milleruntime commented on a change in pull request #2169:
URL: https://github.com/apache/accumulo/pull/2169#discussion_r657153416
##########
File path:
server/manager/src/main/java/org/apache/accumulo/manager/tableOps/compact/cancel/CancelCompactions.java
##########
@@ -54,6 +54,19 @@ public long isReady(long tid, Manager env) throws Exception {
@Override
public Repo<Manager> call(long tid, Manager environment) throws Exception {
+
+ mutateZooKeeper(tid, tableId, environment);
+ return new FinishCancelCompaction(namespaceId, tableId);
+ }
+
+ @Override
+ public void undo(long tid, Manager env) {
+ Utils.unreserveTable(env, tableId, tid, false);
+ Utils.unreserveNamespace(env, namespaceId, tid, false);
+ }
Review comment:
That is good point. Perhaps that is why nothing is done in the undo
impl. I guess the undo doesn't always work for situations like this. This made
me think of a recent undo failure I saw though while doing seom Random walk
testing using the MutlitTable setup.
https://github.com/apache/accumulo/issues/1919. It still think it would be good
to revisit the FATE operations for things like this.
--
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.
For queries about this service, please contact Infrastructure at:
[email protected]
