MFIM 100 - ES3052GP switch - Periodic complete Handset loss of connectivity

[esqi]

Hello,

I look after a iPECS 100, which has about 42 stations which are all LIP8024e units. The problem this customer (and i!) are facing is that once every 1 to 2 days the handsets lose connection to the system.

The MFIM is not randomly restarting, nor is the switch. The switch is a basic setup as there is no hybrid ports used (tagged and untagged ports on a single interface) and it is otherwise a voice only network. In the past it was uplinked to the customer's data network for system management reasons and we decided to unlink it incase it was a data flood and the switch was crashing. This of course proved fruitless and it continued to display the same symptom for all handsets.

The MFIM is running (this is Australia) 6.0cd and the switch is running Op code 1.2.6.0. Ive just upgraded the firmware on the switch to see if that assists and will be reading the changelogs of the file to see if had any bugs in the 1.2.6.0 revision - just updated to 1.2.11.0.

We have of course replaced the MFIM and switch for new items, with no improvement.

The basic port allocation is below, i was suspecting it may have over porting symptoms but i dont think it is indeed the case, as we would lose a module and get a notification of sorts via alarm etc.

MFIM 100
SLTM 4 (1 polycom unit 1 fax)
MCIM 42 handsets approx
PRIM (ive lowered the ports to 20)

Running NMS to the system gives me very little information that i can gather.

I though that maybe IGMP snooping may be the issue, so i checked the global setting and it was off, then i checked the vlan setting in its subtree and it was active, so i modified it to be inactive/disabled and today i chatted to an admin who thinks the global would override any vlan specific setting applied regardless. I did agree with that comment and it does hold true somewhat as it would be an issue with reconnecting the handsets and it should not affect everyone of them simultaneously.

Any ideas? Sort of stumped.

 

[eats4treats]

Esqi

How many switches have you got? It might not be the answer but the 42 phones could be drawing to much power from the switch when the system reaches high capacity.

Can you try to spread the load with another POE switch? I know there expensive but as a trial you can loan a 24port poe to test.

Eats

 

[esqi]

The restarts happen early morning or late afternoon when office is barely attended.

I did check the load power wise and from memory it wasnt near the max threshold.

Although I will check again, thanks.

 

[esqi]

Checked current power usage and its 84 / 400 watts used/available.

That is all idle, so i had a quick look around and the min / max is 2.10 / 2.52. lets multiply that by approx 42 handsets 105.84 max from the switch. Now im pretty sure the individual ports deliver much more (over the min/max) than this if it is required.

 

[xs650]

never had such a problem but for the 3050 series opcode 1.2.16.0 is vailable.
There is a reboot problem with a ping solved but this sound not like your problem...

I would update the switch to 1.2.16.0 and check if the phones use 1.1Ao.
When the problem still appear you could ask your distributor to open a ticket.
regards
xs

 

[esqi]

The handsets have been rebooting still, I had been talking to a colleagues about it off the cuff when it was suggested there may be a loopback happening and that the spanning tree may have something to do with it.

So I took a look at the spanning tree options and there was a global option that was active on the switch, so have disabled that for now. It could be knocking everything out for a minute or two.

Now to wait.