Would it be possible to try with a 5.10 snapshot? There were some recent fixes around leveldb replication, and before trying to reproduce on my side, would be good to know you see it in the latest:
https://repository.apache.org/content/repositories/snapshots/org/apache/activemq/apache-activemq/5.10-SNAPSHOT/ On Thu, Nov 7, 2013 at 5:27 AM, triumvir <peter.na...@alfa.de> wrote: > Hi, > > I have installed a Replicated LevelDB Store with 3 zookeeper Server and 3 > activemq Server. It works well, but if I test it with jmeter and kill a > activemq-master with kill -9 ( simulate power blackout ), often it works > well but at two times following message was logged endless to the logfile of > the slave which should become master ( controlled by the width variable in > the activemq.xml ). > > *Store update waiting on 1 replica(s) to catch up to log position 0* > > At this time, no activemq Instance is started and the activemq-cluster > doesn't work :-( > > After I start the killed activemq server, the cluster works well again and > the slave become master. > > > > > > -- > View this message in context: > http://activemq.2283324.n4.nabble.com/WARN-Store-update-waiting-on-1-replica-s-to-catch-up-to-log-position-0-tp4674001.html > Sent from the ActiveMQ - User mailing list archive at Nabble.com. -- Christian Posta http://www.christianposta.com/blog twitter: @christianposta
