Github user joshelser commented on a diff in the pull request:
https://github.com/apache/accumulo/pull/254#discussion_r114585046
--- Diff:
server/tserver/src/main/java/org/apache/accumulo/tserver/replication/AccumuloReplicaSystem.java
---
@@ -536,8 +570,21 @@ public ReplicationStats execute(Client client) throws
Exception {
// If we have some edits to send
if (0 < edits.walEdits.getEditsSize()) {
+ // Check if we are interrupted before to writing the edits
+ if (Thread.interrupted()) {
+ log.debug("Replication work interrupted before writing edits,
returning empty replication stats");
+ return new ReplicationStats(0L, 0L, 0L);
+ }
+
log.debug("Sending {} edits", edits.walEdits.getEditsSize());
long entriesReplicated = client.replicateLog(remoteTableId,
edits.walEdits, tcreds);
--- End diff --
> May still need some overall logic. If there are multiple client/RPC
calls, the sum of them could exceed the replication timeout. For example if
there are three RPC calls and each takes 9 mins and the replication timeout is
10 min, then it could run for 27 min without timing out.
I don't think that's the scope of what Adam is trying to fix. He's trying
to set, for a single WAL being replicated, is the timeout for that replication
task. So, for the number of threads that a TabletServer will devote to
replicating, each of them would have the same amount of time to try to execute
(10mins).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
