Brian Anderson
Brian Anderson
Still trying to convince myself this patch isn't going to cause regressions. I am seeing ci failures occassionally but they're timeouts: ``` +- run scripts failure 2025-12-08 19:44:49.124Z error: process...
I'm about ready to give up trying to find a reproducible test case and am thinking we should just check this in and hope for the best.
CI failure doesn't seem related.
Pushing some eviction experiments to run them against ci.
I've discovered that it is not possible currently to safely deinit the internal client while leaving the io thread running. This has lead me to look even more suspiciously at...
Continued at https://github.com/tigerbeetle/tigerbeetle/pull/3509