ejabberd - Comments for "moving the database from an old 2.0.1rc1 installation"

I assume that means

rbrcurtis — Sat, 30 Jul 2011 18:27:50 +0000

I assume that means user_caps_resources is also no longer needed? If so, it looks like it worked:

root@jabber:/tmp# ejabberdctl debug
Attaching Erlang shell to node ejabberd@jabber.
To detach it, press: Ctrl+G, q, Return
Erlang R13B03 (erts-5.7.4) [source] [64-bit] [rq:1] [async-threads:0] [hipe] [kernel-poll:false]
Eshell V5.7.4  (abort with ^G)
(ejabberd@jabber)1> mnesia:restore("/tmp/jabber-backup", [{skip_tables, [disco_publish,user_caps]}]).
{aborted,{no_exists,user_caps_resources}}
(ejabberd@jabber)2> mnesia:restore("/tmp/jabber-backup", [{skip_tables, [disco_publish,user_caps,user_caps_resources]}]).
{atomic,[config,privacy,local_config,passwd,roster,
         last_activity,sr_user,offline_msg,route,motd,acl,s2s,vcard,
         caps_features,sr_group,mod_register_ip,vcard_search,
         motd_users,session,private_storage,iq_response]}
(ejabberd@jabber)3>

Thanks so much!

You installed ejabberd 2.1.5,

mfoss — Sat, 30 Jul 2011 16:46:28 +0000

You installed ejabberd 2.1.5, started it, it created the tables with empty content. Then you tried to import a 2.0.1 backup file. Mnesia complains that the tables disco_publish, user_caps, and maybe others don't exist.

Does Mnesia mean that they don't exist in the new DB, or in the old backup file?

Looking at the source code history, those tables are not used since ejabberd 2.1.3. So, they don't exist in the new DB.

Add user_caps also to that skip_tables option, and try again to restore the backup.

Ok, I have it up and running

rbrcurtis — Wed, 27 Jul 2011 16:09:49 +0000

Ok, I have it up and running on the new machine with the same hostname as the old machine:

root@jabber:~# ejabberdctl status
The node ejabberd@jabber is started with status: started
ejabberd 2.1.5 is running in that node
root@jabber:~# telnet localhost 5222
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.

now when I try to restore the db from a backup that I took from the current 2.0.1 machine, I get this:

root@jabber:/tmp# ejabberdctl restore /tmp/jabber-backup 
Can't restore backup from "/tmp/jabber-backup" at node ejabberd@jabber: Table disco_publish does not exist.
root@jabber:/tmp# ejabberdctl debug
Attaching Erlang shell to node ejabberd@jabber.
To detach it, press: Ctrl+G, q, Return
Erlang R13B03 (erts-5.7.4) [source] [64-bit] [rq:1] [async-threads:0] [hipe] [kernel-poll:false]
Eshell V5.7.4  (abort with ^G)
(ejabberd@jabber)1> mnesia:restore("/tmp/jabber-backup", [{skip_tables, [disco_publish]}]).
{aborted,{no_exists,user_caps}}
(ejabberd@jabber)2>

thoughts?

rbrcurtis wrote: a new ubuntu

mfoss — Wed, 27 Jul 2011 14:34:02 +0000

rbrcurtis wrote:

a new ubuntu machine, preferably 2.1.5 as thats what is in the repo
...
I can't connect to the running node with ejabberdctl as [...] there is no debug option

Ok, first of all install the ubuntu package.

Second step is to ensure the new deployment works, and you have full control over it. ejabberdctl must work. Also ejabberdctl live, debug, other commands...

Third step is to import the user data from the old server (either mnesia dump, or SQL dumps, or whatever). But don't attempt this step until you have full control over ejabberd.