[Platform] Trigger trying to take a platform backup when a task has completed #7149
Comments
daniel-yb
added a commit
that referenced
this issue
Feb 25, 2021
…k has completed Summary: Every time a task completes we should optimistically take a platform backup to reduce the window of the latest backups being out of date. This diff also fixes: 1. A small bug with disabling the sync schedule when deleting an HA config. 2. Fix a few instances of file descriptors being left hanging around which would grow unboundedly until we hit the max open files limit. Test Plan: Run a few tasks and see that a one-off sync is triggered. Run tasks with no HA config and ensure there is no regression to legacy behavior. For the additional fixes in this diff: 1. Delete an HA config and ensure the config for sync schedule enabled == false. 2. Set sync to 1m and leave running for an hour or two. Run `ls -l /proc/<PID>/fd` and ensure there aren't dangling file descriptors left around. Reviewers: sanketh, sb-yb Reviewed By: sb-yb Subscribers: jenkins-bot, yugaware Differential Revision: https://phabricator.dev.yugabyte.com/D10703
polarweasel
pushed a commit
to lizayugabyte/yugabyte-db
that referenced
this issue
Mar 9, 2021
…en a task has completed Summary: Every time a task completes we should optimistically take a platform backup to reduce the window of the latest backups being out of date. This diff also fixes: 1. A small bug with disabling the sync schedule when deleting an HA config. 2. Fix a few instances of file descriptors being left hanging around which would grow unboundedly until we hit the max open files limit. Test Plan: Run a few tasks and see that a one-off sync is triggered. Run tasks with no HA config and ensure there is no regression to legacy behavior. For the additional fixes in this diff: 1. Delete an HA config and ensure the config for sync schedule enabled == false. 2. Set sync to 1m and leave running for an hour or two. Run `ls -l /proc/<PID>/fd` and ensure there aren't dangling file descriptors left around. Reviewers: sanketh, sb-yb Reviewed By: sb-yb Subscribers: jenkins-bot, yugaware Differential Revision: https://phabricator.dev.yugabyte.com/D10703
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
This goes hand in hand with work for #7148
The text was updated successfully, but these errors were encountered: