can not reproduce

x86_64

erlang R14B-1
ejabberd 2.1.5-1

I just checked on a freshly installed VM (tried both 32 bit and 64 bit).
All fail to start.
Maybe because your system is older, it has some packages that lack on a freshly installed system?

Could you suggest some way to debug this?

If I execute the command from the init script I get this crash dump

I can reproduce it on x86

erlang R14B-1
ejabberd 2.1.5-1

/etc/rc.d/ejabberd start
:: Starting ejabber daemon [FAIL]

Do these (http://www.ejabberd.im/node/872) suggestions help?

The suggestions like erl -sname qqq, erl -name qqq etc., all go on fine.
I have no reasons to believe that erlang itself is broken. I am a heavy user of couchdb, and it works flawlessly on all my machines, including the VMs from my previous comments.

I am also sure there are no networking issues like conflicting ports.

try to do

$ sudo su - ejabberd
$ erl -pa /usr/lib/ejabberd/ebin -sname ejabberd -s ejabberd -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/spool/ejabberd\" -kernel inetrc \"/etc/ejabberd/inetrc\"

and post output

[DELETED] - See below for the same command executed properly

it looks like file permissions issue. Please check that ejabberd user can read /usr/lib/ejabberd, /usr/lib/erlang, /etc/ejabberd, can write into /var/log/ejabberd, /var/spool/ejabberd

All of these are right. ejabberd user can do all of these.

Just tried downgrading to erlang-R13B04-3 - No good :(

Btw, I found out that last version of the command to start the daemon didn't work because I ran it from /root

Here is what I get when I run it from /tmp

[root@ithiria ~]# su ejabberd
[ejabberd@ithiria root]$ cd /tmp/
[ejabberd@ithiria tmp]$ erl -pa /usr/lib/ejabberd/ebin -sname ejabberd -s ejabberd -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/spool/ejabberd\" -kernel inetrc \"/etc/ejabberd/inetrc\"
Erlang R13B04 (erts-5.7.5) [source] [smp:4:4] [rq:4] [async-threads:0] [hipe] [kernel-poll:false]

Eshell V5.7.5 (abort with ^G)
(ejabberd@ithiria)1> dets: file "/var/spool/ejabberd/offline_msg.DAT" not properly closed, repairing ...
dets: file "/var/spool/ejabberd/private_storage.DAT" not properly closed, repairing ...
dets: file "/var/spool/ejabberd/pubsub_item.DAT" not properly closed, repairing ...
dets: file "/var/spool/ejabberd/vcard.DAT" not properly closed, repairing ...

=INFO REPORT==== 9-Nov-2010::20:11:36 ===
I(<0.41.0>:ejabberd_app:70) : ejabberd 2.1.5 is started in the node ejabberd@ithiria



In other words, it works

But when I put this command in the init script, it doesn't work

If I run the whole command from the current init script, it works

su ejabberd -s /bin/bash -c 'erl -pa /usr/lib/ejabberd/ebin -sname ejabberd -s ejabberd -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/spool/ejabberd\" -detached -kernel inetrc \"/etc/ejabberd/inetrc\"'

the Server starts and works normally

I added some print statements to the init script (attached)
Seems like the start command is never executed

[root@ithiria ~]# /etc/rc.d/ejabberd start
:: Starting ejabber daemon [BUSY] before pid
1956
seems like it fails
[FAIL]
[root@ithiria ~]#

it looks like rc.d script does not work if some erlang instance already running

check ejabberd-2.1.5-2

Yes, now it works with couchdb started. Thanks