I'm not entirely sure if this is an option, but we'd like to make tbpl-dev stop eating up so much of DB disk till we can upgrade the disks on the DB. We'll probably get the disks by the end of the month and DCOps will have to go phx1 to install it, but having tbpl-dev which is using 100GB of disk right now not add more will surely help :) Thanks!
Perfectly happy for you to do the same purge as bug 779290 comment 4 but for anything >1 weeks rather than 4, if just for tbpl-dev.
ie in bug 779290 comment 4 s/1 month/1 week/ for tbpl-dev
Assignee: nobody → server-ops-database
Component: Tinderboxpushlog → Server Operations: Database
Product: Webtools → mozilla.org
QA Contact: cshields
Version: Trunk → other
Brandon, clear to purge as mentioned above in comment #2 :)
Severity: critical → normal
(In reply to Ed Morley [:edmorley UTC+0] from comment #1) > Perfectly happy for you to do the same purge as bug 779290 comment 4 but for > anything >1 weeks rather than 4, if just for tbpl-dev. Thanks Ed!
Deletes are all done. dev1 is out of the load balancer (It was already disabled in both ro and rw pools before I began) and I'm doing the optimize tables now. This will take several hours.
All done! Both dev1/dev2 have been cleaned up and looking good. We went from 90% disk usage to 49% so we recovered 40% disk space in this action alone. I've created bug 835608 to automate this purge so we won't have to do this manually in the future. Closing this one out.
Status: NEW → RESOLVED
Last Resolved: 6 years ago
Resolution: --- → FIXED
(In reply to Brandon Johnson [:cyborgshadow] from comment #6) > I've created bug 835608 to automate this purge Please may I be CCed on bug 835608 (or that bug opened up). Cheers :-)
cc'd on it. I don't want to open it up because I may reveal some internals as I write that automation.
Ok, thank you :-)
Product: mozilla.org → Data & BI Services Team
You need to log in before you can comment on or make changes to this bug.