tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From André Warnier>
Subject Re: Apache 2.2.3 segfault with Jk connector 1.2.27
Date Wed, 17 Dec 2008 23:31:15 GMT
Rainer Jung wrote:
> On 18.12.2008 00:09, Rainer Jung wrote:
>>> Question thus : is my above first configuration invalid ?
>> Most likely not invalid, but exotic and thereby not well tested (the
>> SetHandler trick). Will try to reproduce.
> I can reproduce, but only if the request goes to a virtual server (VHost).
> I've got a nice core, will inspect now.
I use the SetHandler inside of a Location, because in production 
configurations I use other Location sections with, for instance, 
mod_perl handlers or authentication-related snippets.
And I could never figure out clearly from the documentation what the 
order of precedence was between plain JkMount/JkUnMount directives and 
these Location sections, or other Handlers.

I came upon the workaround by being intrigued by the following kind of 
lines in the mod_jk logfile :
(The initial "missing uri map" sounded strange)

[Wed Dec 17 15:52:36 2008] [3519:3081501472] [debug] 
jk_map_to_storage::mod_jk.c (3404): missing uri map for
[Wed Dec 17 15:52:36 2008] [3519:3081501472] [debug] 
jk_handler::mod_jk.c (2288): Single worker (ajp13) configuration for 
[Wed Dec 17 15:52:36 2008] [3519:3081501472] [debug] 
jk_handler::mod_jk.c (2320): Into handler jakarta-servlet worker=ajp13 
[Wed Dec 17 15:52:36 2008] [3519:3081501472] [debug] 
wc_get_worker_for_name::jk_worker.c (116): found a worker ajp13

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message