admin@glassfish.java.net

Re: GlassFish process exit

From: Byron Nevins <byron.nevins_at_oracle.com>
Date: Fri, 29 Oct 2010 20:24:05 -0700

-Xrs should not be a problem. It just screens out signals.

Sent from my iPhone

On Oct 29, 2010, at 6:15 PM, NBW <emailnbw_at_gmail.com> wrote:

> Out of curiosity, how is this all going to play in an environment where the -Xrs option has been added to the cmd that starts the VM spawned by asadmin.bat and also added to the JVM options in domain.xml to in order to prevent the server from terminating on Windows when the user whose process it is running under logs out of the console session? Preventing that termination due to log out is critical on Windows servers with glassfish running as a service.
>
> -Noah
>
> On Fri, Oct 29, 2010 at 7:57 PM, Joe Di Pol <joe.dipol_at_oracle.com> wrote:
> Bill Shannon wrote:
> This change addresses many of the issues below, but it seems like
> it's working against 13593. Is that why you filed this:
> https://glassfish.dev.java.net/issues/show_bug.cgi?id=14304
>
> Seems to me that the --force ought to be done *outside* the app server
> process, so that we can kill it no matter what state it's in. But that
> means lots of platform-dependent code. See for instance this blog:
> http://blogs.mulesoft.org/tcat-server-restarts-extend-improve-and-automate-stock-tomcat-restarts/
>
> There's no "--force" for stop-local-instance is there? Seems
> like that would be a good place for this. It could try to
> stop the server gently, and if that didn't work shoot it.
>
> This would also give us the option to have stop-instance
> run "stop-local-instance --force" over SSH as a last resort
> for stubborn instances.
>
> Joe
>
>
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: admin-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: admin-help_at_glassfish.dev.java.net
>
>
>