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

OpenStack network tap interface random rxdrops for some VMs
by hai wu 22 Jul '21

22 Jul '21

There are random very slow rxdrops for certain OpenStack VMs for their network tap interfaces. Please note that this is rxdrop, NOT txdrop. I know we could tune txqueuelen and multi-queue for tap network interface txdrop issue, but is there any known way to tune for this tap network interface rxdrop issue? Thanks, Hai

3 12

[kolla][wallaby][CEPH] bluestore partions names
by Ignazio Cassano 22 Jul '21

22 Jul '21

Hello, I am going to installa ceph under kolla wallaby and I have some questions about how I must prepare partitions for bluestore on may ceph servers. Suppose I have 6 spinning disk: sdb sdc sdd sde sdf sdf And I have 1 nvme disk: nvme1 I am labelling each spinning disk as follows: sdb KOLLA_CEPH_OSD_BOOTSTRAP_1 sdc KOLLA_CEPH_OSD_BOOTSTRAP_2 and so on On nvme I create 6 partitions with the following names: KOLLA_CEPH_OSD_BOOTSTRAP_1_BS KOLLA_CEPH_OSD_BOOTSTRAP_2_BS and so on Is the above correct ? Thanks Ignazio

2 2

[cinder] Bug deputy report for week of 2021-22-07
by Sofia Enriquez 21 Jul '21

21 Jul '21

Hello, This is a bug report from 2021-14-07 to 2021-22-07. You're welcome to join the Cinder Bug Meeting today. Weekly on Wednesday at 1500 UTC on #openstack-cinder Agenda: https://etherpad.opendev.org/p/cinder-bug-squad-meeting ----------------------------------------------------------------------------------------- High - https://bugs.launchpad.net/cinder/+bug/1936854 "instance is using the wrong scsi disk after a stop/start". Unassigned. Medium - https://bugs.launchpad.net/cinder/+bug/1936909 [Gate failure] "c-bak appears to crash during tempest.api.volume.test_volumes_backup.VolumesBackupsTest.test_backup_create_attached_volume in CentOS 8 stream job" . Unassigned. - https://bugs.launchpad.net/cinder/+bug/1936431 "Failure on restoring imported Ceph backup". Unassigned. Cheers, Sofia -- L. Sofía Enriquez she/her Software Engineer Red Hat PnT <https://www.redhat.com> IRC: @enriquetaso @RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com>

1 0

CI hiccup
by Cédric Jeanneret 21 Jul '21

21 Jul '21

Hello folks, A broadcast to inform you there was some issue in both gerrit (review.opendev.org) and zuul earlier today (around 7am CET). While the issue was corrected in a very short time (thank you Opendev Infra folks!), there was a small issue with zuul - the CI queue got emptied at some point. There was a notification on the IRC: NOTICE: Due to a configuration error unfortunately the Zuul queue was lost. Please recheck any in-flight changes Sooo... yeah. That's about it :). Have a great day all, and be nice to the CI ;) Cheers, C. -- Cédric Jeanneret (He/Him/His) Sr. Software Engineer - OpenStack Platform Deployment Framework TC Red Hat EMEA https://www.redhat.com/

1 0

[zun] Disk quota feature
by Cristina Mayo Sarmiento 21 Jul '21

21 Jul '21

Hi, I've installed Openstack Wallaby of Ubuntu 20.04.2. I'm trying to specify quotas in zun containers but the zun-compute log's show this issue: > zun.common.exception.Invalid: Your host does not support disk quota feature. Does exists anyway to solve that? Currently, zun containers can "see" compute node device's mounted in /var completely: > node@compute:~$df -h > Filesystem Size Used Avail Use% Mounted on > udev 126G 0 126G 0% /dev > tmpfs 26G 2.4M 26G 1% /run > /dev/mapper/vg0-lv--root 98G 12G 82G 13% / > tmpfs 126G 0 126G 0% /dev/shm > tmpfs 5.0M 0 5.0M 0% /run/lock > tmpfs 126G 0 126G 0% /sys/fs/cgroup > /dev/sda3 976M 203M 707M 23% /boot > /dev/sda1 511M 7.9M 504M 2% /boot/efi > /dev/mapper/vg0-lv--var 2.9T 965M 2.8T 1% /var > /dev/loop0 56M 56M 0 100% /snap/core18/2074 > /dev/loop2 68M 68M 0 100% /snap/lxd/20326 > /dev/loop1 56M 56M 0 100% /snap/core18/1944 > /dev/loop3 32M 32M 0 100% /snap/snapd/10707 > /dev/loop5 33M 33M 0 100% /snap/snapd/12398 > /dev/loop4 70M 70M 0 100% /snap/lxd/19188 > tmpfs 26G 0 26G 0% /run/user/1000 > overlay 2.9T 965M 2.8T 1% /var/lib/docker/overlay2/0f88afac44d7a661c31d6b7eea85240a2ed8d84aec057f7f761be4a544d1e089/merged > shm 64M 0 64M 0% /var/lib/docker/containers/a4d60834a7ad753bc2971a446dd65c8acabf4d6df074eec846af0ddb2efc84c9/mounts/shm > test@container:~$df -h > df -h > Filesystem Size Used Available Use% Mounted on > overlay 2.9T 964.3M 2.7T 0% / > tmpfs 64.0M 0 64.0M 0% /dev > tmpfs 125.9G 0 125.9G 0% /sys/fs/cgroup > /dev/mapper/vg0-lv--var > 2.9T 964.3M 2.7T 0% /etc/resolv.conf > /dev/mapper/vg0-lv--var > 2.9T 964.3M 2.7T 0% /etc/hostname > /dev/mapper/vg0-lv--var > 2.9T 964.3M 2.7T 0% /etc/hosts > shm 64.0M 0 64.0M 0% /dev/shm > tmpfs 125.9G 0 125.9G 0% /proc/acpi > tmpfs 64.0M 0 64.0M 0% /proc/kcore > tmpfs 64.0M 0 64.0M 0% /proc/keys > tmpfs 64.0M 0 64.0M 0% /proc/timer_list > tmpfs 64.0M 0 64.0M 0% /proc/sched_debug > tmpfs 125.9G 0 125.9G 0% /proc/scsi > tmpfs 125.9G 0 125.9G 0% /sys/firmware Thanks in advance!

2 1

[puppet][TripleO] Is the OpenDev yum-puppetlabs mirror still used?
by Clark Boylan 21 Jul '21

21 Jul '21

Hello, The OpenDev team has been looking at our AFS disk utilization with an eye towards pruning and trimming out content that is no longer used. We've noticed that the yum-puppetlabs mirror [0] is fairly sizeable [1] and we're wondering if it is still being used. If it is still being used do you need all of the contents of the mirror or can we trim it back? Currently looks like there is content for older Fedora releases, SLES, and various puppet versions. It also looks like we may not have synced this recently (not sure why yet). Any info you can provide on how this is used is much appreciated. [0] https://mirror.iad.rax.opendev.org/yum-puppetlabs/ [1] https://grafana.opendev.org/d/T5zTt6PGk/afs?viewPanel=33&orgId=1 Clark

2 3

[dev][infra][tact-sig] Dropping debian-stretch images from OpenDev
by Jeremy Stanley 21 Jul '21

21 Jul '21

Due to storage pressure in a variety of our subsystems (node images, caches, package mirrors) we're looking at under-utilized aging platforms with an eye for cleaning those up to free some space and reduce our management overhead. High on that list are our debian-stretch nodes and associated mirroring. Stretch was released over four years ago, and became effectively obsolete when Buster was released two years ago. It reached normal EOL one year ago and has limited LTS for only another eleven months. We've had newer debian-buster images available in OpenDev for a couple of years already, and more recently debian-bullseye which is due to be officially released soon. Almost nothing uses the debian-stretch label in master branch configs according to codesearch, though I expect some projects are still relying on it to perform tests on a few of their older stable branches. They may also be relying on the inappropriately-named "debian-stable" nodeset which is based on debian-stretch, though again I see a very low incidence of it in master branches at least. If we were to drop our debian-stretch nodes and debian-stable nodeset definition, how much of an impact would it be on projects? Would a few weeks be enough time for those affected to prepare, or should we be thinking more on the order of months? -- Jeremy Stanley

1 0

[neutron] Virtual PTG October 2021
by Slawek Kaplonski 21 Jul '21

21 Jul '21

Hi, As You know our next PTG will held virtually in October. I just created Doodle [1] to collect best time slots for our team. Please feel it in before Tuesday 20.07 so I can book some slots before deadline which is July 21st. I also created etherpad [2] to collect ideas for the PTG sessions. So if You have anything what You would like to be discussed, please add it there. [1] https://doodle.com/poll/r4rc7z9mnxp4wwxa?utm_source=poll&utm_medium=link [2] https://etherpad.opendev.org/p/neutron-yoga-ptg -- Slawek Kaplonski Principal Software Engineer Red Hat

1 1

[openstack-ansible] Configure API endpoints on port 443
by Gilles Mocellin 21 Jul '21

21 Jul '21

Hi OSA users, Does someone have a guide to configure a deployment via openstack-ansible and have all API endpoints on HTTPS/443 ? Of course to distinguish the services, we should user either : - DNS names : https://image.endpointdomain.tld, https:// compute.endpointdomain.tld... - URI : https://endpointdomain.tld/image, https://endpointdomain.tld/ compute... As I understand, HAproxy configuration is really different for these cases only one frontend instead of on by service ?

2 2

[masakari] Yoga PTG planning
by Radosław Piliszek 21 Jul '21

21 Jul '21

Dears, As discussed during today's meeting [1], I have reserved the October 19 (Tuesday) 07-09 UTC timeslot in the Bexar Zoom room. I have also kickstarted the etherpad for the PTG. Please feel free to add your discussion topics. Finally, please remember to register for the PTG itself. [3] [1] https://meetings.opendev.org/meetings/masakari/2021/masakari.2021-07-20-06.… [2] https://etherpad.opendev.org/p/masakari-yoga-ptg [3] https://openinfra-ptg.eventbrite.com/ Kind regards, -yoctozepto

1 1