MX gone faulty?

JonP
Getting noticed

MX gone faulty?

Hello Community,

 

We have had an MX75 in our new premises since July time last year. In the last month or so, we have been seeing frequent, albeit irregular drops of the primary internet connection for just a brief moment. The MX successfully fails over to the backup circuit, however the drop does kill our telephone sessions which are in progress at the time. 

 

We're looking into routing VoIP traffic over the secondary all the time, so that when the switch happens it doesn't affect calls.

 

We have a case open with Meraki support, and our ISP. Between them we have isolated that the physical L1 connection between the ISP router and the MX is dropping out. We have since placed a managed switch in between the devices, so we can pull logs from that to see which side is dropping. 

 

Fast forward to this morning and the link dropped at around 07:15 BST. Looking at the switch logs, we can see the port the MX is on dropped out for just over a minute, and had a bit of a flap during that period:

 

[Log from switch - show log Ɩ port 13]

I 28/03/22 07:16:39 00077 ports: port 13 is now off-line
I 28/03/22 07:17:21 00076 ports: port 13 is now on-line
I 28/03/22 07:17:24 00077 ports: port 13 is now off-line
I 28/03/22 07:17:27 00076 ports: port 13 is now on-line

 

Has anyone seen this sort of behaviour before? We have already replaced the uplink cable, and the MX is on its own 400VA UPS to isolate it from the mains supply. I have of course updated the support case, but wanted to see if anyone in the community had any ideas or thoughts on this?

 

Thanks all 🙂

9 Replies 9
alemabrahao
Kind of a big deal
Kind of a big deal

Is the MX in Warm Spare? Did you notice any spanning-tree events on the switch?

I am not a Cisco Meraki employee. My suggestions are based on documentation of Meraki best practices and day-to-day experience.

Please, if this post was useful, leave your kudos and mark it as solved.
JonP
Getting noticed

Unfortunately it is not in a warm spare. We have ordered another to set up warm spare, but it is 6+ months away yet.

 

The switch between the MX and the ISP Router has only two devices connected to it so no spanning tree events.

alemabrahao
Kind of a big deal
Kind of a big deal

Well, It looks like a MX issue. What version are you running?

I am not a Cisco Meraki employee. My suggestions are based on documentation of Meraki best practices and day-to-day experience.

Please, if this post was useful, leave your kudos and mark it as solved.
JonP
Getting noticed

It's an MX75 on MX 16.16

OVERKILL
Building a reputation

Are other ports on it flapping at the same time or just the WAN port? 

JonP
Getting noticed

Just the WAN port - WAN1 to be exact. WAN2 is fine.

OVERKILL
Building a reputation

OK, thank you. There are two things that will cause a flap:

1. 100% CPU spikes, but these, in my experience, flap all the active interfaces

2. Energy Efficient Ethernet being enabled on a non-Meraki switch or router. This is a known issue. 

 

What is the make/model of the switch you are using between the ISP and the MX? 

LeeMT
Conversationalist

Do you have Energy Efficient Ethernet (EEE) enabled within your environment? If so, test disabling that and see if the irregular drops stop.

JonP
Getting noticed

I don't believe EEE is enabled - but I can't find how to check?

Get notified when there are additional replies to this discussion.
Welcome to the Meraki Community!
To start contributing, simply sign in with your Cisco account. If you don't yet have a Cisco account, you can sign up.
Labels