I am using websphere version 5.1. I have noticed that the application servers are going down quite frequently. If I restart the server using WAS console, it starts working fine again. I have noticed the following error in the http_plugin.log -
[Wed Oct 29 10:49:17 2008] 0000a1c2 00000001 - ERROR: ws_common: websphereExecute: Failed to create the stream
[Wed Oct 29 10:49:17 2008] 0000a1c2 00000001 - ERROR: ws_server: serverSetFailoverStatus: Marking msnapp01_WebSphere_Portal_vmember3 down
[Wed Oct 29 10:49:17 2008] 0000a1c2 00000001 - ERROR: ws_common: websphereHandleRequest: Failed to execute the transaction to 'msnapp01_WebSphere_Portal_vmember3'on host 'msnapp01.centre1.uk.boots.com'; will try another one
[Wed Oct 29 10:50:32 2008] 00015cba 00000001 - ERROR: ws_common: websphereGetStream: Failed to connect to app server on host 'msnapp01.centre1.uk.boots.com', OS err=78
[Wed Oct 29 10:50:32 2008] 00015cba 00000001 - ERROR: ws_common: websphereExecute: Failed to create the stream
[Wed Oct 29 10:50:32 2008] 00015cba 00000001 - ERROR: ws_server: serverSetFailoverStatus: Marking msnapp01_WebSphere_Portal_vmember3 down
[Wed Oct 29 10:50:32 2008] 00015cba 00000001 - ERROR: ws_common: websphereHandleRequest: Failed to execute the transaction to 'msnapp01_WebSphere_Portal_vmember3'on host 'msnapp01.centre1.uk.boots.com'; will try another one
[Wed Oct 29 10:50:35 2008] 00016356 00000001 - ERROR: ws_common: websphereGetStream: Failed to connect to app server on host 'msnapp01.centre1.uk.boots.com', OS err=78
[Wed Oct 29 10:50:35 2008] 00016356 00000001 - ERROR: ws_common: websphereExecute: Failed to create the stream
[Wed Oct 29 10:50:35 2008] 00016356 00000001 - ERROR: ws_server: serverSetFailoverStatus: Marking msnapp01_WebSphere_Portal_vmember3 down
[Wed Oct 29 10:50:35 2008] 00016356 00000001 - ERROR: ws_common: websphereHandleRequest: Failed to execute the transaction to 'msnapp01_WebSphere_Portal_vmember3'on host 'msnapp01.centre1.uk.boots.com'; will try another one
Can someone please help?
Vikash
That literally means failed to connect() over TCP. I think the
webcontainer use a pretty big listen backlog by default, so it's not
likely that you just have too few webcontainer threads because that
would just result in them being queued up by the OS.
You didn't mention your OS, but on AIX err=78 means Connection Timed
Out. Did you set ConnectTimeout in plugin-cfg.xml too agressively?
With TRACE level logging you'd see the timing info better.
--
Eric Covener