Idea:
- remove the LateralCacheNoWait from the cache list when its state is
STATUS_UNCONNECTABLE
- let UDP discovery add it back when it comes back online

Of course this can only be done when UDP Discovery is in place. When using
plain TCP Lateral Cache, the LateralCacheNoWait should never go into
STATUS_UNCONNECTABLE and never be removed.

Thanks,

Marek Cyzio
Equifax
(770)740-6610
[EMAIL PROTECTED]


                                                                           
             Aaron Smuts                                                   
             <[EMAIL PROTECTED]                                             
             >                                                          To 
                                       JCS Users List                      
             06/28/2006 11:08          <jcs-users@jakarta.apache.org>      
             PM                                                         cc 
                                                                           
                                                                   Subject 
             Please respond to         Re: Error messages filling up log   
             "JCS Users List"                                              
             <[EMAIL PROTECTED]                                             
               a.apache.org>                                               
                                                                           
                                                                           
                                                                           




We will definitely need this to be configurable when
we allow a port range.  This is a bit tricky, since we
want to be able to reconnect to the old server after
we reach the max retries if that server comes back up
on the same port.  Let me look into it.

Aaron

--- [EMAIL PROTECTED] wrote:

>
>
>
>
> Issue:
> - I have multiple caches running on different
> machines. I am using UDP
> Discovery + random socket selection (multiple JVM's
> on each machine).
> Things work great until I restart one of the
> machines. As soon as I do
> that, the other caches start spilling tons of error
> messages:
>
> ERROR|28 Jun 2006 08:36:37,611|Thread-25
> |socket.tcp.LateralTCPService
> |Could not create sender to [172.18.199.218:36764]
> -- Socket is null,
> cannot connect to 172.18.199.218:36764|
> ERROR|28 Jun 2006 08:36:37,611|Thread-25
> |ket.tcp.LateralTCPCacheManager|Can't fix Socket is
> null, cannot connect to
> 172.18.199.218:36764|
> ERROR|28 Jun 2006 08:36:37,612|Thread-25
> |ry.lateral.LateralCacheRestore|Can't fix Can't fix
> Socket is null, cannot
> connect to 172.18.199.218:36764|
> INFO |28 Jun 2006 08:36:38,021|Thread-27
>
|ry.lateral.LateralCacheMonitor|LateralCacheManager.instances.size()
> = 3|
> INFO |28 Jun 2006 08:36:38,022|Thread-27
> |ry.lateral.LateralCacheMonitor|
>  1- mgr.getCaches().size() = 1|
> INFO |28 Jun 2006 08:36:38,022|Thread-27
> |ry.lateral.LateralCacheMonitor|Lateral Cache No
> Wait not in error|
> INFO |28 Jun 2006 08:36:38,022|Thread-27
> |ry.lateral.LateralCacheMonitor|
>  2- mgr.getCaches().size() = 1|
> INFO |28 Jun 2006 08:36:38,022|Thread-27
> |ry.lateral.LateralCacheMonitor|found
> LateralCacheNoWait in error,
> LateralCacheNoWait  Status = 3 cache = [
>  LateralCache
>  Cache Name [eid_session_data]
>  cattr =  [172.18.199.218:36773:36762]]|
>
>
> I did some code checking and I found that there is
> no max number of retries
> defined in LateralCacheNoWait. Would it be possible
> to:
>
> a) define a new status in CacheConstants -
> STATUS_UNCONNECTABLE
> b) add a new member variable to LateralCacheNoWait -
> numberOfRepairAttempts
> c) modify LateralCacheRestore to increment
> numberOfRepairAttempts on
> LateralCacheNoWait in case repair attempt failed and
> d) modify LateralCacheMonitor to not try to repair
> LateralCacheNoWait that
> are not repairable
> e) add maxNumberOfReconnectRetries to
> ILateralCacheAttributes
>
> That would make my live so much easier... The other
> option I have is to set
> LOG4J output level on org.apache.jcs to FATAL, but
> this way I cannot see
> any other issues.
>
> Thanks,
>
> Marek Cyzio
> Equifax
> (770)740-6610
> [EMAIL PROTECTED]
> This message contains information from Equifax Inc.
> which may be
> confidential and privileged.  If you are not an
> intended recipient, please
> refrain from any disclosure, copying, distribution
> or use of this
> information and note that such actions are
> prohibited.  If you have
> received this transmission in error, please notify
> by e-mail
> [EMAIL PROTECTED]
>
>
>
---------------------------------------------------------------------
> To unsubscribe, e-mail:
> [EMAIL PROTECTED]
> For additional commands, e-mail:
> [EMAIL PROTECTED]
>
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]






This message contains information from Equifax Inc. which may be
confidential and privileged.  If you are not an intended recipient, please
refrain from any disclosure, copying, distribution or use of this
information and note that such actions are prohibited.  If you have
received this transmission in error, please notify by e-mail
[EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to