dasahcc commented on a change in pull request #1413:
URL: https://github.com/apache/helix/pull/1413#discussion_r496115113
##########
File path:
helix-core/src/main/java/org/apache/helix/controller/dataproviders/BaseControllerDataProvider.java
##########
@@ -253,9 +254,34 @@ private void refreshIdealState(final HelixDataAccessor
accessor,
private void refreshLiveInstances(final HelixDataAccessor accessor,
Set<HelixConstants.ChangeType> refreshedType) {
if
(_propertyDataChangedMap.get(HelixConstants.ChangeType.LIVE_INSTANCE).getAndSet(false))
{
+ // Keep a copy of old live instances in case of maintenance mode
+ Map<String, LiveInstance> oldLiveInstances = getLiveInstances();
_liveInstanceCache.refresh(accessor);
_updateInstanceOfflineTime = true;
refreshedType.add(HelixConstants.ChangeType.LIVE_INSTANCE);
+
+ // If maintenance mode is enabled and timeout window is specified,
filter 'new' live nodes
+ // for timed-out nodes
+ long timeOutWindow = _clusterConfig.getMaintenanceOfflineNodeTimeOut();
+ if (timeOutWindow >= 0 && isMaintenanceModeEnabled()) {
+ for (String instance : _liveInstanceCache.getPropertyMap().keySet()) {
+ // For every 'new' live node, check if it's timed-out
+ if (!oldLiveInstances.containsKey(instance) &&
isInstanceTimedOutDuringMaintenance(
Review comment:
Then that may not be the snapshot we need. Because, this change could
cause race condition and let old machine starting bootstrapping, which breaks
our rule.
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
