DanielLeens commented on PR #12081:
URL: https://github.com/apache/seatunnel/pull/12081#issuecomment-6574854525

   Thanks @Rangsh, that is the pointer I was missing.
   
   I checked `FileMapStore` at head `ef99d22d1`, and it matches what you 
describe (static read only, nothing run locally):
   
   - `store` and `delete` throw `persistenceFailure(...)` when the 
`IMapStorage` call returns `false`.
   - `storeAll` and `deleteAll` throw it when the returned failure set is 
non-empty.
   - `persistenceFailure` (around L121-L134) returns an `IMapStorageException` 
with the explicit "permanently fail-closed, restart this engine node" message 
when `isAppendPermanentlyBlocked()` is true, and a generic "failed to persist 
durably" message otherwise.
   - The Javadoc above `store` also states the write-through 
(`write-delay-seconds=0`) assumption this relies on.
   
   So a permanent block does become an observable failure to the Hazelcast 
caller instead of a silent in-memory-only update. That closes my question about 
the caller outside the storage module, and my assessment of `ef99d22d1` is 
unchanged: no source blocker on my side.
   
   The remaining gates are not source issues: Apache has no adopted `Build` 
check for this SHA (the fork run is green), and final approval or dismissal of 
the earlier changes-requested review needs a write-capable maintainer.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to