Homer,

What version are you on?

This was orginal behavior (WAY back in 3.2) I'm glad they changed it.  I
don't think the version matters, but later versions have a smarter
behavior when they detect two active CAS/CAMs

I would check network connectivity between the CAS's.  I would also
check your Spanning tree.

It sounds like the 1st CAS is coming up, since it's spanned out (Span
tree blocking) it can't contact the secondary.  So it goes active.

Spanning tree goes to forwarding state.

CAS 1 and CAS 2 see each other, and notice both are active.  Primary
always wins in this case, and Secondary backs down to standby.

Idle question.  Why are you failing over enough to notice this problem?
:-)


-----Original Message-----
From: Perfigo SecureSmart and CleanMachines Discussion List
[mailto:[log in to unmask]] On Behalf Of Homer Manila
Sent: Tuesday, April 04, 2006 4:16 PM
To: [log in to unmask]
Subject: HA behavior: expected?

All,

Just wanted to pose a question to everyone. Wanted to see if it was
common behavior for a CAS, running in virtual gateway mode with High
Availability monitoring heartbeat over eth0, to always want to stay
active on the primary server in the failover pair?  For example, if my
primary CAS is active and the secondary is on standby, rebooting the
primary forces a failover to the secondary, but the secondary only stays
active until the primary is back up and running again, at which point it
fails back over to the primary. 

I'm concerned for 2 reasons:

- this didn't happen with heartbeat being monitored over just serial(if
it failed to the other CAS, it would stay active there until it was
rebooted), and
- this causes 2 noticeable interruptions in service for the user;
granted only about 30 seconds each, but still an outage.


Thanks for any feedback!

--Homer Manila
Network Security Administrator
Office of Information Technology
American University
202-885-2209



Bowman, Jeff wrote:
> Thank you.
>
> Jeff
>
> -----Original Message-----
> From: Perfigo SecureSmart and CleanMachines Discussion List 
> [mailto:[log in to unmask]] On Behalf Of Anthony Maszeroski
> Sent: Monday, April 03, 2006 11:27 AM
> To: [log in to unmask]
> Subject: Re: Bandwidth Restriction
>
> Go to : Device Management->CCA 
> Servers->Manage->Filter->Roles->Bandwidth
>
> Is Enable Bandwidth Management checked?
>
> Bowman, Jeff wrote:
>   
>> Has anyone else experienced problems with restricting bandwidth?  We 
>> have a role setup for "bandwidth abuser", however it seems that the 
>> users are unaffected by the settings for that role.  I do seem to 
>> remember someone else on the list having the same problem, but could 
>> not remember what the resolution was.
>>
>> Thanks,
>>
>> Jeff Bowman
>> Network Technician\Wireless Manager
>> Davidson College
>> (704) 894-2000
>>
>>     
>
> --
> - Anthony Maszeroski
> -----------------------------------
> Network Security Specialist
> The University of Scranton
> email : [log in to unmask]
> phone : 570-941-4226
> -----------------------------------
>