Hi James,

I guess, the size of the shared memory is not big enough.
You can increase it using the JkShmSize directive for Apache. The value
is an integer value which is the provided shared memory size in Kilobytes.

During startup with JkLogLevel debug you get messages like:

... do_shm_open::jk_shm.c (327): Truncated shared memory to 28800
... do_shm_open::jk_shm.c (360): Initialized shared memory size=28800
free=28672 addr=0x2a98f02000
... do_shm_open_lock::jk_shm.c (266): Opened shared memory lock
/var/tmp/install/apache2p/logs/jk-runtime-status.lock
... init_jk::mod_jk.c (2562): Initialized
shm:/var/tmp/install/apache2p/logs/jk-runtime-status

from which you can see, which file system object is used for the shared
mem and which size is used. Usually the size should be big enough to
handle 64 workers (including ajp13, load balancers and lb sub workers).

You can check, the size that is being used on your system and then
double it via JkShmSize and try again. Let us know your results.

One question:

James Masson wrote:
>
> Hi,
>
> I'm working on creating a pair of Apache mod_jk load balancers for an
> Tomcat web-app farm.
>
> Redhat ES4
> Server Version:   Apache/2.0.52 (Red Hat)
> JK Version:   mod_jk/1.2.20
>
> apache-tomcat-5.5.20-1
>
> Everything works fine, until I exceed 15 load balancing workers on
> mod_jk, then I get a 500-Internal Server Error, for every web-app.
>
> Note that the 16th worker config works fine, if specified as part of the
> 15.

I don't understand, what you mean by specifying a worker as part of
another worker.

Regards,

Rainer

---------------------------------------------------------------------
To start a new topic, e-mail: users@(protected)
To unsubscribe, e-mail: users-unsubscribe@(protected)
For additional commands, e-mail: users-help@(protected)