Closed bbockelm closed 6 months ago
@matyasselmeci has been running this, patched into a production cache, over the course of the day today. The cache was seeing a crash every ~30 minutes before but has been running stably for about 8 hours.
All good signs.
Avoid potentially using the DB object after it has been closed. Noticed this while trying to cleanup valgrind errors in a test host.