Accelerate backup abort when no disk space is available to prevent db from becoming unresponsive
Details
| Detail name | Value |
|---|---|
| Changelog Number | 20173 |
| Type | Bug |
| Status | Resolved |
| Affected Versions | Exasol 7.1.0 |
| Fix Versions | Exasol 8.29.0, Exasol 7.1.27 |
| Resolution Date | 2024-05-21 |
Description
If a large archive volume runs out of disk space, it can take hours to cancel the running backup. This can lead to several serious problems on the database side:
- The network is used asynchronously in this situation, which can lead to poor performance on the database side.
- If a defragmentation or shrink operation is triggered in the database, the database may even stop responding because the operations have to wait until the backup is cancelled.
Workaround
In this case, the only reliable way to undo the backup is to restart the db.
Fix
The backup automatically stops if there is not enough space in the archive volume.
Attention:
This bugfix has to be explicitly enabled by running several shell commands.
Please contact Exasol support to be informed how it can be done.
