ivakegg commented on code in PR #5070:
URL: https://github.com/apache/accumulo/pull/5070#discussion_r1846659760
##########
server/manager/src/main/java/org/apache/accumulo/manager/Manager.java:
##########
@@ -1057,17 +1057,36 @@ private long balanceTablets() {
int attemptNum = 0;
do {
log.debug("Balancing for tables at level {}, times-in-loop: {}", dl,
++attemptNum);
- params = BalanceParamsImpl.fromThrift(tserverStatusForBalancerLevel,
- tserverStatusForLevel, partitionedMigrations.get(dl));
+
+ SortedMap<TabletServerId,TServerStatus> statusForBalancerLevel =
+ tserverStatusForBalancerLevel;
+ if (attemptNum > 1 && (dl == DataLevel.ROOT || dl ==
DataLevel.METADATA)) {
+ // If we are still migrating then perform a re-check on the tablet
+ // servers to make sure non of them have failed.
+ Set<TServerInstance> currentServers =
tserverSet.getCurrentServers();
+ tserverStatus = gatherTableInformation(currentServers);
+ // Create a view of the tserver status such that it only contains
the tables
+ // for this level in the tableMap.
+ tserverStatusForLevel = createTServerStatusView(dl, tserverStatus);
+ final SortedMap<TabletServerId,TServerStatus>
tserverStatusForBalancerLevel2 =
+ new TreeMap<>();
+ tserverStatusForLevel.forEach((tsi, status) ->
tserverStatusForBalancerLevel2
+ .put(new TabletServerIdImpl(tsi),
TServerStatusImpl.fromThrift(status)));
+ statusForBalancerLevel = tserverStatusForBalancerLevel2;
+ }
+
+ params = BalanceParamsImpl.fromThrift(statusForBalancerLevel,
tserverStatusForLevel,
+ partitionedMigrations.get(dl));
wait = Math.max(tabletBalancer.balance(params), wait);
- migrationsOutForLevel = params.migrationsOut().size();
- for (TabletMigration m :
checkMigrationSanity(tserverStatusForBalancerLevel.keySet(),
- params.migrationsOut())) {
+ migrationsOutForLevel = 0;
+ for (TabletMigration m :
checkMigrationSanity(statusForBalancerLevel.keySet(),
+ params.migrationsOut(), dl)) {
final KeyExtent ke = KeyExtent.fromTabletId(m.getTablet());
if (migrations.containsKey(ke)) {
log.warn("balancer requested migration more than once, skipping
{}", m);
continue;
}
+ migrationsOutForLevel++;
Review Comment:
I like this. now we are only counting the migrations that are specific to
the level.
--
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]
