Details
-
Bug
-
Status: Closed
-
Blocker
-
Resolution: Fixed
-
None
-
None
-
None
-
Reviewed
Description
Many of this in the logs
WARN util.Sleeper: We slept 59993ms, this is likely due to a long garbage collecting pause and it's usually bad, see http://wiki.apache.org/hadoop/Hbase/Troubleshooting#A9
except this is the MetaScanner and the sleeper has a period of 60000ms. I have one of 30000ms in Stargate which is doing the same thing now.
I see the intent but these spurious warnings defeat the intent here. Marked as blocker as this will cause no end of users writing in on the list. Should be a test of the actual sleep time against the configured interval, not a constant.