Failure to start nodes after scaling via the ConfD job "infra_db_scale" leaves an inconsistent state
Details
| Detail name | Value |
|---|---|
| Changelog Number | 24037 |
| Type | Bug |
| Status | Resolved |
| Affected Versions | Exasol 8.0.0, Exasol 2025.1.0 |
| Fix Versions | Exasol 2025.1.9, Exasol 2026.1.0, Exasol 2025.2.1 |
| Resolution Date | 2026-02-18 |
Description
If the ConfD job infra_db_scale is not able to finish successfully, a rollback to the previous state is performed. However, this rollback procedure is incomplete if the failure is produced by a timeout in the initialization step after updating the compute instances, since the nodes are not brought back to their original instance types.
Workaround
None
Fix
Additional steps are performed in the rollback procedure, in order to bring the nodes back to their previous compute instance types.