[asterisk-users] Strange res_config_odbc error messages in 1.6.1.1

[Tilghman Lesher]

On Friday 19 June 2009 10:25:15 Benny Amorsen wrote:
> When I try to use 1.6.1.1 with ODBC and MySQL, I get these:
>
> [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c: Realtime table
> supporten_sip at asterisk: column type (-9) unrecognized for column 'name'
> [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c: Realtime table
> supporten_sip at asterisk: column type (-9) unrecognized for column 'ipaddr'
> [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c: Realtime table
> supporten_sip at asterisk: column type (-9) unrecognized for column 'port'
> [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c: Realtime table
> supporten_sip at asterisk: column 'regseconds' is not long enough to contain
> realtime data (needs 11) [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c:
> Realtime table supporten_sip at asterisk: column type (-9) unrecognized for
> column 'defaultuser' [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c:
> Realtime table supporten_sip at asterisk: column type (-9) unrecognized for
> column 'fullcontact' [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c:
> Realtime table supporten_sip at asterisk: column type (-9) unrecognized for
> column 'regserver' [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c:
> Realtime table supporten_sip at asterisk: column type (-9) unrecognized for
> column 'useragent' [Jun 19 17:19:22] WARNING[5882] res_config_odbc.c:
> Realtime table supporten_sip at asterisk: column type (-9) unrecognized for
> column 'lastms'
>
> I don't understand the "column type (-9)" messages, and I'm also
> confused about regseconds, because:
>
> `regseconds` int(11) NOT NULL default '0'
>
> With 1.6.0.9 everything works fine.
>
> It is of course possible that I somehow miscompile 1.6.1.1 but not
> 1.6.0.9, but I'm trying to keep the configurations the same.
>
> Has any of you seen this?

Looks like you're using a widevarchar column, which is something I didn't plan
for.  It should be in SVN shortly, however.  On the regseconds, it's a typo in
the code (UINTEGER4 should be INTEGER4).

-- 
Tilghman