starman-modcluster - Starman launcher with mod_cluster support
starman-modcluster --mc-node-name=MyApp1 --mc-uri=http://127.0.0.1:6666 --mc-context="/myapp" --mc-alias="myapp1" --mc-host=127.0.0.1
Specified unique name of you server instance, in mod_cluster it is equivalent to JvmRoute. Max site 80 characters
Contexts that your application can server.
Specifies the hostname/IP address of a node
Specifies the port that application is hosted on, if not specified, will attempt to figure it out.
Specifies scheme that is used for your node (http/https), if not specified, will attempt to figure it out.
Virtual host alias that will be added to httpd. Defaults to StarmanServer
Domain/Load balancer name
How often in seconds server is going to communicate its status to mod_cluster. Defaults to 30 seconds.
Allow Starman to listen for mod_clusters multicast traffic for dynamic subscription
Address to listen multicast traffic on. Defaults to 126.96.36.199
Port that mod_cluster is broadcasting its information. Defaults to 23364
Stick a request to a node "yes"/"no" (Defaults to "yes")
Name of the cookie containing the session id (Defaults to "JSESSIONID")
remove the session id (cookie or parameter) when the request can't be routed to the right node "yes"/"no" (Defaults to "no")
Return an error if the request can't be routed according to JVMRoute (Defaults to "yes")
time in seconds to wait for an available worker. (Defaults to 0, no wait)
number of attemps to send the request to the backend server (Defaults to 1)
Tell how to flush the packets. On: Send immediately, Auto wait for flushwait time before sending, Off don't flush. (Defaults to "off")
Time to wait before flushing. Value in seconds (Defaults to 10)
Time to wait for a pong answer to a ping. 0 means we don't try to ping before sending. Value in secondes (Defaults to 10)
soft max inactive connection over that limit after ttl are closed. Default depends on the mpm configuration
max time in seconds to life for connection above smax. (Defaults to 60)
Max time httpd will wait for the backend connection. (Defaults to 0, no timeout)
Display Net::MCMP debug information, and show configuration options of mod_cluster in the logs.
Starman::ModCluster passes through other options given to Starman and Plack::Runner, the common backend that starmna and plackup uses, so the most options explained in
plackup -h such as
--daemonize works fine in starman-modcluster too.