[RADIATOR] Radiator unable to talk to database

[Heikki Vatiainen]

On 01/31/2011 07:39 AM, Zaeem Arshad wrote:

> We are running Radiator 4.5.1 on RHEL 4.8 kernel 2.6.9-89. Radiator
> talks to a backend Oracle 10G server. Recently, we ran into an issue
> where Radiator just could not talk to the database and logged this message:
> 
> Sat Jan 22 04:57:16 2011: ERR: Could not connect to SQL database with
> DBI->connect
> dbi:Oracle:(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=192.168.x.x)(PORT=
> 1533))(CONNECT_DATA=(SID=db123))), db123, db123:
> Sat Jan 22 04:57:16 2011: ERR: Could not connect to any SQL database.
> Request is ignored. Backing off for 600 seconds
> 
> By the look of it, it appeared to be a connectivity/database issue.

It does. You did not mention if the Oracle logs showed anything out of
ordinary happening at the same time. If the log is still available, that
might one place to check why Radiator had problems.

> However, the network connectivity was up and we could reach the database
> using Oracle client from the same machine. We suspected the JDBC drivers
> and wrote a separate java program using the same driver and environment
> variables and it connected to the database just fine. During this
> period, Radiator kept logging the above mentioned message. Multiple
> process restarts and a server reboot did not fix the issue.

At least the reboot should have taken care of any possible host IP stack
problems.

> Our Radiator installation is configured to log the messages to
> /var/log/radius/radiator.log rotated manually. The log file's size was
> around 270MB at the time of the problem. I moved the logfile and created
> a new one and restarted the Radiator process and everything started
> working fine. The log file's size is currently 356MB and the server is
> running fine. Is this a bug or a known issue?

It is not a known issue. There should not be any connection between log
file size and db problems especially if disk space was not a problem.

Can you check Oracle logs if for example, there were connectivity errors
which caused oracle to put Radiator on hold for a certain time. Those
errors might have happened before Radiator started logging the
connectivity problems shown above.

Thanks for reporting this. Please let us know if you find more
information about what happened or if the problem happens again.

Thanks!

-- 
Heikki Vatiainen <hvn at open.com.au>

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.