maedhroz commented on code in PR #2660:
URL: https://github.com/apache/cassandra/pull/2660#discussion_r1317851295
##########
src/java/org/apache/cassandra/net/Verb.java:
##########
@@ -159,22 +159,23 @@ public enum Verb
SCHEMA_VERSION_REQ (20, P1, rpcTimeout, MIGRATION, () ->
NoPayload.serializer, () -> SchemaVersionVerbHandler.instance,
SCHEMA_VERSION_RSP ),
// repair; mostly doesn't use callbacks and sends responses as their own
request messages, with matching sessions by uuid; should eventually harmonize
and make idiomatic
+ // for the repair messages that implement retry logic, use rpcTimeout so
the single request fails faster, then retries can be used to recover
REPAIR_RSP (100, P1, repairTimeout, REQUEST_RESPONSE, () ->
NoPayload.serializer, () -> ResponseVerbHandler.instance
),
- VALIDATION_RSP (102, P1, longTimeout, ANTI_ENTROPY, () ->
ValidationResponse.serializer, () -> RepairMessageVerbHandler.instance,
REPAIR_RSP ),
Review Comment:
This previously had a long timeout...
```
static final ToLongFunction<TimeUnit> longTimeout = units ->
Math.max(DatabaseDescriptor.getRpcTimeout(units), units.convert(5L,
TimeUnit.MINUTES));
```
...because the validation response could have a large payload/Merkle trees,
and we didn't want to time out too aggressively? I'm not sure what the impact
of that will be, although we have talked about the fact that we might want to
retry less aggressively for this verb...?
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
