cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3136
Views
10
Helpful
5
Replies

3850 stack crash running Denali 16.3.6

N3t W0rK3r
Level 3
Level 3

We have a 5 switch stack of all WS-C3850-48P switches running Denali IOS-XE 16.3.6.

Yesterday, while working on a minor config change via SSH, my console session locked up on me.

I had just created a new vlan and gave it a name and then poof! It hung.

I opened a new ssh session with no problem and was able to get into the switch.  I ran a show logg without issue, along with a show proc cpu but when I issue a show run, it locked again!

I had a colleague look at it.  He logged in ok and saw my multiple sessions still active.  He could not clear them.  And then the session locked on him as well.

At this point the stack was still forwarding traffic to clients so no big deal, yet.

I went into the closet an consoled in to the top switch and ran a show log again and then a show vlan, at which point it hung again.

Then it seemed the switch stack started dropping clients.

I had no choice but to power cycle the stack.  I booted back up fine 10 minutes later.

When I reviewed the syslogs, I saw an odd entry from around the same time, that I have not been able to find any information on (see below).

 

2018-05-03 13:38:58        Local7.Warning switchb    7326: May  3 13:38:57.397 EDT: %EM-4-AGED: The specified EM client (EM_TYPE_VTP_DOWNLOAD_INFO type=18, id=22131)

2018-05-03 13:38:58        Local7.Warning switchb    7327:  did not close the EM event within the permitted amount of time (480000 msec).

 

Any ideas what happen and why?  Should we abandon this code?  I could not find any reported bugs on this topic.

 

Thanks, in advance.

 

John

 

1 Accepted Solution

Accepted Solutions

Reza Sharifi
Hall of Fame
Hall of Fame

https://bst.cloudapps.cisco.com/bugsearch/bug/CSCuq38216

 

There is a bug associated with EEM but not the same exact type you are seeing.

I would open a ticket with TAC as this might be a known bug to them.

HTH

 

View solution in original post

5 Replies 5

Reza Sharifi
Hall of Fame
Hall of Fame

https://bst.cloudapps.cisco.com/bugsearch/bug/CSCuq38216

 

There is a bug associated with EEM but not the same exact type you are seeing.

I would open a ticket with TAC as this might be a known bug to them.

HTH

 

b-brugger
Level 1
Level 1

Since we have upgraded to 16.3.6 on our C3850-48P Stack and did a "show vlan" (before we deleted a vlan). SSH sessions locked.

Now we get " %SNMP-3-INPUT_QFULL_ERR: Packet dropped due to input " and %EM-4-AGED: The specified EM client (EM_TYPE_VTP_DOWNLOAD_INFO type=18, id=217)

 

Switch is ping reachable, but not snmp reachable any more.

 

How to configure workaround "Remove IPv6 Source guard" ? We have some automatic config on switchports from APIC "ip device tracking maximum 10" which i read, causes implied ACLs.

Ok, found another workaround:

Telnet Sessions Hang/Become unavailable at execution of "show run"
CSCvj49476

cclan
Level 1
Level 1

Hello, this is marked as solved but when it happenned to us (after creating a new vlan as well) the IPv6 source guard was already disabled. How did you managed to solve it?

Thank you

In our case it was a combination of global enabled "ip dhcp snooping" and "IPv6 source guard" on ports. After disabling dhcp snooping the switch did not Hang/Become unavailable at execution of "show run" any more.

 

We tried to upgrade to 16.6.4, but got another bug with snooping https://bst.cloudapps.cisco.com/bugsearch/bug/CSCvk16813

 

Maybe next try we upgrade to 16.9.x and hope no bug affect on our configuration

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: