An OCI process running on the compute instance in OCI is using high continuously increasing amount of memory and swap
This approximately started 10 days and has kept increasing when it reached ~90% of available swap today. ( ~48 later from start)
This is seen happening suddenly in multiple OCI instance starting at almost the same time on all.
OS flavour: Rocky linux 9.3
The process is
/usr/bin/python3 /usr/lib/python3.9/site-packages/oci_utils/impl/ocid_main.py --no-daemon
It’s using 72% of memory(16GB) and
It was using 3.6GB of swap on Instance1 and 2.9GB on Instance2 of the available 4GB
Sounds like you need to report a bug here: Reporting Bugs and RFE's - Rocky Linux Wiki as it’s a SIG pay special attention to the information relating to SIG’s and where the bugs should be reported.
I’ve produced some updates for the python-oci-sdk and oci-utils packages: please give them a try and let me know if they resolve the situation. Our packages were a bit out of date, and in the oci-utils case, I’ve packaged a snapshot from a pull request made in December with significant changes.
Glad I ran across this post, also experiencing this issue starting around the same time (Mar 9). Thank you for the updated packages.
I was going the route of trying to use the updated oci-utils-0.14.0-14.el9.noarch.rpm from Oracle Linux 9.5. Would that be an option as well (since its noarch?)