Hi I fixed the issue. I used ConnectionPools with one server each, with send and recv timeout to 20Sec. Catching Cassandra Exception. This worked for me. Thanks for the response


On Mon, Feb 6, 2012 at 5:38 PM, Sasha Dolgy <sdolgy@gmail.com> wrote:
Tamil, what is the underlying purpose you are trying to achieve?  To
have your webpages know and detect when a node is down?  To have a
monitoring tool detect when a node is down?  PHPCassa allows you to
define multiple nodes.  If one node is down, it should log information
to the webserver logs and continue to work as expected if an alternate
node is available.

Parsing the output of "nodetool ring" is OK if you want the status at
that very moment.  Something more reliable should be considered,
perhaps using JMX and a proper monitoring tool, like Nagios or

On Mon, Feb 6, 2012 at 8:59 AM, R. Verlangen <robin@us2.nl> wrote:
> You might consider writing some kind of php script that runs nodetool "ring"
> and parse the output?
> 2012/2/6 Tamil selvan R.S <tamil.3758@gmail.com>
>> Hi,
>>  What is the best way to know the cluster status via php?
>>  Currently we are trying to connect to individual cassandra instance with
>> a specified timeout and if it fails we report the node to be down.
>>  But this test remains faulty. What are the other ways to test
>> availability of nodes in cassandra cluster?
>>  How does datastax opscenter manage to  do that?
>> Regards,
>> Tamil Selvan

Sasha Dolgy