Will Not Reconnect After Network Outage

[Rogertee]

The P164 seems to have trouble reconnecting after a network outage.  Initially, I found that what appeared to be brief glitches in the network connection, more specifically it appeared that if the ISP had a glitch or some other momentary glitch that caused a network interruption, it would cause the unit to disconnect and would not reconnect to the client-Magic RDS4.  The first remedy was to install a UPS.  That seemed to solve the problem for the most part. 

However, a power outage of approximately 45minutes at the client end (studio facility) appears to have disrupted communication so that the P164 has not reconnected.  When it "sticks" like this, I have found it necessary to power cycle the unit.  It is in a fairly remote location so a special trip for a power cycle is quite inconvenient. 

The connection is established over an open port on the firewall/router and therefore all communication is over the public internet.

In another post, I read about a time-out in the TCP connection on another model.  I presume the same feature is in operation on the P164.  Is there a way to have the unit re-open the port after some specific time?

[Jan]

The P164 always reconnects after the network outage, no matter what was its duration or reason. This is guaranteed by the design and integrated watch-dog. You will never need to travel to the remote location.

The only thing you need to do is to enable the Ethernet fail-safe monitor in the Magic RDS 4, as described in the manual. The timeout is implemented as well. Both may be enabled for faster recovery. Related forum thread is at https://pira.cz/forum/index.php?topic=709.0
Sometimes, there is useful information in the manuals

[Rogertee]

Now that you mention it, I remember reading that in the manual but did not give it much attention because the default is "Disabled" and I did not enable it.

When it is operating properly, I can use Windows Powershell to "ping" it.  As of this morning, I am unable to ping it, indicating that the TCP port is locked up. 

These are VERY reliable units and this seems to be an outlying issue.  For now it would seem a trip to the site to power cycle the unit will be necessary, but I would welcome any further suggestions you may have.

[Jan]

The Ethernet fail-safe monitor and watch-dog, if enabled, effectively does the same - resets the module as in case of power cycle, if there is no meaningful incoming data for the specified period of time. Due to simplicity of this solution, it works reliably.

Some issues of course cannot be solved from the encoder side. I have already experienced many cases like:
broken cable,
Ethernet switch stuck,
Ethernet router which apparently operates but no data can pass through,
DHCP assigning another IP address after power drop out (it is preferable to configure the IP address manually)
etc.

As the result, although it is the RDS encoder which becomes unavailable, the cause of failure is usually in another network component. Anyway, any experience is still welcome.