AP7930 SNMP command issue - need to reboot interface

[DCIM_Support]

I have a bunch of AP7930 remoted power strips and I have been loosing control of them. I am not able to control them via DCE or by sending SNMP commands to them. When this happens, I reboot the network management  interface card and then I get all control back.

It appears to be random and I don't notice there is an issue until I go to command an outlet off.

Is there a lockout if it receives too many connections? I have two systems access the same device via the same SNMP strings. Would that pose an issue? It doesn't seem to affect 90% of my devices. But it seems to be happening more and more often.

AOS = 3.7.3

APP = 3.7.3

Regards,

J

(CID:105461491)

[DCIM_Support]

Hi John,

Here are a few questions I had upon reading your post:

1.) Is this SNMPv1 or SNMPv3? Depending on your answer, are both interfaces enabled on the PDUs?

2.) When SNMP does not work and you reboot the management interface, how are you doing this? Web, telnet, etc? Those interfaces still work okay and just SNMP does not? Can we limit the access issue to the SNMP interface?

3.) Do you have a ballpark on how frequently this happens on each PDU? Like once a day on each PDU, or once a week, etc? We did have an issue in an earlier firmware version where SNMP could timeout after about 50 days and would require a reboot of the management interface.

On the PDU, you can query it with multiple systems. But, I think of it as a little PC/CPU. If you are doing a full walk on the device repeatedly from multiple locations one after the other, then it could get bogged down. A get/set here and there from multiple locations should be fine. You could experience an issue if both systems are trying to do similar actions on the PDU at the same exact time depending on what gets/sets/walks they are doing.

Anyway, you are a few firmware revisions behind. Just as a test, I am not sure if you want to consider upgrading a few PDUs to see if it changes the behavior. The latest available for these devices is AOS 3.9.0/rpdu APP v3.7.4. (You can get it here: ftp://restrict:Kop$74!@ftp.apcc.com/restricted/hardware/nmcard/firmware/rpdu/374/apc_hw02_aos390_rpd...) 

The other thing you could do is check the PDU event log during the time you have an issue and see if it is reporting anything strange. If you can give me the time/date and post a PDU event log here, I could take a look for you. Instructions on downloading it in the preferred format are here: http://www.apc.com/support/index?page=content&country=ITB〈=en&locale=en_US&id=FA156131 

Let me know if you have any questions on what I suggested and asked.

(CID:105461519)

[DCIM_Support]

1) It is SNMP V1. Some devices are using both but in the cases I have been able to review, only SNMP V1 has been enabled to my knowledge. 2) The devices have been rebooted using telnet into the device as it is the quickest way. On the one device I was able to really play with, it seemed all other functions worked. Just the SNMP seemed to fail. I think the device is still in it's hung up state. Most others we had to reboot and move on. 3) I have not been able to determine a time frame between device hang ups. The unit reports normal until we go to control it and that is random at best. I have held up doing any further upgrading to the units because when i push 3.7.4 firmware I loose all web connectivity to the units. I have not been able to determine the reason why. Port setting, change of protocol, etc. I have looked though the event log and haven't seen anything I could relate to this issue. There is only login/out events and some other minor items like set date/time.

(CID:105461527)

[DCIM_Support]

Hi John, Nothing jumps out at me with what we've discussed so far beyond it being related to an issue we saw at 49.7 days and fixed it. I thought it was fixed by AOS 3.7.3 though. Next time it happens, could you make note of the device's management interface uptime prior to issuing the reboot to fix SNMP? This would be right on the main screen after you log in via Telnet or it is also in the web interface under Administration->General->About. And to clarify, when this issue happens, no OIDs respond to queries, not just outlet control SETs? (not sure if you can try to check any other OIDs or do a walk) Do you get a timeout or some other type of error message?

(CID:105461529)

[DCIM_Support]

This question is closed for comments. You're welcome to start a new topic if you have further comments on this issue.