787705/ Error while starting kuryr : #> kuryr-k8s-controller--config-file /etc/kuryr/kuryr.conf http://paste.openstack.org/show/787706/ Let me what I am missing in “/etc/kuryr/kuryr.conf” Thanks, Veera.

1 0

[scientific-sig] No IRC meeting today
by Stig Telfer 18 Dec '19

18 Dec '19

Hi All - Apologies, in the rush to end of year we have no agenda items for today. Given the season, I suggest we reconvene with a fresh start on Tuesday 7th January 2020. Cheers, Stig

1 0

[Watcher]IRC meeting at 8:00 UTC today
by licanwei 18 Dec '19

18 Dec '19

1 0

[nova][ironic] need help understanding 'cpu_arch' in nova ironic driver
by melanie witt 18 Dec '19

18 Dec '19

Hey all, Recently I'm trying to figure out how to fulfill a use case in the nova ironic driver around treating an ironic node's 'cpu_arch' as optional. This is coming up as a result of a downstream issue [1] and recent change on the ironic side [2] to make cpu_arch optional for iscsi deployments. The change makes it so that ironic will _not_ include a 'cpu_arch' attribute as part of a node's properties if iscsi. On the nova side, we have a filter scheduler ImagePropertiesFilter which will only match a node if the architecture, hypervisor_type, and vm_mode in the glance image properties match the cpu_arch of the node. We have always pulled the cpu_arch of the from the ironic node properties since the original ironic driver code was added to nova. Now, the use case I'm trying to solve today [3] is where an iscsi ironic deployment has no cpu_arch specified on the ironic side and the deployer wants to use glance images with architecture specified in their image properties. Today (and historically always) the request to create an instance in this situation will fail with NoValidHost because the nova ironic driver could not determine a cpu_arch from the ironic node. My questions are: * Is this a valid thing to want to do? * If if is valid, what is the correct way that we should handle missing cpu_arch in the nova ironic driver? Should we guess at a default cpu_arch? If so, how? Or should we add a new config option where a default cpu_arch can be specified? Or, other? Thanks in advance for any input you can offer. Cheers, -melanie [1] https://bugzilla.redhat.com/show_bug.cgi?id=1653788 [2] https://review.opendev.org/620634 [3] https://bugzilla.redhat.com/show_bug.cgi?id=1688838

3 3

[TC] 'V' Technical Elections
by Kendall Nelson 18 Dec '19

18 Dec '19

Hello! /me takes TC hat off and puts election official hat on Having run our handy dandy tool to project dates[1] for the elections, we have some decisions to make again. At this point we have threeish options: 1. Run the elections with a week in between with the TC election happening BEFORE the PTL election like the script generated. The TC election would run March 10 to March 31 and the PTL Election would run April 7 to April 21. The primary concern here is that it might lead to some election burnout. 2. Run the elections back to back with the PTL election first (closer to the historic norm). This would be tweaking the generated dates by pulling the PTL election three weeks up and pushing back the TC dates by two weeks. The PTL election would run March 3 to March 17 and the PTL Election would run March 17 to April 7. Both elections would be concluded by milestone 3. Similar concerns with this option to the first with election burnout. 3. Run the elections in parallel as a combined election, similar to how we did for the Train elections. This is much easier for election officials because we only need to generate one electorate. The two elections would start March 10 and be done by March 31 following the TC election model with a nominations period, a campaigning period (optional for PTL elections), and then a polling period (assuming we need a runoff). Other info to keep in mind when formulating your opinions: - Extra ATC deadline currently is March 23-27 which will likely need to get bumped earlier for any of these options - RC1 is the week of April 20-24 - Milestone 3 is April 9 - Release schedule for reference[2] - For TC member terms, those closing in on their ~1 year terms would be closer to 13 months since the train elections concluded March 5th of last year -Kendall (diablo_rojo) & the Election Officials [1]http://paste.openstack.org/show/787114/ [2]https://releases.openstack.org/ussuri/schedule.html

9 24

[searchlight] Elasticsearch 5 and latest support
by Mathieu Gagné 18 Dec '19

18 Dec '19

Hi, I'm reading the documentation for Searchlight and it's recommending ElasticSearch 5.x: https://docs.openstack.org/searchlight/latest/install/elasticsearch.html I installed and tried to use Elasticsearch 5 and it doesn't work. The queries sent by Searchlight aren't accepted by Elasticsearch. For examples: When listing facets: TransportError(400, u'parsing_exception', u'[terms] failed to parse field [size]'): RequestError: TransportError(400, u'parsing_exception', u'[terms] failed to parse field [size]') When searching for images: No registered plugin for type owner: Exception: No registered plugin for type owner Is there any work in progress to add support for Elasticsearch 5? In fact, Elasticsearch 5 is EOL. Are there plans to support later versions? Thanks -- Mathieu

2 4