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

[telemetry] Gnocchi: Aggregates Operation Syntax
by Blom, Merlin, NMU-OI 06 Aug '19

06 Aug '19

Hey, I would like to aggregate data from the gnocchi database by using the gnocchi aggregates function of the CLI/API The documentation does not cover the operations that are available nor the syntax that has to be used: <https://gnocchi.xyz/gnocchiclient/shell.html?highlight=reaggregation#aggreg ates> https://gnocchi.xyz/gnocchiclient/shell.html?highlight=reaggregation#aggrega tes Searching for more information I found a GitHub Issue: <https://github.com/gnocchixyz/gnocchi/issues/393> https://github.com/gnocchixyz/gnocchi/issues/393 But I cannot use the syntax from that ether. My use case: I want to aggregate the vcpus hours per month, vram hours per month, . per server or project. - when an instance is stopped only storage is counted - the exact usage is used e.g. 2 vcpus between 1st and 7th day 4vcpus between 8th and last month no mean calculations Do you have detailed documentation about the gnocchi Aggregates Operation Syntax? Do you have complex examples for gnocchi aggregations? Especially when using the python bindings: conn_gnocchi.metric.aggregation(metrics="memory", query=[XXXXXXXX], resource_type='instance', groupby='original_resource_id') Can you give me advice regarding my use case? Do's and don'ts. Thank you for your help in advance! Merlin Blom

2 1

[kuryr] [ptl] [tc] Stepping down as Kuryr PTL
by Daniel Mellado 05 Aug '19

05 Aug '19

As I have taken on a new role in my company I won't be having the time to dedicate to Kuryr in order to keep as the PTL for the current cycle. I started working on the project more than two cycles ago and it has been a real pleasure for me. Helping a project grow from an idea and a set of diagrams to a production-grade service was an awesome experience and I got help from my awesome team and upstream contributors! I would like to take this opportunity to thank everyone who contributed to the success of Kuryr – either by writing code, suggesting new use cases, participating in our discussions, or helping out with Infra! Michal Dulko (irc: dulek) has been kind enough to accept replacing me as the new Kuryr PTL [1]. I’m sure he'll make an excellent work as he's knowledgeable about every piece of the code and is tightly connected to the community. I will still be around to help if needed. Please join me congratulating Michal on his new role! Best! Daniel [1] https://review.opendev.org/674624

6 5

stestr Python 2 Support
by Matthew Treinish 05 Aug '19

05 Aug '19

Hi Everyone, I just wanted to send a quick update about the state of python 2 support in stestr, since OpenStack is the largest user of the project. With the recent release of stestr 2.4.0 we've officially deprecated the python 2.7 support in stestr. It will emit a DeprecationWarning whenever it's run from the CLI with python 2.7 now. The plan (which is not set in stone) is that we will be pushing a 3.0.0 release that removes the python 2 support and compat code from stestr sometime in early 2020 (definitely after the Python 2.7 EoL on Jan. 1st). I don't believe this conflicts with the current Python version support plans in OpenStack [1] but just wanted to make sure people were aware so that there are no surprises when stestr stops working with Python 2 in 3.0.0. Thanks, -Matt Treinish [1] https://governance.openstack.org/tc/resolutions/20180529-python2-deprecatio…

3 3

[neutron] Bug deputy report for week of July 29th
by Brian Haley 05 Aug '19

05 Aug '19

Hi, I was Neutron bug deputy last week. Below is a short summary about reported bugs. -Brian Critical bugs ------------- None High bugs --------- * https://bugs.launchpad.net/neutron/+bug/1838699 - Removing a subnet from DVR router also removes DVR MAC flows for other router on network - Confirmed, needs owner * https://bugs.launchpad.net/neutron/+bug/1838760 - Security groups don't work for trunk ports with iptables_hybrid fw driver - Confirmed, needs owner * https://bugs.launchpad.net/neutron/+bug/1838793 - "KeepalivedManagerTestCase" tests failing during namespace deletion - Rodolfo took ownership * https://bugs.launchpad.net/devstack/+bug/1838811 - /opt/stack/devstack/tools/outfilter.py failing in neutron functional jobs since 8/2 - https://review.opendev.org/#/c/674426/ merged Medium bugs ----------- * https://bugs.launchpad.net/neutron/+bug/1838396 - update port receive 500 - https://review.opendev.org/#/c/673486/ * https://bugs.launchpad.net/neutron/+bug/1838403 - Asymmetric floating IP notifications - Seems events for floating IPs/routers are not sent in some cases - Additional information was supplied in bug, need to reproduce - Need neutron version information - Needs owner * https://bugs.launchpad.net/neutron/+bug/1838449 - Router migrations failing in the gate - Miguel assigned to himself, needs further investigation * https://bugs.launchpad.net/neutron/+bug/1838431 - [scale issue] ovs-agent port processing time increases linearly and eventually timeouts - Confirmed, needs owner * https://bugs.launchpad.net/neutron/+bug/1838563 - Timeout in executing ovs command crash ovs agent - https://review.opendev.org/#/c/674085/ * https://bugs.launchpad.net/neutron/+bug/1838587 - request neutron with Incorrect body key return 500 - https://review.opendev.org/#/c/674153/ * https://bugs.launchpad.net/neutron/+bug/1838689 - rpc_workers default value ignores setting of api_workers - https://review.opendev.org/674125 merged, backport in progress Low bugs -------- Wishlist bugs ------------- * https://bugs.launchpad.net/neutron/+bug/1838621 - [RFE] Configure extra dhcp options via API and per network - Discussed at neutron drivers meeting 8/2 - Slawek will investigate questions that came up at meeting Invalid bugs ------------ Further triage required ----------------------- * https://bugs.launchpad.net/neutron/+bug/1838617 - ssh connection getting dropped frequently - Pike, but not much info given, asked for logs and/or a reproducer on more recent code. * https://bugs.launchpad.net/neutron/+bug/1838697 - DVR Mac conversion rules are only added for the first router a network is attached to - Asked for additional information * https://bugs.launchpad.net/neutron/+bug/1839004 - Rocky DVR-SNAT seems missing entries for conntrack marking - Looks like a possible mis-configuration - Asked for additional information

1 0

[Manila] Python3 support in Manila 3rd Party CI
by Tom Barron 05 Aug '19

05 Aug '19

We worked in the Pike release to get Manila unit tests running under Python3. In Train we have completed the work begun in Stein to get all first party functional test jobs running under Python 3. Now we need to push to get third party jobs converted to Python 3 since Train will be the last OpenStack release to keep support for Python2 and it will itself support Python3 first [1]. None of this will be news to those who attend the weekly Manila team meeting at 1500 UTC on Thursdays on #openstack-meetings-alt on freenode [2], but not every back end driver has regular representation at those meetings so we are communicating the need for third party CI job Python 3 support here as well. We are tracking this work on the Manila wiki [3], where you may also find or add tips to help other CI maintainers. Also, please feel free to follow up with me via email or on freenode #openstack-manila. -- Tom Barron email: tpb@dyncloudnet irc: tbarron [1] https://governance.openstack.org/tc/resolutions/20180529-python2-deprecatio… [2] https://wiki.openstack.org/wiki/Manila/Meetings [3] https://wiki.openstack.org/wiki/Manila/TrainCycle#Python3_Testing

1 0

[Airship-Seaworthy] Deployment of Airship-Seaworthy on Virtual Environment
by Anirudh Gupta 05 Aug '19

05 Aug '19

Hi Team, We want to test Production Ready Airship-Seaworthy in our virtual environment The link followed is https://airship-treasuremap.readthedocs.io/en/latest/seaworthy.html<https://ind01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fairship-t…> As per the document we need 6 DELL R720xd bare-metal servers: 3 control, and 3 compute nodes. But we need to deploy our setup on Virtual Environment. Does Airship-Seaworthy support Installation on Virtual Environment? We have 2 Rack Servers with Dual-CPU Intel® Xeon® E5 26xx with 16 cores each and 128 GB RAM. Is it possible that we can create Virtual Machines on them and set up the complete environment. In that case, what possible infrastructure do we require for setting up the complete setup. Looking forward for your response. Regards अनिरुद्ध गुप्ता (वरिष्ठ अभियंता) Hughes Systique Corporation D-23,24 Infocity II, Sector 33, Gurugram, Haryana 122001 DISCLAIMER: This electronic message and all of its contents, contains information which is privileged, confidential or otherwise protected from disclosure. The information contained in this electronic mail transmission is intended for use only by the individual or entity to which it is addressed. If you are not the intended recipient or may have received this electronic mail transmission in error, please notify the sender immediately and delete / destroy all copies of this electronic mail transmission without disclosing, copying, distributing, forwarding, printing or retaining any part of it. Hughes Systique accepts no responsibility for loss or damage arising from the use of the information transmitted by this email including damage from virus.

3 5

[manila] Enabling CephFS snapshot support by default
by Goutham Pacha Ravi 05 Aug '19

05 Aug '19

Hi Zorillas and interested parties, (Copying a couple of folks explicitly as they may not be part of openstack-discuss) Manila's CephFS driver has a configuration option "cephfs_enable_snapshots" that administrators can toggle in manila.conf to allow project users to take snapshots of their shares. [1][2] This option has defaulted to False, since the time the CephFS driver was introduced (Mitaka release of OpenStack Manila) IIUC, CephFS snapshots have been "stable" since the Mimic release of CephFS, which debuted in June 2018. [3] Since then, ceph deployers/administrators don't have to toggle anything on the backend to enable snapshots. So, can we consider changing the default value of the config opt in manila to "True" in the Train release? This will allow the driver to report the capability "snapshot_support" as enabled, and makes life easier for OpenStack deployers/administrators. I would also love to deprecate this option and remove it in a future release, since it isn't really necessary, read on: Users cannot arbitrarily take snapshots in manila, there are a couple of things that can be used to prevent them: Quotas and Share Types. Unless users create their shares using a share type that explicitly requests "snapshot_support", users can't snapshot their shares. So, we're covered by the service's API in case someone is afraid this has an impact on someone in the wild using newer versions of Manila with older versions of Ceph, where snapshots were unstable. Thoughts? [1] https://opendev.org/openstack/manila/src/commit/721bb70a81b80bf065705e8790b… [2] https://docs.openstack.org/manila/queens/admin/cephfs_driver.html#enabling-… [3] http://docs.ceph.com/docs/mimic/dev/cephfs-snapshots/

2 1

[nova] local ssd disk performance
by Budai Laszlo 05 Aug '19

05 Aug '19

Hello all, we have a problem with the performance of the disk IO in a KVM instance. We are trying to provision VMs with high performance SSDs. we have investigated different possibilities with different results ... 1. configure Nova to use local LVM storage (images_types = lvm) - provided the best performance, but we could not migrate our instances (seems to be a bug). 2. use cinder with lvm backend and instance locality, we could migrate the instances, but the performance is less than half of the previous case 3. mount the ssd on /var/lib/nova/instances and use the images_type = raw in nova. We could migrate, but the write performance dropped to ~20% of the images_types = lvm performance and read performance is ~65% of the lvm case. do you have any idea to improve the performance for any of the cases 2 or 3 which allows migration. Kind regards, Laszlo

4 8

[tc] fast-approve change to update voting schedule for u release
by Mohammed Naser 05 Aug '19

05 Aug '19

Hi everyone, Doug has proposed a change to make changes to the voting schedule for the U release. I will be fast tracking this change and this serves as a notification to the TC (and the community as whole). It's a trivial change which adjusts scheduling only, nothing more. The current date is already passed so we wouldn't have documentation that makes sense in the first place. https://review.opendev.org/#/c/674465/1 If anyone objects, feel free to push a revert to that patch. Thanks, Mohammed -- Mohammed Naser — vexxhost ----------------------------------------------------- D. 514-316-8872 D. 800-910-1726 ext. 200 E. mnaser(a)vexxhost.com W. http://vexxhost.com

1 0

[openstack-dev][magnum] Project updates
by feilong 05 Aug '19

05 Aug '19

Hi all, Now we're very close to the milestone-2 of Train release. I'd like to take this opportunity to update the project status of Magnum. So far, we have done some great work in this cycle which make Magnum to achieve to a higher level. The first feature I'd like to highlight is the k8s version rolling upgrade. With this feature, user can upgrade their k8s version with zero downtime for their application. Another one is auto scaling, now Magnum can support k8s cluster auto scaling by leveraging the Cluster Autoscaler (there is a OpenStack Magnum driver contributed by CERN team). Besides, Magnum now can support auto repair, aka auto healing, by using a combination of Draino, Cluster Autoscaler and Node Problem Detector. Except above work we have done, there are still a lot of cool features coming: 1. The first one I'd like to mention is the boot volume support. With this feature, user can boot k8s cluster node from volume and user can even set the volume type which is super useful if the cloud provider can support high performance storage, e.g. NVME. And a new useful label named "etcd_volume_type" will be introduced by the way so that user can set the volume type for etcd volume to get a great performance for etcd cluster. 2. Private cluster. With this feature, user can set the network, subnet and floating IP when creating cluster with a public cluster template, which is very handy when user want to use public templates but still want to create the cluster in their existing network. 3. Another super fancy feature is introducing a new auto healing controller except the Draino+CA+NPD combination. Recently, we (Catalyst Cloud) contributed our Magnum auto healer[1] to Cloud-Provider-Openstack repo and we're trying to introduce it to Magnum as another auto healing controller, it would be also very useful for end user to get a zero downtime k8s cluster. As you can see, there are quite a lot of exciting features are being introduced into Magnum, however all of them are done by such a small team. What a cool team! As you know, Kubernetes is still evolving very fast, so we're very keen to get more help from the community. We're working hard to add some new core members to our existing team, but we do need more. In short, if you're interested in provisioning Kubernetes on OpenStack, please try Magnum and feel free to pop up on #openstack-containers for any question. Cheers. -- 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 ------------------------------------------------------

3 4