Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Zookeeper >> mail # user >> observed dataloss while replacing zookeeper severs (zookeeper 3.5.0)


Copy link to this message
-
observed dataloss while replacing zookeeper severs (zookeeper 3.5.0)
Hi,

I am experimenting with zookeeper server node replacement using 3.5.0
version.
In my setup I have 3 zookeeper servers forming a quorum and want to replace
all 3 zookeeper servers with brand new instances of servers.

If I replace all 3 nodes in quick succession (using reconfig -remove x -add
y).
Is there a possibility  zookeeper data may not be resynced completely on
the new instances of zookeeper servers due to quick succession. As I
observed that some data nodes (znodes) are not present in the zookeeper
database after replacing all members in the quorum.

Is there any way I can avoid this and how should I ensure data has been
resynced before proceeding for next server replacement.

Appreciate you help.

Thanks & Regards,
Deepak
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB