We have just set up a new secondary DPM server to protect our Primary.  What 
I've noticed, is several times all the protection group members reset to 
"Agent not reachable" and generate a warning alert under monitoring.

If I check the agent on the Primary for the protected member, and also the 
agent for the primary in the secondary console, all are ok with no errors.  
If I inactivate the alert, then it continues fine for another few hours and 
throws the error again.

While the error is active, I can perform a CC, which works perfectly.

Is there something I'm missing with how the secondary DPM talks to the 
protected members through the Primary DPM server?

As with most things in DPM, the actual error message is completely useless 
in providing any further troubleshooting info, and internet/forums searches 
turn up nothing at all.

Anybody had similar experiences?

I might also add that this server is on the same LAN as the primary and 
there are no firewalls intervening.

Hi,
we have 30 min job which checks connectivity with all the agent, it can just 
happen that during the connectivity check a network glitch might have 
happened causing the alert to be raised which can happen frequently in a WAN 
sort of scenario. we raise the alert in the warning state saying there could 
be issues when the backup job actually runs if it’s a scenario like network 
loss or agent got uninstalled on the PS.

-- 
Thanks,
C.Prashanth [MSFT]
This posting is provided “AS IS” with no warranties, and confers no rights

"Caillin" <Caillin@discussions.microsoft.com> wrote in message 
news:DE8CDFA7-ACAC-4E3A-ACB9-C6743E6BE282@microsoft.com...
> We have just set up a new secondary DPM server to protect our Primary. 
> What
> I've noticed, is several times all the protection group members reset to
> "Agent not reachable" and generate a warning alert under monitoring.
>
> If I check the agent on the Primary for the protected member, and also the
> agent for the primary in the secondary console, all are ok with no errors.
> If I inactivate the alert, then it continues fine for another few hours 
> and
> throws the error again.
>
> While the error is active, I can perform a CC, which works perfectly.
>
> Is there something I'm missing with how the secondary DPM talks to the
> protected members through the Primary DPM server?
>
> As with most things in DPM, the actual error message is completely useless
> in providing any further troubleshooting info, and internet/forums 
> searches
> turn up nothing at all.
>
> Anybody had similar experiences?
>
> I might also add that this server is on the same LAN as the primary and
> there are no firewalls intervening. 

Found the issue.  It looks as though with the rollup update on the Primary 
server, the agents needed to be updated on the protected computers, and now 
the error isn't showing up on the secondary agent.

"C.Prashanth [MSFT]" wrote:

> Hi,
> we have 30 min job which checks connectivity with all the agent, it can just 
> happen that during the connectivity check a network glitch might have 
> happened causing the alert to be raised which can happen frequently in a WAN 
> sort of scenario. we raise the alert in the warning state saying there could 
> be issues when the backup job actually runs if it’s a scenario like network 
> loss or agent got uninstalled on the PS.
> 
> -- 
> Thanks,
> C.Prashanth [MSFT]
> This posting is provided “AS IS” with no warranties, and confers no rights
> 
> "Caillin" <Caillin@discussions.microsoft.com> wrote in message 
> news:DE8CDFA7-ACAC-4E3A-ACB9-C6743E6BE282@microsoft.com...
> > We have just set up a new secondary DPM server to protect our Primary. 
> > What
> > I've noticed, is several times all the protection group members reset to
> > "Agent not reachable" and generate a warning alert under monitoring.
> >
> > If I check the agent on the Primary for the protected member, and also the
> > agent for the primary in the secondary console, all are ok with no errors.
> > If I inactivate the alert, then it continues fine for another few hours 
> > and
> > throws the error again.
> >
> > While the error is active, I can perform a CC, which works perfectly.
> >
> > Is there something I'm missing with how the secondary DPM talks to the
> > protected members through the Primary DPM server?
> >
> > As with most things in DPM, the actual error message is completely useless
> > in providing any further troubleshooting info, and internet/forums 
> > searches
> > turn up nothing at all.
> >
> > Anybody had similar experiences?
> >
> > I might also add that this server is on the same LAN as the primary and
> > there are no firewalls intervening. 
>