Hi all,
I am having a problem with my loadbalancing plugin. I installed glassfish cluster on two machines, with sun java web server in front of them (on a third machine). When I tried to check my configuration with sample clusterjsp application, after stopping one instance that had served requests, the other one did not take them over. Instead, I got a violet screen showing problem with server.
Here's an excerpt from webserver's error log:
[01/Apr/2010:15:39:32] warning (12940): reports: lb.runtime: RNTM2024: Daemon
https://esb2:38181 is unhealthy.
[01/Apr/2010:15:39:32] warning (12940): reports: lb.runtime: RNTM2030: Daemon Monitor :
https://esb2:38181 : could be because daemon is down
[01/Apr/2010:15:39:55] warning (12940): reports: lb.healthchecker: HLCK3003: Listener:
https://esb2:38181 is detected to be still unHealthy in cluster: alias-cluster
[01/Apr/2010:15:40:02] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:40:25] warning (12940): reports: lb.runtime: RNTM2025: Daemon
https://esb2:38181 is healthy.
[01/Apr/2010:15:40:25] info (12940): reports: lb.healthchecker: HLCK3004: Listener:
https://esb2:38181 is detected to be healthy now in cluster : alias-cluster
[01/Apr/2010:15:40:33] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:40:50] warning (12940): reports: lb.runtime: RNTM2024: Daemon
https://esb2:38181 is unhealthy.
[01/Apr/2010:15:40:50] warning (12940): reports: lb.runtime: RNTM2030: Daemon Monitor :
https://esb2:38181 : could be because daemon is down
[01/Apr/2010:15:40:58] warning (12940): reports: lb.healthchecker: HLCK3003: Listener:
https://esb2:38181 is detected to be still unHealthy in cluster: alias-cluster
[01/Apr/2010:15:41:04] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:41:31] warning (12940): reports: lb.healthchecker: HLCK3003: Listener:
https://esb2:38181 is detected to be still unHealthy in cluster: alias-cluster
[01/Apr/2010:15:41:35] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:42:04] warning (12940): reports: lb.healthchecker: HLCK3003: Listener:
https://esb2:38181 is detected to be still unHealthy in cluster: alias-cluster
[01/Apr/2010:15:42:06] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:42:37] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:42:37] warning (12940): reports: lb.healthchecker: HLCK3003: Listener:
https://esb2:38181 is detected to be still unHealthy in cluster: alias-cluster
[01/Apr/2010:15:43:07] warning (12940): reports: lb.runtime: RNTM2025: Daemon
https://esb2:38181 is healthy.
[01/Apr/2010:15:43:07] info (12940): reports: lb.healthchecker: HLCK3004: Listener:
https://esb2:38181 is detected to be healthy now in cluster : alias-cluster
[01/Apr/2010:15:43:08] finest (12940): Validating accelerator cache entries
[01/Apr/2010:15:43:17] warning (12940): reports: lb.runtime: RNTM2024: Daemon
https://esb2:38181 is unhealthy.
[01/Apr/2010:15:43:17] warning (12940): reports: lb.runtime: RNTM2030: Daemon Monitor :
https://esb2:38181 : could be because daemon is down
These entries appear in log all the time, firstly listener becomes unhealthy, after few minutes turns healthy and so on.
What I also found out, there are periods of time, when I cannot telnet on a machine esb2 on port 38181.
My configuration is:
Glassfish 2.1.1 esb on Solaris 10
SJWS 7.0U8 B01/20/2010 on OpenSolaris 2009.06 x64
lbplugin - I used lbconfigurator 1.0 beta
What can be reason for this situation? How can I fix it?
Any help will be very appreciated.
Bart
[Message sent by forum member 'timtowdi']
http://forums.java.net/jive/thread.jspa?messageID=394894