AnsweredAssumed Answered

cronjob for disaster recovery

Question asked by sriharsha on Jul 29, 2013
Latest reply on Jul 30, 2013 by sriharsha
Hi ,

We are running a cronjob for disaster revocery everything was running fine ,but since last friday we are facing  some issues we found this in /opt/mapr/logs/maprcli-user-5082.log it says

2013-07-29 11:01:42,541 ERROR com.mapr.baseutils.zookeeper.ZKDataRetrieval init [main]: Could not connect to ZK within: 30000 ms. Check if ZK connection defined correctly: hostname:3888. No data from ZK will be returned.

and zookeeper.log says this

2013-07-29 10:22:05,784 - INFO  [NIOServerCxn.Factory:] - Closed socket connection for client /ipaddress:37058 which had sessionid 0x3eaee94bb01cf6
2013-07-29 10:31:02,546 - INFO  [NIOServerCxn.Factory:$Factory@251] - Accepted socket connection from /
2013-07-29 10:31:02,546 - INFO  [NIOServerCxn.Factory:] - Processing srvr command from /
2013-07-29 10:31:02,547 - INFO  [Thread-20:NIOServerCnxn@1435] - Closed socket connection for client / (no session established for client)
2013-07-29 10:37:09,937 - INFO  [NIOServerCxn.Factory:$Factory@251] - Accepted socket connection from /ipaddress:37131
2013-07-29 10:37:09,941 - INFO  [NIOServerCxn.Factory:] - Client attempting to establish new session at /ipaddress:37131
2013-07-29 10:37:09,944 - INFO  [CommitProcessor:0:NIOServerCnxn@1580] - Established session 0x3eaee94bb01cf7 with negotiated timeout 30000 for client /ipaddress:37131
2013-07-29 10:37:10,482 - WARN  [NIOServerCxn.Factory:] - EndOfStreamException: Unable to read additional data from client sessionid 0x3eaee94bb01cf7, likely client has closed socket

on all nodes i verfied the qstatus on zookeeper , and status , i was able to do telnet to the port:3888, was able to ping the host

my cron commands looks like this

/opt/mapr/bin/maprcli dump cldbnodes -zkconnect hostname:3888 > $SEARCH_PATH

/opt/mapr/bin/maprcli volume dump create -name mapr.cldb.internal -dumpfile

any pointers to this are apprecited