July 2021 - openstack-discuss - lists.openstack.org

[neutron] Drivers meeting - agenda for 9.07.2021
by Slawek Kaplonski 09 Jul '21

09 Jul '21

Hi, I will be on PTO tomorrow but Miguel will chair the meeting. Agenda for is at [1]. There are 3 RFEs to discuss: * https://bugs.launchpad.net/neutron/+bug/1933517 - [RFE][OVN] Create an intermediate OVS bridge between VM and intergration bridge to improve the live-migration process * https://bugs.launchpad.net/neutron/+bug/1933222 - [RFE] Add distributed datapath for metadata * https://bugs.launchpad.net/neutron/+bug/1932154 - Off-path SmartNIC Port Binding [1] https://wiki.openstack.org/wiki/Meetings/NeutronDrivers#Agenda -- Slawek Kaplonski Principal Software Engineer Red Hat

2 1

VM fail over mechanism to Remote DR site
by KK CHN 09 Jul '21

09 Jul '21

Members , I am using OpenStack Ussuri for our datacentre for virtualization requirements. we are using KVM hypervisor on Debian . We are in a requirement that our DR site is 200+ KMs apart from our office premises and connectivity over WAN. This DR site also going to deploy with Ussuri version. How we can setup the DR site so effectively so that, when VMs or Physical hosts in our DC clusters crashes or abruptly power to one physical machine got break due to electrical wire pulled out/unplugged , then how automatic failover to the DR site which is 300 miles away from our DC . what will be the solution for this setup so that automatically the VMs crashed in our DC will be served from our DR site with minimum latency or zero downtime (is zero down time possible ? kindly enlight on this or how to achieve near zero down time ) using OpenStack without any data loss ( is this possible no Data Loss ?). How ? Whats the best architecture for this using openstack, what all are the tools components need to be used ? and what type of solutions need to be deployed for the above requirement. VMs are nearly 100 nos and Linux and Windows majority of them with around 100 GB to 200 GB size for each VM. Kindly share your thoughts and expertise in this regard and enlighten me on what techniques / components of OpenStack need to be employed for achieving this. Any hints and tips and reference links highly appreciated. Thanks in advance, Kris

1 0

[tripleo] Retiring tripleo-common-tempest-plugin
by Takashi Kajinami 09 Jul '21

09 Jul '21

Hello, I have raised this in #tripleo channel yesterday but I noticed we have a tempest plugin[1] for tripleo but it has never been used actually. The repo has not been updated for a while and it's documentation still mentions very old items like python2, usage of mistral in undercloud, and so on. Because I'm not aware of any plan to use this repo for our testing, I proposed retiring repo[1]. If anybody has concern with this retirement, please give your feedback on the proposed change. [1] https://review.opendev.org/c/openstack/project-config/+/800154 Thank you, Takashi

2 1

[release] Release countdown for week R-12, Jul 12 - Jul 16
by Herve Beraud 09 Jul '21

09 Jul '21

Development Focus ----------------- The Xena-2 milestone is next week, on July 15, 2021! Xena-related specs should now be finalized so that teams can move to implementation ASAP. Some teams observe specific deadlines on the second milestone (mostly spec freezes): please refer to https://releases.openstack.org/xena/schedule.html for details. General Information ------------------- Libraries need to be released at least once per milestone period. Next week, the release team will propose releases for any library that has not been otherwise released since milestone 1. PTL's and release liaisons, please watch for these and give a +1 to acknowledge them. If there is some reason to hold off on a release, let us know that as well. A +1 would be appreciated, but if we do not hear anything at all by the end of the week, we will assume things are OK to proceed. Remember that non-library deliverables that follow the cycle-with-intermediary release model should have an intermediary release before milestone-2. Those who haven't will be proposed to switch to the cycle-with-rc model, which is more suited to deliverables that are released only once per cycle. Next week is also the deadline to freeze the contents of the final release. All new 'Xena' deliverables need to have a deliverable file in https://opendev.org/openstack/releases/src/branch/master/deliverables and need to have done a release by milestone-2. Changes proposing those deliverables for inclusion in Xena have been posted, please update them with an actual release request before the milestone-2 deadline if you plan on including that deliverable in Xena, or -1 if you need one more cycle to be ready. Upcoming Deadlines & Dates -------------------------- Xena-2 Milestone: 15 July, 2021 -- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud -----BEGIN PGP SIGNATURE----- wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----

1 0

qemu-kvm-rhev >= 2.10.0
by Pete Zhang 09 Jul '21

09 Jul '21

We have qemu-kvm-ev in yum.repos.d but it still complains of missing qemu-kvm-rhev. Any idea where we can find qemu-kvm-rhev (NOT qemu-kvm-ev), thx. Error: Execution of '/bin/yum -d 0 -e 0 -y install openstack-nova-compute' returned 1: Error: Package: 1:openstack-nova-compute-20.6.0-1.el7.noarch (local_openstack-tnrp) Requires: qemu-kvm-rhev >= 2.10.0 sh-4.2# yum list | grep qemu-kvm qemu-kvm-ev.x86_64 10:2.10.0-21.el7_5.1.1 local_openstack-tnrp qemu-kvm.x86_64 10:1.5.3-160.el7_6.1 @ORB-updates qemu-kvm-common.x86_64 10:1.5.3-160.el7_6.1 @ORB-updates qemu-kvm-tools.x86_64 10:1.5.3-160.el7_6.1 ORB-updates -- <https://smart.salesforce.com/sig/peiyong.zhang//us_mb/default/link.html>

2 1

[i40e][vf][sriov] Disabling i40evf driver at the compute level.
by Laurent Dumont 09 Jul '21

09 Jul '21

Hey everyone, We are troubleshooting some weirdness with X710 Intel cards and i40evf. Looking around, it seems that most vendors recommend disabling the i40evf driver from the compute side (and keeping the i40e driver). Is anyone aware of any funkiness if both the i40e driver and the i40evf driver is present on computes running VMs with SRIOV ports. Thanks!

3 3

[Cinder] Active/Active Support
by Swapnil Kulkarni 09 Jul '21

09 Jul '21

Hello Cinder Team, Can someone provide a reference documentation for active/active support for a Cinder Driver w.r.t. configuration and any driver changes expected? Best Regards, Swapnil

2 1

[Octavia dashboard] Impossible to fill the subnet field when creating a load balancer
by Taltavull Jean-Francois 09 Jul '21

09 Jul '21

Hi All, On my OSA Victoria deployment (on Ubuntu 20.04), the "Create Load Balancer" form does not allow me to fill the mandatory "subnet" field and therefore I can't create a load balancer with Horizon. On the other hand, Octavia works fine and I can create load balancers with the CLI, Terraform, etc. Has one of you already faced such a situation ? Best regards, Jean-François

3 6

[nova] Applying custom policies at VM creation time
by gustavo panizzo 09 Jul '21

09 Jul '21

Hello I'd like to reject VM creation or mutate the request depending on custom (business) logic for example - reject duplicate names within the same neutron network(s) - change the VM name to append a number or increase it (VM-1 requested, VM-1 already exists so it gets created as VM-2) - reject some values for particular keys on metadata (role=db rejected for project foo) How can I accomplish this? I really don't want to patch nova thanks PS: I found https://github.com/roaet/wafflehaus but it looks it hasn't been touched in years -- IRC: gfa GPG: 0x27263FA42553615F904A7EBE2A40A2ECB8DAD8D5 OLD GPG: 0x44BB1BA79F6C6333

2 2

VM failover mechanism to DR site at remote
by KK CHN 09 Jul '21

09 Jul '21

Hi list members , I am using OpenStack Ussuri for our datacentre for virtualization requirements. we are using KVM hypervisor on Debian . We are in a requirement that our DR site is 200+ KMs apart from our office premises and connectivity over WAN. This DR site also going to deploy with Ussuri version. How we can setup the DR site so effectively so that, when VMs or Physical hosts in our DC clusters crashes or abruptly power to one physical machine got break due to electrical wire pulled out/unplugged , then how automatic failover to the DR site which is 300 miles away from our DC . what will be the solution for this setup so that automatically the VMs crashed in our DC will be served from our DR site with minimum latency or zero downtime (is zero down time possible ? kindly enlight on this or how to achieve near zero down time ) using OpenStack without any data loss ( is this possible no Data Loss ?). How ? Whats the best architecture for this using openstack, what all are the tools components need to be used ? and what type of solutions need to be deployed for the above requirement. VMs are nearly 100 nos and Linux and Windows majority of them with around 100 GB to 200 GB size for each VM. Kindly share your thoughts and expertise in this regard and enlighten me on what techniques / components of OpenStack need to employed for achieving this. Any hints and tips and reference links hightly appreicated. Thanks in advance, Kris

1 0