Restored the local network

[andre.claudino_apc]

Hello Dear!

I have a problem with my APC Nobreaks. A message is displayed from nobreak logs:

05/25/2015	22:58:48	UPS: Restored the local network management interface-to-UPS communication.
05/25/2015	22:58:33	System: Network service started. IPv6 address FE80::2C0:B7FF:FE96:4BAD assigned by link-local autoconfiguration.
05/25/2015	22:58:30	System: Network service started. System IP is 192.168.42.143 from manually configured settings.
05/25/2015	22:58:24	System: Network Interface restarted.

I don't know why this occurs, but it's only on the VLAN 42, in the others the problem doesn't occur.

Thnk You in Advance!

[BillP]

Please review your full event log in conjunction with this knowledge base article -> http://www.schneider-electric.us/en/faqs/FA156063 

What we need to look at is why and how often the system network interface is restarting. It may be due to the watchdog mechanism as explained in the article if your VLAN 42 has a lot of network traffic or very little. 

If you're still not sure or have further questions, please provide the full event logs as instructed here so we can provide more detailed feedback specific to what's in your logs -> http://www.schneider-electric.us/en/faqs/FA156131 

[andre.claudino_apc]

We have these nobreaks in monitoring with NNM. When I added them for Syslog, the error occur.

Following attached the log files requested.

[BillP]

Hello,

I only see this message once in your event log and it was after a large amount of "duplicate IP address detected" messages which seems normal to me where it may needed to have reboot and see if the duplicate IP was present.

Did you resolve that issue? It appears so. We'll have to observe moving forward. There are 4-5 messages that are normal after a reboot (system interface restart) where the system is booting and establishing communication and starting processes internally. The "restored the local network communication to UPS (or any device type the management card is inside of)" is normal and part of that sequence.

Let me know if you have any other questions but if you see another network interface restart, let's take a look then.

[andre.claudino_apc]

Hello!

In this case can be normal, but the problem occurs all the time when I added the nobreak for monitoring with Syslog Server.

I can't to do the test now, because I will have problems with the alerts.

In attached is the "Print 1" that's as I have configured the Syslog Server.

But when I send a test as the "Print 2", I received several messages informing:

UPS: Restored the local network management interface-to-UPS communication.

And the messages doesn't stop. So I have that turn off the network cable and remove the Syslog settings.

[BillP]

Hi Andre,

I haven't seen that before and it is really hard to comment further without seeing what is going on in the event log at the same time you see this issue. Do the messages repeat at a certain interval? Does the test Syslog message work properly? I would expect that if you receive this message via Syslog repeatedly, it should be repeating the same amount in the event log too which I only saw it there once. 

I also don't know if you want to try upgrading your firmware to see if the problem stops as you are several revisions behind. I am not suggesting it will definitely resolve the issue but it is worth a try and would be the next step I suggest if it were possible. Ideally, before upgrading though, I would take a look at the event log after you try the test but I understand it causes an annoyance to do that. If you were to try the test again, I'd want both the event.txt and config.ini files to check the syslog and event notification configuration together beyond what you showed in your screenshots.

If you do want to try upgrading to test, it will be a relatively major upgrade since you're on version AOS 5.1.7/application v5.1.7 and the latest is AOS 6.4.0/application 6.4.0 for most devices. This new version has a completely new web interface and many new features and fixes within it but there are things to beware of when upgrading as noted here -> http://www.schneider-electric.us/en/faqs/FA167693 

You could always go back to the older firmware if you needed to but you'll have to consider the items mentioned in that knowledge base for downgrading in addition to those for upgrading.

Firmware is available @ http://www.apc.com/us/en/tools/download/ and you can select Filter by Software/Firmware - UPS Network Management Cards and find the right application for your UPS - presumably something like a Smart-UPS or Symmetra model UPS (what you call a no break)?

[BillP]

Hi Andre,

The first thing you need to do is track down the duplicate IP address - during the time you're doing the testing, all I see is constant error messages in the event log showing the following:

System: Network service information. Detected duplicate IP address on D4 3D 7E 0A DD 48.

These messages are spamming the log and interfering with the network operation of the UPS Network Management Card likely causing it to be inaccessible and periodically reboot. This type of behavior is what you see when you have duplicate IP addresses on a network with any type of network node - erratic connectivity. You need to find out what the MAC address listed above is and make sure it does not have the same IP address as your UPS Network Management Card. Once you resolve this issue, it should be tested again to verify if the problem still occurs or if only a subset of it does.

[andre.claudino_apc]

Hi Angela!

I discovered that this MAC (D4 3D 7E 0A DD 49) is the same MAC that Syslog and NNM Server.

But the IP address is different. I don't understand why is showing as IP duplicated.

Thank You!

[BillP · Accepted solution]

Hmm, is it possible the Syslog and NNM server has multiple network interfaces on it and you only checked one that was not identical to that of the APC NMC? Or, I am not sure what OS it runs but is it possible that it was changed on your server so it is displaying differently but has not taken affect yet because it needs a reboot? (On our product, if that was the case, you'd see a yellow ! next to the log off option showing pending changes.)

These are just a few ideas/thoughts that came to mind without being there to see. You could also check in a packet capture to see what IP is using that MAC to be absolutely sure something crazy strange isn't happening.