In wsgi_execute_remote(), there are error cases that share the exact same error
log text and are therefore hard to diagnose. Users would be more able to
diagnose their own problems with more descriptive messages.
These two cases --
#1:
/*
* Only allow the process group to match against a restricted
* set of processes if such a restricted set has been defined.
*/
if (config->restrict_process) {
if (!apr_table_get(config->restrict_process,
config->process_group)) {
wsgi_log_script_error(r, apr_psprintf(r->pool, "Daemon "
"process called '%s' cannot be "
"accessed by this WSGI application",
config->process_group), r->filename);
return HTTP_INTERNAL_SERVER_ERROR;
}
}
and #2:
/*
* Only allow the process group to match against a daemon
* process defined within a virtual host with the same
* server name or a daemon process defined at global server
* scope.
*/
if (group->server != r->server && group->server != wsgi_server) {
if (strcmp(group->server->server_hostname,
r->server->server_hostname) != 0) {
wsgi_log_script_error(r, apr_psprintf(r->pool, "Daemon "
"process called '%s' cannot be "
"accessed by this WSGI application",
config->process_group), r->filename);
return HTTP_INTERNAL_SERVER_ERROR;
}
.. have the same error messages.
The first could use more detail in the message to say "because the process
group ... (whatever)". Sorry, I'm not knowledgeable enough about the error
condition to suggest improved text on this one.
The second could use more detail to say "because the same daemon is used by
more than one virtual server, but the ServerName parameters of the virtual
servers do not match".
mod_wsgi is awesome, thank you so much
Original issue reported on code.google.com by stephani...@gmail.com on 13 Jun 2012 at 9:45
Original issue reported on code.google.com by
stephani...@gmail.com
on 13 Jun 2012 at 9:45