Starting a proxy server
Starting a proxy server starts a new server process based on the process definition settings of the current proxy server configuration.
Before we start a proxy server, verify that all of the resources that the proxy requires are available. We must also start all prerequisite subsystems.
All proxy server processes are performed in the controller because we cannot configure a servant for a proxy server. Therefore, because application modules cannot be deployed into a controller, we cannot deploy an application module into a proxy server. If an application that we are running on a proxy server includes an application module, such as a module that serves an error page if a proxy server routing error occurs, that application module must be deployed to the servant of an application server with which the proxy server can communicate, instead of to the proxy server.gotcha
This procedure for starting a server also applies to restarting a server. However, if a server fails and you want the recovery functions to complete their processing prior to new processes being started on the server, we must restart the server in recovery mode.
Using one of the following options to start a proxy server.
Tasks
- For the z/OS and distributed platforms, except AIX, we can issue the startServer command
from the command line to start a single proxy server.
(Dist) We can issue the startServer command from the C:\WebSphere\AppServer\profiles\AppSrv02\bin directory.
# .\startServer.sh proxyserver1
We can issue the startServer command
from the /usr/WebSphere/AppServer/bin directory.
# ./startServer.sh proxyserver1
(ZOS) When a server is started, a new temp directory is created off of the servant process token, such as profile_root/default/temp/node/server. When the server is cleanly stopped, these temp directories are removed. However, if you infrequently perform graceful shutdowns, which happens if we cancel rather than stop the server, then these temp directories are not removed and the hierarchical file system (HFS) used for these temp directories eventually becomes full. You prevent this storage problem from occurring if you precompile your JavaServer pages (JSPs) when we install a proxy server or if we use the JspBatchCompiler function to precompile them before they are invoked.
- We can use the administrative console to start a proxy
server.
- In the administrative console, click...
Servers > Server Types > WebSphere proxy servers > .
- Select the proxy server that you want to start, and click Start.
- Confirm that you want to start the proxy server.
- View the Status value and any messages or logs to see whether the proxy server was started.
- In the administrative console, click...
- (iSeries) We can use the startServer Qshell command to start a proxy server.
- (iSeries) We can use the Submit Job (SBMJOB)
CL command to start a proxy server. Enter the following line of code to use the SBMJOB CL command.
SBMJOB JOB(jobNumber/QEJBSVR/jobName) OPTION(*CNTRLD) DELAY(delayTime)
In this example, jobNumber is the job number, jobName is the name of the proxy server job, and delayTime is the amount of time, in seconds, to wait for the job to start. We can initially set delayTime to 600 seconds and then adjust the value, if necessary, to a value that is more appropriate for the environment.
The specified proxy server starts. To verify that the proxy server was started, in the administrative console, click...
Servers > Server Types > WebSphere proxy servers > .
What to do next
If we need to start the proxy server with standard Java debugging enabled, then complete these steps:
- From the administrative console, expand Servers > Server Types > WebSphere proxy servers > proxy_server.
- From Server Infrastructure, click Java and process management > Process definition.
- Select Control.
- Click Java virtual machine.
- Select Debug mode to enable the standard Java debugger. Set Debug mode arguments, if needed, and click OK.
- Save the changes to a configuration file.
- Stop the proxy server, and restart the proxy server.