Eric -
Yes, /etc/init.d/james is just a link to phoenix.sh - I followed the
instructions in the james wiki, including the fixed JAVA_HOME and
PHOENIX_HOME, and it works. The arguments to phoenix.sh (start/stop
etc.) are exactly those that are used in all /etc/init.d scripts - it
looks like it was written to function as a standard startup script.
The wiki instructions work around the PHOENIX_HOME detection issue by
simply commenting it out and setting it manually. I got curious and
looked at what the script actually does, and it turns out this is just a
bug, which happens both at bootup and from the command line when the
script is run using a relative path. The patch fixes this bug in any
case, so this is no longer an issue.
Don -
u might want to try adding some debugging info into the script (echo
redirected using >> into a file) with the variables and progress
messages, to see if the script is being called at all during startup,
and if so, where it's failing. This is the advice I got when
experiencing the same symptom of james apparently being completely
ignored during startup, and it helped me pinpoint the problem.
Amichai
Eric MacAdie wrote:
To: A. Rothman
Not to beat a dead horse, but is your /etc/init.d/james file really
just a link to /path/to/james/bin/phoenix.sh? Or is it a script that
calls /path/to/james/bin/phoenix.sh? On my system,
/etc/init.d/james.sh is an actual script that calls
"path/to/james/bin/phoenix.sh start". (I previously linked to it on
the list.)
I think that phoenix.sh needs to get a "start" argument in order to
actually run James, and calling phoenix.sh directly on bootup might
not do that.
Eric MacAdie
A. Rothman wrote:
I don't know anything about Centos or chkconfig, but I set up James
to start as daemon in ubuntu a short while ago, and had similar
symptoms (no trace of what's happening). I found that I have to
update the phoenix.sh script (which was linked from
/etc/init.d/james) and add an explicit export of JAVA_HOME and fix
the detection of PHOENIX_HOME as well (either override it manually in
the script, or apply the patch I submitted a couple weeks ago which
was applied to 2.3.2 which fixes it's automatic detection).
I'm not sure if this is relevant to you, but I hope it helps :-)
Amichai
Don Smith wrote:
I realize this might be more of a linux question, but my problem is
only
with James, so I'm wondering if there is something James specific I'm
missing. I've added James to initd via the chkconfig --add james
command.
[r...@web01 ~]# chkconfig --list | grep james
james 0:off 1:off 2:on 3:on 4:on 5:on 6:off
I did the virtually the same thing with Jetty, a web app server:
[r...@web01 ~]# chkconfig --list | grep jetty
jetty 0:off 1:off 2:off 3:on 4:off 5:off 6:off
The difference is that on boot Jetty is started up, but James isn't.
And
there is nothing in the James or Phoenix logs indicating there was
even an
attempt to start up. Has anyone else had success getting James to
start on
boot on Linux, like Centos5? Did you do anything different than what
I've
done?
Thanks,
Don
---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscr...@james.apache.org
For additional commands, e-mail: server-user-h...@james.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscr...@james.apache.org
For additional commands, e-mail: server-user-h...@james.apache.org
