Hi All,

I set up a three node cluster and testing various features as per our 
requirements.

Not able to test the resilience successfully. We tried to test the hard 
fail over  for a single node in a cluster.

Doing  parallel  reads and writes to the cache, while doing fail over and 
the test was targeted for 1 million objects. Means 1 million reads and 
writes are parallel.

Finally observed that the read thread was stuck and not able to make any 
progress , write thread was ok , and adding items to the cache, with data 
loss.

Couchbase server software : 
couchbase-server-enterprise_2.5.1_x86_64_openssl098.rpm

Couchbase client software : Couchbase-Java-Client-1.4.4

Here is the cluster info

Node1 : xxx.xx.xx.144 4 cores , 50 GB RAM , 20 GB Hard Disk

Node2 : xxx.xx.xx.145 4 cores , 50 GB RAM , 20 GB Hard Disk

Node3 : xxx.xx.xx.146 4 cores , 50 GB RAM , 20 GB Hard Disk

Each node have default Couch base bucket, having 39 GB RAM as storage, with 
one replica, indexing and flushing enabled.

Each node configured for auto fail over of 30 seconds.

Any help is highly appreciated.

Thanks & Regards,

Subba

Contact : 310-384-2796

-- 
You received this message because you are subscribed to the Google Groups 
"Couchbase" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to