That's better than the impression I'd gotten last time I investigated the
question.

Do you get more useful information at DEBUG?  And do you get the same
behavior if you wait to start 2c till 2a is fully up?
On Mon, Feb 29, 2016 at 7:08 PM, artnaseef <a...@artnaseef.com> wrote:

> Something sounds very wrong there.  The NFS lock file should prevent more
> than one broker writing to the store at a time.
>
> Is all of /var/log/activemq/activemq-data/ shared across all of the
> brokers?
>

​Hi,

Everything under
/var/log/activemq/activemq-data/amq-dev-2
​ is shared between
 3 nodes:

$ ls -l /var/log/activemq/activemq-data/amq-dev-2
total 396
-rw-r--r--. 1 activemq activemq 33554432 Feb 29 19:27 db-1.log
-rw-r--r--. 1 activemq activemq   176128 Feb 29 19:25 db.data
-rw-r--r--. 1 activemq activemq   176464 Feb 29 19:25 db.redo
-rw-r--r--. 1 activemq activemq        8 Feb 29 19:17 lock
drwxrwxr-x. 2 activemq activemq     4096 Feb 24 00:54 tmp_storage

​L​
ooking at the logs, sorting and merging the logs based on timestamp, I see
that
​
 amq-dev-2b correctly identified itself as a slave but
​that both amq-dev-1a and amq-dev-2c appear to think they were OK to
proceed as the  master.


[amq-dev-2a]
2016-02-29 19:08:25,083 [erSimpleAppMain] INFO  XBeanBrokerFactory$1
    - Refreshing org.apache.activemq.xbean.XBeanBrokerFactory$1@616d071a:
startup date [Mon Feb 29 19:08:25 PST 2016]; root of context hierarchy
[amq-dev-2b]
2016-02-29 19:08:27,162 [erSimpleAppMain] INFO  XBeanBrokerFactory$1
    - Refreshing org.apache.activemq.xbean.XBeanBrokerFactory$1@616d071a:
startup date [Mon Feb 29 19:08:27 PST 2016]; root of context hierarchy
[amq-dev-2a]
2016-02-29 19:08:28,830 [erSimpleAppMain] INFO  PListStoreImpl
    - PListStore:[/var/log/activemq/activemq-data/amq-dev-2/tmp_storage]
started
[amq-dev-2a]
2016-02-29 19:08:28,937 [erSimpleAppMain] INFO  BrokerService
   - Using Persistence Adapter:
KahaDBPersistenceAdapter[/var/log/activemq/activemq-data/amq-dev-2]
[amq-dev-2a]
2016-02-29 19:08:29,176 [JMX connector  ] INFO  ManagementContext
   - JMX consoles can connect to
service:jmx:rmi://localhost:2020/jndi/rmi://localhost:2020/jmxrmi
[amq-dev-2b]
2016-02-29 19:08:29,692 [erSimpleAppMain] INFO  PListStoreImpl
    - PListStore:[/var/log/activemq/activemq-data/amq-dev-2/tmp_storage]
started
[amq-dev-2b]
2016-02-29 19:08:29,760 [erSimpleAppMain] INFO  BrokerService
   - Using Persistence Adapter:
KahaDBPersistenceAdapter[/var/log/activemq/activemq-data/amq-dev-2]
[amq-dev-2b]
2016-02-29 19:08:29,857 [JMX connector  ] INFO  ManagementContext
   - JMX consoles can connect to
service:jmx:rmi://localhost:2020/jndi/rmi://localhost:2020/jmxrmi
[amq-dev-2c]
2016-02-29 19:08:30,285 [erSimpleAppMain] INFO  XBeanBrokerFactory$1
    - Refreshing org.apache.activemq.xbean.XBeanBrokerFactory$1@616d071a:
startup date [Mon Feb 29 19:08:30 PST 2016]; root of context hierarchy
[amq-dev-2c]
2016-02-29 19:08:32,345 [erSimpleAppMain] INFO  PListStoreImpl
    - PListStore:[/var/log/activemq/activemq-data/amq-dev-2/tmp_storage]
started
[amq-dev-2c]
2016-02-29 19:08:32,427 [erSimpleAppMain] INFO  BrokerService
   - Using Persistence Adapter:
KahaDBPersistenceAdapter[/var/log/activemq/activemq-data/amq-dev-2]
[amq-dev-2c]
2016-02-29 19:08:32,462 [JMX connector  ] INFO  ManagementContext
   - JMX consoles can connect to
service:jmx:rmi://localhost:2020/jndi/rmi://localhost:2020/jmxrmi
[amq-dev-2b]
2016-02-29 19:08:46,086 [erSimpleAppMain] INFO  SharedFileLocker
    - Database /var/log/activemq/activemq-data/amq-dev-2/lock is locked by
another server. This broker is now in slave mode waiting a lock to be
acquired
[amq-dev-2c]
2016-02-29 19:13:32,105 [erSimpleAppMain] INFO  Journal
   - ignoring zero length, partially initialised journal data file:
db-1.log number = 1 , length = 0
[amq-dev-2c]
2016-02-29 19:13:32,471 [erSimpleAppMain] WARN  MessageDatabase
   - Index corrupted. Recovering the index through journal replay.
Cause:java.lang.IllegalStateException: File size is pre allocated

Reply via email to