Chatted a lot with Elliot around his API wrt auth/permissions.
Chatted with Tom about cors issues.
DDM assessment presentation and attendence.
Chatted a lot with Elliot around his API wrt auth/permissions.
Chatted with Tom about cors issues.
DDM assessment presentation and attendence.
Commented on issue around site-capabilities new services, FTS completion (4133) and API ESAP backend (4160).
Submitted PR.
Refactored unittests a bit to clean them up. Need to now squash + rebase and open up a PR (https://tech.robbarnsley.com/Rucio%20development)
Met with Matt M. about perfsonar integration with monitoring infra.
Need to check list-dids is working! testing:metadata1/2/3 - can't seem to list-dids with filter on key1/2/3
Need to look over site-capabilities services. Has anything been added?
Stuck adding docker-compose interpreters due to bug. Need to fix broken aliases (docker -f and docker compose)
CAPI failure (https://confluence.skatelescope.org/display/SRCSC/Tech+Debt%3A+CAPI+infrastructure+issues) most of day.
Writing tests for postgres metadata deletion. Need to figure out how to get config as a fixture into parameterize.
Added missing functionality, got autotests to all pass.
Just need to add bits in for postgres plugin (and possibly mongo) + refactor the tests.
Chats with Yan about compute api.
Fixing issues with Cedric's changes (getting CI autotests to pass) before I add anything.
Set up a better Rucio dev environment for local testing (docker python interpreters, run configurations to include flask, "local" workflow running on github)
Fixed issue with postgres PV not binding. Underlying cinder volume had been detached from instance (presumably following a pod eviction from a node) and had not reattached. Had to get STFC to unstuck it as it wouldn't attach.
Had a look at our DNS issues:
For our AENEAS subnet we have a dhcp server and two dns servers running (automatically managed by openstack i think). The DNS servers look broken (192.168.50.13, 192.168.50.16).
For our newer subnets (e.g. srcnet) this list of dns servers looks to be more global (i.e. not created per subnet), one is running at 130.246.56.240.
so to fix:
vim /etc/systemd/resolved.conf
change:
[Resolve]
DNS=130.246.56.240 # added 15/04 RMB```
run:
```systemctl restart systemd-resolved```
# 12/04/2024 - AL
# 11/04/2024 - Support & Rucio metadata deletion - Hospital (0.5)
# 10/04/2024 - Support & Rucio metadata deletion
Supporting Yan with Permissions API work.
Starting to work on Rucio metadata deletion that Cedric started.
# 09/04/2024 - Rucio metadata deletion - Sick (0.5)
Followed up with Cedric:
"Hi Rob,
Sorry I missed your previous message. I won't have much time to work on it during the next month, so if you want to do it, you're welcome.
BTW, I think I know what happened : I implemented the logic before we moved all our queries to temporary tables, then when I rebased, the old method was removed and the code was gone."
# 08/04/2024 - Sick (0)
# 02/04/2024 to 04/04/2024 - Sick (0)