action #77842
closed
openQA Project (public) - coordination #103938: [saga][epic] Scale up: Efficient handling of large storage on o3
coordination #68923: [epic] Use external videoencoder in production auto_review:"External encoder not accepting data"
Use external videoencoder in production on all o3 machines size:M
Added by okurz about 4 years ago.
Updated over 1 year ago.
Description
Motivation¶
See #68923
Acceptance criteria¶
- AC1: All o3 workers run the same external video encoder configured (exceptions allowed if required and documented) as in #77839
- AC2: All o3 workers are not overstressed by the external video encoder, i.e. not more jobs failing or incompleting due to an overstressed worker
Suggestions¶
- Research what has been done in #77839
- Ensure ffmpeg is installed on all worker machines (should be done on all of them as as of #77842#note-6)
- Configure external video encoder as documented in https://progress.opensuse.org/issues/68923#note-17, just see in #77842-10
- Monitor the CPU usage of ffmpeg e.g. via
htop and also the overall performance of each worker, e.g. check for incomplete jobs due to performance issues
- Wait for feedback for some days
- Copied from action #77839: Use external videoencoder in production on one o3 worker added
- Copied to action #77845: Use external videoencoder in production on all osd machines size:M added
- Tracker changed from coordination to action
- Status changed from Workable to New
moving all tickets without size confirmation by the team back to "New". The team should move the tickets back after estimating and agreeing on a consistent size
- Target version changed from Ready to future
- Description updated (diff)
I needed to install a new package on o3 workers anyways so I went ahead and installed ffmpeg-4 on all workers.
- Description updated (diff)
- Target version changed from future to Ready
$ ssh o3
Last login: Fri May 12 06:59:23 2023 from scar.infra.opensuse.org
okurz@ariel:~> for i in aarch64 openqaworker4 openqaworker7 openqaworker19 openqaworker20 qa-power8-3 rebel; do echo $i && ssh root@$i "grep EXTE /etc/openqa/workers.ini" ; done
aarch64
openqaworker4
openqaworker7
openqaworker19
EXTERNAL_VIDEO_ENCODER_CMD=ffmpeg -y -hide_banner -nostats -r 24 -f image2pipe -vcodec ppm -i - -pix_fmt yuv420p -c:v libvpx-vp9 -crf 35 -b:v 1500k -cpu-used 0
EXTERNAL_VIDEO_ENCODER_OUTPUT_FILE_EXTENSION=webm
#EXTERNAL_VIDEO_ENCODER_ADDITIONALLY=1
openqaworker20
EXTERNAL_VIDEO_ENCODER_CMD=ffmpeg -y -hide_banner -nostats -r 24 -f image2pipe -vcodec ppm -i - -pix_fmt yuv420p -c:v libvpx-vp9 -crf 35 -b:v 1500k -cpu-used 0
EXTERNAL_VIDEO_ENCODER_OUTPUT_FILE_EXTENSION=webm
#EXTERNAL_VIDEO_ENCODER_ADDITIONALLY=1
qa-power8-3
rebel
so we already use the external video encoder on w19+20, just use the same on the other machines.
- Subject changed from Use external videoencoder in production on all o3 machines to Use external videoencoder in production on all o3 machines size:M
- Description updated (diff)
- Status changed from New to Workable
- Related to action #129673: Change the default of the video encoder options in os-autoinst added
- Status changed from Workable to Feedback
I've ensured ffmpeg-4 is installed on all workers (it was missing on some) and configured the external video encoder on all of them (using the command line that has already been configured on workers 19 and 20). I'll checkout whether new jobs look still good.
It looks generally still good.
It looks still good.
I've just fixed the worker aarch64 where one of the cache services failed and thus no tests ran. So I'll keep this a little bit longer in feedback to wait until more jobs on that worker are executed.
Note that all the PowerPC jobs I've found have NOVIDEO=1 set. That means this change does not make a difference for the architecture/qa-power8-3 (for the better or worse).
- Status changed from Feedback to Resolved
This seems to work. If there were bigger problems we would know by now.
Also available in: Atom
PDF