Is anyone (trying to) running the official debian packaged ejabberd on testing? Mine is not working at all, basically. The erlang version seems to be named something with 11 in it. Is this too new, I saw references to downgrading to 10 in another thread?
Looking for ejabberd docs?
To access the most up-to-date ejabberd documentation, please visit docs.ejabberd.im »
Ejabberd on Debian testing
Try to be more specific in
Try to be more specific in "is not working at all". Are there epmd and beam processes after ejabberd start? Are there log files in /var/log/ejabberd? Could you show sasl.log from there?
Its not logging anything,
Its not logging anything, thats the problem. It for some reason stops logging after awhile, but epmd/beam is still running. It also seems to close its ports. (This could be it happend only after restart, not sure)
I tried compiling erlang (0.10, following the guide) myself, and was planning on compiling ejabberd aswell, but the erlang compile failed with some error that Google was not able to help me with. :(
There were similar symptoms
There were similar symptoms in Debian bug tracking system:
Can you show the output of
and
Working now! ;)
ps axu | egrep '(beam|epmd)'
ejabberd 2944 0.0 0.0 1740 416 ? S Oct03 0:00 /usr/lib/erlang/erts-5.5.1/bin/epmd -daemon
ejabberd 6445 0.0 1.3 22500 10084 pts/5 Sl Oct19 0:17 /usr/lib/erlang/erts-5.5.1/bin/beam -- -root /usr/lib/erlang -progname erl -- -home /var/lib/ejabberd -sname ejabberd -s ejabberd -kernel inetrc "/etc/ejabberd/inetrc" -ejabberd config "/etc/ejabberd/ejabberd.cfg" log_path "/var/log/ejabberd/ejabberd.log" -sasl sasl_error_logger {file,"/var/log/ejabberd/sasl.log"} -mnesia dir "/var/lib/ejabberd" -noshell -noshell -noinput
root 5655 0.0 0.0 1956 620 pts/8 R+ 12:09 0:00 grep -E (beam|epmd)
Weirdly enough, I've done _nothing_ to the machine, but now its working again! I'm sure it will stop working sometime again though...
--- /usr/share/ejabberd/ejabberd.cfg 2006-10-08 12:12:21.000000000 +0200
+++ /etc/ejabberd/ejabberd.cfg 2006-10-19 22:51:27.000000000 +0200
@@ -6,10 +6,10 @@
%% ucf section
%% Admin user
-{acl, admin, {user, "__USER__", "__HOSTNAME__"}}.
+{acl, admin, {user, "sandos", "jabber.sandos.se"}}.
%% Hostname
-{hosts, ["__HOSTNAME__"]}.
+{hosts, ["jabber.sandos.se"]}.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
@@ -150,9 +150,9 @@
% {host, "muc.localhost", [{password, "secret"}]}]},
% Jabber ICQ Transport
-% {5555, ejabberd_service, [{ip, {127, 0, 0, 1}},
-% {access, all},
-% {hosts, ["icq.localhost", "sms.localhost"], [{password, "secret"}]}]},
+ {5555, ejabberd_service, [{ip, {127, 0, 0, 1}},
+ {access, all},
+ {hosts, ["icq.jabber.sandos.se", ], [{password, "majs"}]}]},
% AIM Transport
% {5556, ejabberd_service, [{ip, {127, 0, 0, 1}},
I wonder if this is somehow timing related. Any timing I can think of is DNS-lookups, but [icq.]jabber.sandos.se does resolve. Might be that it takes some time (hours??) to start ejabberd?
I will have to investigate a bit more I think.
It _seems_ its not saving my
It _seems_ its not saving my roster for some reason.
Revise: It was _not_ working at all. I was connecting to jabber.org instead. My server is still unusable. =(
I found a file called
I found a file called erl_crash.dump in /var/lib/ejabberd
I guess this means something crashed? :)
Show top few lines of
Show top few lines of erl_crash.dump. Also, who is the owner of /var/lib/ejabberd, of /var/log/ejabberd. Show the result of
and
I actually reinstalled the
I actually reinstalled the package, and Im now using the HIPE enabled erlang debian package. It's running flawlessly now....
I can't imagine that
I can't imagine that erlang-base-hipe can run ejabberd and erlang-base not. But anyway, happy ejabbering!
Actually, I dont think that
Actually, I dont think that hipe vs. non-hipe is the problem, either. Somehow the package was just corrupted is my guess...