Re: Apache Failover problem in mod_jk 1.2.15

[Jayamurugan <jayamurug@xxxxxxxxx>]

Hi

 

   Does anybody have any idea this problem? Please let me know. Thanks in advance

 

Jai

 

On 2/26/08, Jayamurugan <jayamurug@xxxxxxxxx> wrote:

Hi there,

 

      We are using Apache 2.0.59 and Jboss.4.0.4 in our production solaris 10 box. We have mod_jk 1.2.15 connector between apache and jboss. We have cluster configured with Jboss instances. One of Our jboss process stopped responding abruptly. But the process was running. Users experienced slow response time. In mod_jk.log I noticed below error

 

[Wed Jan 30 10:08:00 2008] [info]  jk_open_socket::jk_connect.c (444): connect to 10.32.107.102:8109 failed with errno=145
[Wed Jan 30 10:08:00 2008] [info]  ajp_connect_to_endpoint::jk_ajp_common.c (889): Failed opening socket to (10.32.107.102:81
09) with (errno=145)
 

         My question is if apache mark one server to error state it should failover to other server in LB. But here in my case it didnt. Could some body please help me to find the cause? Thanks in advance. Here is my workers.properties configuration

 

worker.jbpr01direct.type=lb
worker.jbpr01direct.balance_workers=jbpr01direct1,jbpr01direct2
worker.jbpr01direct.sticky_session=1

worker.jbpr01direct1.type=ajp13
worker.jbpr01direct1.host=jbpr01direct1
worker.jbpr01direct1.port=8109
worker.jbpr01direct1.lbfactor=50
worker.jbpr01direct1.cache_timeout=600
worker.jbpr01direct1.socket_keepalive=1
worker.jbpr01direct1.recycle_timeout=300
worker.jbpr01direct1.jvm_route=jbpr01direct1

worker.jbpr01direct2.type=ajp13
worker.jbpr01direct2.host=jbpr01direct2
worker.jbpr01direct2.port=8109
worker.jbpr01direct2.lbfactor=50
worker.jbpr01direct2.cache_timeout=600
worker.jbpr01direct2.socket_keepalive=1
worker.jbpr01direct2.recycle_timeout=300
worker.jbpr01direct2.jvm_route=jbpr01direct2

 

Thanks

Jai