- Email: email@example.com
- Registered on: 10/27/2014
- Last connection: 12/17/2018
Applied in changeset commit:b30b5e2914c5d40f760fac2e77e2d40407902aaf. Issue #4267 (MODIFIED): force_full is not honored for distribution unitsIf you attempt to sync a repo like a KS repo with distribution units, the "force_full" option is not honored. This can cause issues if you have units that are not downloaded to disk, and you want to force a sync.
I was able to repro this on 2.17.1 and 2.19 nightly. I created a tar with "skopeo copy --format v2s1 docker://busybox dir:/tmp/busybox; cd /tmp/busybox; tar cf busybox.tar *", then tried to upload the tar and it failed with the same mes... Issue #4095: Can't import containers created with BuildahHi!
I tried the tool Buildah (https://www.projectatomic.io/blog/2017/06/introducing-buildah/) to build containers without having docker binary. Containers are built (not sure if they work), but I can't upload them to the pulp repo:
-so there wouldn't be a 'celery' queue anymore, and all work would go to the dedicated queues?-
ah i did not parse it correctly at first. Ya this sounds reasonable to me as well and doesn't involve config changes. Issue #2995: Workers work exclusively on a single qpid queue before moving onto the next queueUsing Celery 3, I have the following code in a blah.py file:...
this sounds reasonable to me. Would there be pulp_workers_reserved and pulp_workers_unreserved, or would it all be handled under pulp_workers? Issue #2995: Workers work exclusively on a single qpid queue before moving onto the next queueUsing Celery 3, I have the following code in a blah.py file:...
It could be similar to https://pulp.plan.io/issues/2861. I would not be surprised if fixing that fixes this as well. Issue #2734: task group never quite finishes if pulp is restarted in the middle of task runNOTE: this bug was found in a product that uses pulp 2.8. I have not yet reproduced w/ plain pulp.
If you cancel a task in a task group, it appears that the task group is not updated.
* create a task group with many...
that sounds good to me as well. It's already done in the task anyway so there's no harm in not threading. Issue #2903: Process Recycling feature causes HTTP event notifier to be unreliableFor some non-negligible probability, the HTTP event notifier will fail to send its notification when process recycling is enabled.
1. Enable "Process Recycling":http://docs.pulpproject.org/user-guide/tuning.html#proce...
Also available in: Atom