Simulating Read and WriteTimeouts programatically

["Charulata Sharma (charshar)" <ch...@cisco.com>]

Hi All,

Recently I learnt that in C* driver versions 3.1.x and up,  there is a change in retry. If the query statement is not idempotent, there won't be any retry and by default the queries are considered to be non idempotent.
Hence, even though Retry policy is mentioned while setting up the C* connection, incase of a C* read and write timeouts, the driver does not retry the read/write operation.
I am now trying to test this by setting the default idempotent to true . Unfortunately I cannot change the read and write timeout settings in yaml to test this, so am trying to see if there is
An easier way to raise the ReadTimeOut and WriteTimeOutExceptions programmatically and then test. However, I haven't had much success .

Any pointers on this would be appreciated.

Thanks and Regards,
Charu

Re: Simulating Read and WriteTimeouts programatically

[Jeff Jirsa <jj...@apache.org>]

On 2017-07-06 10:08 (-0700), "Charulata Sharma (charshar)" <ch...@cisco.com> wrote: 
> An easier way to raise the ReadTimeOut and WriteTimeOutExceptions programmatically and then test. However, I haven't had much success .
> 

http://www.scassandra.org/

"Stubbed Cassandra is an open source tool that enables you to test applications that use Cassandra in a quick, deterministic way.

It is especially aimed edge case testing such as read and write timeouts.

It acts as a real Cassandra instance and can be primed to respond with results or with exceptions like read timeouts. It does this by implementing the server side of the CQL binary protocol."


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@cassandra.apache.org
For additional commands, e-mail: user-help@cassandra.apache.org

Re: Simulating Read and WriteTimeouts programatically

[Nitan Kainth <ni...@bamlabs.com>]

You can set timeout value in cqlshrc file. But you will have to overload the cluster to take these into effect.

Sent from my iPhone

> On Jul 6, 2017, at 10:08 AM, Charulata Sharma (charshar) <ch...@cisco.com> wrote:
> 
> Hi All,
>        
> Recently I learnt that in C* driver versions 3.1.x and up,  there is a change in retry. If the query statement is not idempotent, there won't be any retry and by default the queries are considered to be non idempotent.
> Hence, even though Retry policy is mentioned while setting up the C* connection, incase of a C* read and write timeouts, the driver does not retry the read/write operation.
> I am now trying to test this by setting the default idempotent to true . Unfortunately I cannot change the read and write timeout settings in yaml to test this, so am trying to see if there is
> An easier way to raise the ReadTimeOut and WriteTimeOutExceptions programmatically and then test. However, I haven't had much success . 
> 
> Any pointers on this would be appreciated.
> 
> Thanks and Regards,
> Charu
> 
> 
>