Hi,
Using ActiveMQ 5.8.
I observed a behavior while restarting ActiveMQ broker.
Details: Two days ago my disk space filled completely and in some time I
have cleaned it up. KahaDB was also on the same storage. Today when I
restarted my broker it was not able to come up and throw the exception:
"broker.BrokerService.start():556 ERROR]: Failed to start Apache ActiveMQ
(seedEmbeddedBroker, null). Reason:
org.apache.activemq.protobuf.InvalidProtocolBufferException: Protocol
message contained an invalid tag (zero).
org.apache.activemq.protobuf.InvalidProtocolBufferException: Protocol
message contained an invalid tag (zero).
at
org.apache.activemq.protobuf.InvalidProtocolBufferException.invalidTag(InvalidProtocolBufferException.java:48)
"
But when I restarted again, it broker was able to come up properly.
My question is: If KahaDB is corrupted than why broker was able to come up
seconds time when I restarted broker without clearing kahaDB. Ideally It
should not come up without clearing kahaDB.
Can you please help me here or correct me If I am wrong.
P.S. : I am not setting any of the flags like checksumJournalFiles,
checkForCorruptJournalFiles, ignoreMissingJournalfiles. Default values are
here
Thanks,
Anuj
--
View this message in context:
http://activemq.2283324.n4.nabble.com/ActiveMQ-kahaDB-recovery-tp4680998.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
