In article <199808180753.QAA01503@pluto.wow.co.jp> you wrote:
>> > on Solaris 2.6 with what appears to be no problems. However, httpsd
>> > doesn't want to start up. Starting the binary prints out a message
>> > thusly:
>>
>> > "Skip first time initialisation"
>>
>> > ...and then nothing happens.
>> Please look inside Apache's error logfile.
>> There should be at least some hints why the startup failed.
> Thank your for your response.
> Unfortunately, nothing is written (at all) to the error file, nor to the
> access_log. There is also no core file.
The core file can be missing even when httpd SIGSEGV'ed. Because there are
platforms (don't know currently if Solaris 2.6 is one of them, but Linux and
FreeBSD are those) where the kernel doesn't create core dumps for daemons
running under root for security reasons. So, don't expect to find one.
But when you say nothing is written at all to the error log, this indicates
that perhaps the error log is not setup correctly. Because usually when you
have a "LogLevel info" inside httpd.conf you at least should see some Apache
startup message there. And after this Apache-SSL's errors should follow. So,
first try make sure the error log works correctly by checking the "LogLevel"
and "ErrorLog" directives and the directory and file permissions on the error
logfile.
When then there are still no errors, no core dumps, it's really stange.
Because usually Apache-SSL writes an error message in front of every exit(1).
So, either Apache starts up correctly or there have to be at least one error
message. If this is still not the case, I'm out of "remote-ideas" (then you
only can add some printf's to the code ;-)...
Greetings,
Ralf S. Engelschall
rse@engelschall.com
www.engelschall.com