RE: [RadiusNT] HELP

greg Lowthian ( (no email) )
Tue, 27 Jul 1999 16:55:10 -0700

> > We talked to ascend and they assure us there is nothing wrong
> with the boxes there
> > configured the same as all the other Ascends we have with the
> same OS version. I
> > think this is a database problem but I can't figure out what.
> This was a simple
> > update to 2.5.175 and should have been smooth.
> > Keep in mind it is Emerald 2.2.
> Is it possible that you have some old records in your database with
> an AcctSessionID of some of the new ones? I've seen this happen on
> larger databases where the calls table was allowed to sit for quite
> a while and either the gear was upgraded (in which some reset their
> SessionIDs on an upgrade) or was given the IP of an retired piece of
> equipment and now they are clashing.

The SQL server lost a power supply on Sunday morning. I replaced it and
there didn't seem to be any problems. Monday I did a consolidation before I
did the upgrade of RadiusNT somewhere around that time is when it all went
bad. No hardware upgrades or change of IP's for some time although before I
did the consolidation I did a Report on all the boxes to clear Emerald
online so we could go to text only mode.

How do I check to see if the AcctSessionID's are messed up?
The database is 250Meg or so on a Duel PIII 450/384Meg (consolidation now
takes 30min instead of 14 hours).

For more information about this list (including removal) go to: