Probably it requires a mechanism to detect if a reload operation is ongoing -- there are two structures caching the records from the database, one active and one to build the new list on reload, rotating once the reload is finished. But if many reload operations happen at the same time, then it can be a conflict. I have to check the code, I don't recall who written that part and what safety checks are in place.

Normally, if the rpc reload command is sent via fifo file, it cannot be more than one executed at a time, but if you do it over http, then it can be a race.


You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.