we have 4 nodes cluster with JBoss SOA 5.2 on production, lately we started seeing following errors on 2 nodes
2013-07-25 04:56:05,529 WARN [org.jboss.messaging.core.impl.tx.TransactionRepository] (Thread-20) Failed to replay transaction (XID: MessagingXid (1589486249 bq:220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.54 formatID:131075 gtxid:18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.54.101.102, LocalID: 3410309207520201539) during recovery.
when this happens - nothing happens on those nodes - meaning all other threads stop and sometimes we experience timeouts on http gateway or request response services. I believe we have
JBoss Messaging 1.4.8.SP5 on the server.
My Qs is - what are the immediate steps we can take to stop these errors on servers. would cleaning data and tmp directory stop these errors and apply any JBoss Messaging patches later?
Any suggestions guys here? is there any impact if we clean data/tmp directories on servers and restart?