- openstack-discuss - lists.openstack.org

[infra][qa] multiline tracebacks not being indexed anymore
by Matt Riedemann 05 Nov '19

05 Nov '19

We used to be able to query for things like this: message:"in reserve_block_device_name" AND message:"MessagingTimeout" AND tags:"screen-n-api.txt" to fingerprint a traceback in logstash like this [1] but that no longer works. The multiline logstash filter is at [2] but doesn't seem to be getting applied anymore. I asked about this in -infra today and fungi said: "(4:44:00 PM) fungi: mriedem: i suspect that coincided with switching away from osla, we may need some means of parsing tracebacks out of logs in the indexer" I don't know what that means (what's osla? is [2] no longer used?) but if someone could point me at some things to look at I could see if I can generate a fix. [1] https://zuul.opendev.org/t/openstack/build/ef0196fe84804b44ac106d011c8c29ea… [2] https://opendev.org/openstack/logstash-filters/src/branch/master/filters/op… -- Thanks, Matt

2 1

[blazar] Shanghai Summit and PTG activities for Blazar; no IRC meetings
by Pierre Riteau 04 Nov '19

04 Nov '19

Hello, Several of the Blazar core reviewers and contributors will be in Shanghai this week. Don't hesitate to talk to them if you are interested in resource reservation as a service. On Tuesday there will be a project update: https://www.openstack.org/summit/shanghai-2019/summit-schedule/events/24373… And on Friday at the PTG, there will be a project onboarding session, as well as technical discussions. Since most of the team is in Shanghai, IRC meetings are cancelled this week. Best wishes, Pierre

1 0

Cinder multi backend quota update
by M Ranga Swami Reddy 04 Nov '19

04 Nov '19

Hello, We use 2 types of volume, like volumes and volumes_ceph. I can update the quota for volumes quota using "cinder quota-update --volumes=20 project-id" But for volumes_ceph, the above CLI failed with volumes_ceph un recognised option.. Any suggestions here?

2 2

Openstack third pary CI Implementation(Zuul V3 , Nodepool, Queens)
by Sikka, Rishabh 04 Nov '19

04 Nov '19

Dear Team , I am installing Zuul V3 ,Nodepool for our openstack third party CI ,Please let me know if any documentation is available for the same as I am struggling with the steps shared on zuul official page. Also please do let me know if #openstack-third-party-ci is correct IRC channel related to third party ci implementation, if it is correct please refer some of the name who had already implemented it. I had already tried posting my questions on same IRC channel but did not get the desired reply. Note -: If above PDL is not correct , please refer it to the correct PDL. Regards Rishabh Sikka

2 1

[self-healing][autohealing][ptg] self-healing session 11/5@1:40PM
by Eric K 04 Nov '19

04 Nov '19

Looking forward to seeing you tomorrow at the PTG! The self-healing SIG will meet to continue the work on making self-healing easy and available. We’re especially eager to hear/discuss user feedback, feature requests, and ideas for making self-healing easier and better. The session will take place on Tuesday 11/5 at 1:40 - 3:10 PM in room 431 [1]. Planned topics include predictive analytics, cross-project testing, user feedback, and unified bug reporting. Additional topics and comments welcome in the etherpad: https://etherpad.openstack.org/p/SHA-self-healing-SIG [1] Map and schedule https://object-storage-ca-ymq-1.vexxhost.net/swift/v1/6e4619c416ff4bd19e1c0…

1 0

[ptg][storlets] Shanghi PTG plan for Storlets
by Kota Tsuyuzaki 04 Nov '19

04 Nov '19

Hello guys, The Shanghai PTG will happen soon! The Storlets core team prepared the etherpad for Shanghi PTG here, https://etherpad.openstack.org/p/storlets-ptg-shanghai Please feel free to add any topics you want to discuss there. Note that, Storlets reserved the room for 1.5 days but we'll use the latter half day actually because of the other schedules of core team members. If anyone want to catch me in other time, please let me know via E-mail, or something (I'm not sure IRC and any other tools would work in Shanghi PTG network, though) -------------------------------------------- 露崎　浩太 (Kota Tsuyuzaki) kota.tsuyuzaki.pc(a)hco.ntt.co.jp NTTソフトウェアイノベーションセンタ分散処理基盤プロジェクト 0422-59-2837 ---------------------------------------------

3 2

[ospurge] looking for project owners / considering adoption
by Adam Harwell 04 Nov '19

04 Nov '19

I haven't seen much activity on this project in a while, and it's been moved to opendev/x since the opendev migration... Who is the current owner of this project? Is there anyone who actually is maintaining it, or would mind if others wanted to adopt the project to move it forward? Thanks, --Adam Harwell

5 10

[OpenInfra Summit Shanghai] Some wrong information at Shanghai Summit
by Trinh Nguyen 04 Nov '19

04 Nov '19

Dear OSF, I don't attend the Shanghai summit but I notice some pictures of the presentation which Jonathan Bryce is delivering. The attached photo is about the OpenInfra Days around the world and the one on the left is *OpenInfra Days Vietnam*, not Korea. I know it's just a little thing but because it already upset some of our community members I would like to ask for correction. The Vietnam OpenInfra Community has been putting lots of effort to be recognize so please understand. Bests, Trinh -- *Trinh Nguyen* *www.edlab.xyz <https://www.edlab.xyz>*

2 2

[glance] PTG Schedule
by Abhishek Kekane 04 Nov '19

04 Nov '19

Hi All, Welcome to China for the OpenInfra Summit and PTG. I have prepared a PTG schedule for Glance [1]. We will have a small meet with QA team on Wednesday 6th November, full day session on Thursday 7th November and half day on Friday 8th November. I have kept 1 hour for Open discussion on Thursday and Friday, so If someone wants to join us with their topics can do the same during this time. Friday 11:30 to 12:00, we have an interview scheduled to share updates on Glance. Have a productive Summit and PTG. [1] https://etherpad.openstack.org/p/Glance-Ussuri-PTG-planning Thanks & Best Regards, Abhishek Kekane

1 0

[neutron] Bug deputy report for week of November 01
by Lajos Katona 04 Nov '19

04 Nov '19

Hi, On the week from October 28 to November 3 I was the bug deputy of neutron, here's a short summary of the bugs arrived. Critical bugs - https://bugs.launchpad.net/neutron/+bug/1850288: scenario test test_multicast_between_vms_on_same_network fails - workaround to skip unstable test is merged, more investigation is needed. - https://bugs.launchpad.net/neutron/+bug/1850626 neutron-dynamic-routing: TypeError: bind() takes 4 positional arguments but 5 were given - assigned, workaround to skip falling test is there ( https://review.opendev.org/692372) need to check how to fix the issue with https://review.opendev.org/288271 High bugs - https://bugs.launchpad.net/neutron/+bug/1850639 FloatingIP list bad performance - Assigned, in progress - https://bugs.launchpad.net/neutron/+bug/1850800 UDP port forwarding test is failing often High - The workaround to make the test unstable is there, more investigation is needed Medium bugs - https://bugs.launchpad.net/neutron/+bug/1850558 "AttributeError: 'str' object has no attribute 'content_type' in functional tests - assigned, in progress - https://bugs.launchpad.net/neutron/+bug/1850557 DHCP connectivity after migration/resize not working - More investigation is necessary - https://bugs.launchpad.net/neutron/+bug/1850779 [L3] snat-ns will be initialized twice for DVR+HA routers during agent restart medium assigned - assigned, in progress - https://bugs.launchpad.net/neutron/+bug/1850864 DHCP agent takes very long time to report when port is provisioned Medium - Assigned In progress Low bugs - https://bugs.launchpad.net/neutron/+bug/1849980: Do not inherit from built-in "dict" - assigned, in progress - https://bugs.launchpad.net/neutron/+bug/1850602: remove firewall_v1 exceptions in neutron-lib - assigned, in progress RFE - https://bugs.launchpad.net/neutron/+bug/1850818 [RFE][floatingip port_forwarding] Add description field - assigned More investigation is needed - https://bugs.launchpad.net/neutron/+bug/1850630 firewall rule update validating func is not robust enough，missing considering the stock data - https://bugs.launchpad.net/neutron/+bug/1850137: Hosts in a VPNaaS-VPNaas VPN lose their interconnect. Regards Lajos

1 0

[cinder] core list update
by Brian Rosmaita 03 Nov '19

03 Nov '19

In preparation for the PTG, it's time to revise the list of Cinder core contributors. The following people have made great contributions to Cinder, but unfortunately Cinder is no longer their current focus, and they've indicated that they no longer have sufficient time to act as core reviewers for the project: - John Griffith - TommyLike Hu - Xing Yang - Yikun Jiang On behalf of the entire Cinder project team, I thank John, TommyLike, Yikun, and Xing for their past service to Cinder, and hope that they may find more time to spend on Cinder in the future. While I'm thanking people, I should also thank the current members of the cinder-core group for all their work during the Train cycle (and thank them in advance for all the work I look forward to them doing during Ussuri!): - Eric Harney - Gorka Eguileor - Ivan Kolodyazhny - Jay Bryant - Rajat Dhasmana - Sean McGinnis - Walter A. Boring IV There will leave some openings for new core contributors during the Ussuri cycle. If you're interested in getting some advice about how to position yourself to become a Cinder core, please seek out me or any of the active cores listed above during the PTG. For people who won't be at the PTG, you can always look for us in #openstack-cinder. Let's have a productive PTG! cheers, brian

2 1

[tempest] Train release and zero-disk error
by Nikita Belov 02 Nov '19

02 Nov '19

Hello! I try to use rally with Train release of Openstack. I have 842 success tests and 436 failures because of "Only volume-backed servers are allowed for flavors with zero disk". What can I do with this error? Report attached.

1 0

[ops] Shanghai Sessions
by Erik McCormick 02 Nov '19

02 Nov '19

Hello all, For anyone attending the Shanghai Summit, I wanted to point out a couple of interesting Ops things. First, we are redoing the ever popular Ops War Stories at the Forum on Monday. These are a series of lightning talks by anyone with a good story to tell. No presentations needed. Sign up here: https://etherpad.openstack.org/p/shanghai-ptg-ops-war-stories Second, we have space in the PTG area Thursday afternoon (1:30pm - 4:30pm, Room 430) to do a mini Ops Meetup. Please come join us! Share ideas for topics here so we can hit the ground running. https://etherpad.openstack.org/p/shanghai-ptg-ops-meetup Cheers, Erik

1 0

[ops] Shanghai Sessions
by Erik McCormick 02 Nov '19

02 Nov '19

Hello all, For anyone attending the Shanghai Summit, I wanted to point out a couple of interesting Ops things. First, we are redoing the ever popular Ops War Stories at the Forum on Monday. These are a series of lightning talks by anyone with a good story to tell. No presentations needed. Sign up here: https://etherpad.openstack.org/p/shanghai-ptg-ops-war-stories Second, we have space in the PTG area Thursday afternoon (1:30pm - 4:30pm, Room 430) to do a mini Ops Meetup. Please come join us! Share ideas for topics here so we can hit the ground running. https://etherpad.openstack.org/p/shanghai-ptg-ops-meetup Cheers, Erik

1 0

[ptg][neutron] Onboarding during the PTG
by Slawek Kaplonski 02 Nov '19

02 Nov '19

Hi all new (and existing) Neutrinos, During the PTG in Shanghai we are planning to organize onboarding session. It will take place on Wednesday in morning sessions. It is planned to be started at 9:00 am and finished just before the lunch on 12:30. See on [1] for details. All people who wants to learn about Neutron and contribution to it are welcome on this session. Also all existing team members are welcome to be there to show to new contributors and help them with onboarding process :) See You all in Shanghai! [1] https://etherpad.openstack.org/p/Shanghai-Neutron-Planning -- Slawek Kaplonski Senior software engineer Red Hat

1 1

[neutron][ptg] Team dinner
by Slawek Kaplonski 02 Nov '19

02 Nov '19

Hi neutrinos, Thanks to LIU Yulong who helped me a lot to choose and book some restaurant, we have now booked restaurant: Expo source B2, No.168, Shangnan Road, Pudong New Area, Shanghai, TEL: +86 21 58882117 书院人家(世博源店) 上海市浦东新区上南路168号世博源B2 The Dianping page: http://www.dianping.com/shop/20877292 Dinner is scheduled to Tuesday, 5th Nov at 6pm. Restaurant is close to the Expo center. It's about 15 minutes walk according to the Google maps: https://tinyurl.com/y2rc83ej -- Slawek Kaplonski Senior software engineer Red Hat

2 2

[openstack-dev][kuryr] Error in creating LB
by VeeraReddy 02 Nov '19

02 Nov '19

Hi, I am trying to install openstack & Kubernetes using devstack https://docs.openstack.org/kuryr-kubernetes/latest/installation/devstack/ba… Error Log : http://paste.openstack.org/show/785670/ Local.conf : Paste #785671 | LodgeIt! Regards, Veera.

4 4

[auto-scaling] SIG PTG schedule. please join us :)
by Rico Lin 02 Nov '19

02 Nov '19

Hi all PTG is right next week, so if you're interested in Auto-scaling features, please join us. Our session will be hosted on 11/6 for Half-day from 9:00 am to 12:30 pm in room 431. You can suggest PTG sessions in our etherpad: https://etherpad.openstack.org/p/PVG-auto-scaling-sig Feel free to join our IRC channel as well: #openstack-auto-scaling Also, you can check out our new documents for autoscaling: https://docs.openstack.org/auto-scaling-sig And provide any feedback/ bug report for any Auto-scaling issue related to OpenStack in https://storyboard.openstack.org/#!/project/openstack/auto-scaling-sig Here's my Wechat ID: RicoLinCloudGeek See you all in Shanghai! -- May The Force of OpenStack Be With You, *Rico Lin*irc: ricolin

1 2

[cinder][ptg] team dinner in Shanghai
by Brian Rosmaita 01 Nov '19

01 Nov '19

This event is open to anyone who will be in Shanghai on Thursday 7 November and who has a constructive interest in Cinder. I know we have a few contributors in Shanghai who won't be at the PTG; hopefully, they'll be able to join us so we can meet face-to-face. We're planning for a 5:00 pm dinner; it's kind of early, but the restaurant is a close walk from the Expo Center, so I figured it would be easier to go directly there instead of dispersing to our hotels and regrouping later. Plus, an early dinner will enable people to attend the (unofficial) Game Night [0]. Details and a signup sheet (we need a head count) are at the top of the Cinder PTG etherpad: https://etherpad.openstack.org/p/shanghai-ptg-cinder I haven't been able to confirm the time and location yet, but I will keep the above etherpad updated with that info. cheers, brian [0] https://etherpad.openstack.org/p/pvg-game-night

1 0

[cinder] changing the weekly meeting time
by Brian Rosmaita 01 Nov '19

01 Nov '19

(Just to be completely clear -- we're only gathering information at this point. The Cinder weekly meeting is still Wednesdays at 16:00 UTC.) As we discussed at today's meeting [0], a request has been made to hold the weekly meeting earlier so that it would be friendlier for people in Asia time zones. Based on the people in attendance today, it seems that a move to 14:00 UTC is not out of the question. Thus, the point of this email is to solicit comments on whether we should change the meeting time to 14:00 UTC. As you consider the impact on yourself, if you are in a TZ that observes Daylight Savings Time, keep in mind that most TZs go back to standard time over the next few weeks. (I was going to insert an opinion here, but I will wait and respond in this thread like everyone else.) cheers, brian [0] http://eavesdrop.openstack.org/meetings/cinder/2019/cinder.2019-10-23-16.00…

3 3

[ops] preliminary ops meetup proposal, jan 2020, London
by Chris Morgan 01 Nov '19

01 Nov '19

Please see https://twitter.com/osopsmeetup/status/1190365967106412544?s=20 We find this method of getting news out about operators meetups gets good engagement, so please also follow if of interest to you. Chris -- Chris Morgan <mihalis68(a)gmail.com>

1 0

[tripleo] TripleO CI Summary: Sprint 38
by Rafael Folco 01 Nov '19

01 Nov '19

Greetings, The TripleO CI team has just completed Sprint 38 / Unified Sprint 17 (Oct 10 thru Oct 30). The following is a summary of completed work during this sprint cycle: - Tested the temporary manifest implementation in the new promoter for not breaking promotion workflow. Issues to be bootstrapped in the next sprint. - Implemented CI jobs in zuul to build and run tests against ceph-ansible and podman pull requests in github. A PoC has been created to validate the usage of zuul-distro-jobs project for dealing with RPM builds. - Ceph-ansible: initial standalone job added to test pull requests - Podman: integration with RDO software factory is done - Closed-out Train release branching in both upstream and periodic realms as part of the mid-cycle technical debt. - Addressed required changes for building a CentOS8 node for upcoming distro release support across TripleO CI jobs. - All RDO third party and upstream multinode (master/train) jobs are now moved to os_tempest ansible role provided by Openstack-ansible team. - (Pushed to next sprint): Improve tests for verifying a full promotion workflow running on the staging environment. The planned work for the next sprint [1] are: - Evaluate and implement CI jobs in Zuul that deal with RPM build artifacts for ceph-ansible and podman 3rd party testing. - Design and create a PoC for individual component testing in the promotion pipeline. This effort will add an additional verification layer to check OpenStack components (compute, networking, storage, etc) with stable builds, and ease root cause determination when it breaks the code. - Continue to improve and fix the new promotion code by deploying and bootstrapping an isolated promoter server. The Ruck and Rover for this sprint are Sagi Shnaidman (sshnaidm) and Ronelle Landy (rlandy). 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-18 [2] https://etherpad.openstack.org/p/ruckroversprint18

1 0

[keystone] Federated users who wish to use CLI
by Jason Anderson 31 Oct '19

31 Oct '19

Hi all, I'm in the process of prototyping a federated Keystone using OpenID Connect, which will place ephemeral users in a group that has roles in existing projects. I was testing how it felt from the user's perspective and am confused how I'm supposed to be able to use the openstacksdk with federation. For one thing, the RC files I can download from the "API Access" section of Horizon don't seem like they work; the domain is hard-coded to "Federated", and it also uses a username/password authentication method. I can see that there is a way to use KSA to use an existing OIDC token, which I think is probably the most "user-friendly" way, but the user still has to obtain this token themselves out-of-band, which is not trivial. Has anybody else set this up for users who liked to use the CLI? Is the solution to educate users about creating application credentials instead? Thank you in advance, -- Jason Anderson Chameleon DevOps Lead Consortium for Advanced Science and Engineering, The University of Chicago Mathematics & Computer Science Division, Argonne National Laboratory

4 8

Permamently delete double / wrong openflow entry from br-tun bridge in OpenvSwitch
by Pawel Konczalski 31 Oct '19

31 Oct '19

Hi, i have a VM with ARP entries in both OpenVswitch bridges br-tun and br-int on the compute node where is it executed: ovs-ofctl dump-flows br-int | egrep "arp.*10.20.0.34" cookie=0x117c25cd8a3f96, duration=47029.368s, table=24, n_packets=20, n_bytes=840, priority=2,arp,in_port="qvod9cdff27-9c",arp_spa=10.20.0.34 actions=resubmit(,25) ovs-ofctl dump-flows br-tun | egrep "arp.*10.20.0.34" cookie=0x1597c76aa2fd74f2, duration=47015.961s, table=21, n_packets=0, n_bytes=0, priority=1,arp,dl_vlan=27,arp_tpa=10.20.0.34 ... 3e:42:af:1d,IN_PORT The unnecassary / faulty entry can by deleted manually from br-tun with: ovs-ofctl --strict del-flows br-tun "priority=1,arp,dl_vlan=27,arp_tpa=10.20.0.34" but after the port is shutdown and up again the entry will by recreated again. When the VM is moved to another compute node, only the br-int entry is created, moving back to the initial host creates both again. Any idea how to remove permamenty this br-tun entry? Thanks Pawel

1 0

[nova] No meeting during summit/PTG week
by Eric Fried 31 Oct '19

31 Oct '19

And that's all I have to say about that. https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting

1 0