SDFS retries on 503 when accessing to S3
Details
| Detail name | Value |
|---|---|
| Changelog Number | 27677 |
| Type | Bug |
| Status | Resolved |
| Affected Versions | Exasol 8.29.0, Exasol 2025.1.0 |
| Fix Versions | Exasol 2025.1.5, Exasol 2025.2.0 |
| Resolution Date | 2025-12-05 |
Description
When a backup to a remote S3 bucket is rate limited (due to a large number of requests being done simultaneously) it fails. This is caused by S3 responding with an http code 503. On remote volumes using the legacy cloud_backend_v1 option for accessing S3, this is not handled well and the backup aborts.
Workaround
It is possible to slow down the request rate by adding option -backupWriteThreads=4 as a database parameter (see Add Database Parameters). This should reduce the likelihood of hitting a rate limit.
Fix
On hitting a rate limit, the backup should continue normally. Now the backend that uploads to S3 will retry using a truncated exponential back-off strategy.
