Rare RPC thread leak caused by failed object retrieval
Details
| Detail name | Value |
|---|---|
| Changelog Number | 29973 |
| Type | Bug |
| Status | Resolved |
| Affected Versions | Exasol 7.1.0 |
| Fix Versions | Exasol 2026.1.0, Exasol 2025.1.11 |
| Resolution Date | 2026-05-15 |
Description
In rare error scenarios during object retrieval, internal RPC worker threads can remain blocked. Sessions depending on blocked object retrieval will hang forever and eventually time out.
Over time increased RPC thread leaks can cause significant system performance degradation as number of available RPC workers get reduced.
Workaround
There is no permanent workaround. If the issue occurs, a database restart is best mitigation.
Fix
The exception handling and resource cleanup in the object retrieval path was fixed so waiting requests are always released, even when an error occurs.
