Description of problem: During a stack deploy with a high number of object in swift, the service tripleo_swift_account_reaper can take several minutes to start creating a timeout in the deployment: ~~~ time systemctl restart tripleo_swift_account_reaper.service real 7m0.675s user 0m0.021s sys 0m0.027s ~~~ Client changed the timeout value on the server where the service is running and it worked fine after: grep Timeout /etc/systemd/system.conf DefaultTimeoutStartSec=1800s DefaultTimeoutStopSec=1800s I feel it is doing a scan when the service is started. So the number of objects (millions if its handling telemetry data) influence the start time. Maybe it should start then do scan/verification. I will share the error message in the next private comment. Also have sosreport from the node in question. If you need anything else please let me know. Version-Release number of selected component (if applicable): Red Hat OpenStack Platform release 16.1.1 GA (Train) rhosp-rhel8/openstack-swift-account:16.1-50 How reproducible: 100% Steps to Reproduce: 1. Have a lot of objects in swift 2. Try to do a stack deploy and tripleo_swift_account_reaper might take too long to start. 3. Actual results: Stack deploy failing (timeout) Expected results: Stack deploy don't fail. Additional info: sosreport on supportshell.
Verified in core_puddle: RHOS-16.1-RHEL-8-20210205.n.0
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Red Hat OpenStack Platform 16.1.4 director bug fix advisory), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2021:0817