Github user matthpeterson commented on a diff in the pull request:
https://github.com/apache/accumulo/pull/270#discussion_r123045425
--- Diff:
server/base/src/main/java/org/apache/accumulo/server/util/MetadataTableUtil.java
---
@@ -894,12 +894,19 @@ public static void removeBulkLoadEntries(Connector
conn, String tableId, long ti
BatchWriter bw = conn.createBatchWriter(MetadataTable.NAME, new
BatchWriterConfig())) {
mscanner.setRange(new KeyExtent(tableId, null,
null).toMetadataRange());
mscanner.fetchColumnFamily(TabletsSection.BulkFileColumnFamily.NAME);
+ boolean shouldTrace = log.isTraceEnabled();
+ String tidString = Long.toString(tid);
for (Entry<Key,Value> entry : mscanner) {
- log.debug("Looking at entry " + entry + " with tid " + tid);
- if (Long.parseLong(entry.getValue().toString()) == tid) {
- log.debug("deleting entry " + entry);
- Mutation m = new Mutation(entry.getKey().getRow());
- m.putDelete(entry.getKey().getColumnFamily(),
entry.getKey().getColumnQualifier());
+ if (shouldTrace) {
+ log.trace("Looking at entry " + entry + " with tid " +
tidString);
--- End diff --
Yes it was related to the forced logger:
org.apache.log4j.Category.callAppenders(Category:204)
waiting for lock .. (a org.apache.log4j.Logger)
..Category.forcedLog..
..Category.debug..
... removeBulkEntries(MetadataTableUtil.java:911)
Yes, it looks like the slf4j implementation I checked avoids the forceLog
call and instead checks against the level which would not block.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
