Simultaneous disconnection of SM

Hi,

We have detected a problem in serveral APs from our network. Sometimes, once a day or every two days we have a simultaneus disconnection of most of the SM from one AP (not all the SMs). We have investigated on this and we´ve seen that all these SMs has been rebooted (we check uptime).

Do you have had a similar problem? What could be the reason of this?

This happens in some APs of our network at the same time (4 or 5 APs from 150) and this happens in 450, 450i and 450m with versions 16.0.0.1 and 15.2.

Regards,

1 Like

In the forum setting, we will need a lot more info to be able to help a situation like this... have you opened a support ticket?  The support team will ask for the data to help drill into this.

Yes,

I opened a ticket. I put in the forum too to see if someone have had a similar problem.

It´s very strange. Most of the SM from an AP rebooted at the same time.

Regards,

I had an issue last week with one of our 450i AP's on 15.2.1. All SM's dropped off at the same exact time and to fix it I rebooted the AP.

Were the devices rebooted? In my case i don´t have to reboot the AP because the SM rebooted and later connect again with the AP. I´ve updated some APs with this problem from 15.2 to 16.0.0.1 and I go on with the same problem. So the reboot in the AP didn´t solve the problem.

Anyway thank you for your comments.

Regards,

are these APs and SMs monitored by cnMaestro?

Any more info on this? We are having a similar issue after updating to 16.0.0.1 were most sm will disconnect at the same time, then link up again. Nothing reboots, just looses the link. A reboot of the AP fixed this on most AP, but there are still two seeing it happen. On one of them up to 14 times in the last 4 days.

1 Like

Same problem on a PMP 450i AP.

Partially solved downgrading version from 16.0.0.1 to 15.1.5 and now the disconnections are less frequent but still present

2 Likes

Could you please collect engineering.cgi from SM's or CNUT capture and sent it to subi.s@cambiumnetwoks.com

Done but your e-mail reurn me with "Connection timed out"

You can send to me also if you'd like... email or pm is fine.

matt @ cambiumnetworks.com

Please try to private message.

Done both!

Thanks!

Not received yet :(  ( no PM and no email ). 

I can't attach files on PM, i write to Matt

f.y.i.

Subi had a typo in his email in the previous message ( netwoks instead of networks)

it should be

subi.s@cambiumnetworks.com

1 Like

Done!

Thanks!

We are seeing what may be the same issue.  In one case a couple of weeks a PMP450 was dropping all sessions every 60 minutes, like clockwork.  We sent cnMaestro screen shots, logs from engineering.cgi and CNUT and the response was basically "Gee that's interesting. We'll contact the product group." but nothing really came back.  The AP went dead about a week later. It's still on the tower given its winter and can't climb. 


I have the same dropping problem right now on a 3.6GHZ PMP 450 with 34 subs on it.  Drops every session at least once or twice a day and sometimes they are a long time coming back.  We are 16.0.0.1 from 15.2.1 and not improved. 

And yes, sent in two sets of logs so far.  Here's a cnMaestro screen shot of the past 24 hours. 

1 Like

Hello Nick.  I believe I found your support ticket 171185 and looking at the last CNUT capture the AP was running for 56 days and only 5 of 25 SM's registration counters were below 10 and mostly less than 5.  Is it possible the issue is a cnMaestro OnPremises polling issue and the SM's are actually not losing registration?   I will try to contact you directly.

Dave