cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Announcements

CISCO SWITCHES FOR SMALL and MEDIUM BUSINESS

Introducing the next generation of Cisco Small and Medium Business Switches. Cisco is refreshing its SMB Switch portfolio. Click here  to learn more.


1463
Views
0
Helpful
11
Replies
ddavis002
Beginner

SG500-52p disappears

We have a SG500-52p Managed/Stackable switch that will just disappear from the network.  Can't connect new computers and the web interface can't be found and it can't be pinged.  After a restart all is well again.  Can't find any issues in the log.  Any ideas?

~Don

11 REPLIES 11
Glenn Martin
Cisco Employee

moved post to Small Business > Switches community

mdobiac
Participant

Hello ddavis002,

I am sorry you are experiencing this issue.

First check to make sure that the switch this is losing communication that on its stack port that the speed is manually set to either 1G/5G/10G, whichever you are using.  

After that I make sure that all the switches in the stack are on the same firmware 1.4.1.3 and its associated boot code 1.4.0.02.

Hope this helps,

 

Michael D.

If this post is helpful please rate or mark as correct.

Mark Findlay
Beginner

We are also getting this exact problem.

 

Same model of switch. Firmware version 1.4.1.3. Boot version : 1.4.0.02

Existing connected clients stay connected, new ones don't connect. Unable to ping/telnet/ssh to switch. Unable to access web interface

To fix we need to restart. This resolves problems until it happens again. Nothing in log afterwards.

 

We don't stack the switches so why do we need to change the stacking speed?

Our fix was to just upgrade to the firmware and boot version you stated.  Ours has been fine since.

Oh..

We are getting the issue even on this new firmware. Doesn't look like there is a newer one available either.

 

We didn't get the problem on version 1.3.x..

 

You didn't change the stack port speed did you?

 

Nope - did not change port speed.  We're not stacking either...

Hi,

Did anyone manage to get packet capture on PC when not able to ping switch?

Also was the switch accessed via console to check MAC table, IP interfaces, VLAN membership, port counters?

Looking forward to some more information.

Aleksandra

Hi Aleksandra

 

We can't access the switch via a console cable either. Just to make sure we weren't doing something stupid we plugged the console cable into the switch below (without removing it from the laptop) and it worked fine.

 

It looks like the control plane has crashed? It's still forwarding traffic for existing clients.. I bet it can't add additional entries to the mac address table.

 

I'm not sure how we can do any more debugging. I'd say it was a hardware problem but it didn't happen on any of the switches on the 1.3 firmware. It now happens to multiple switches, all on 1.4..

 

We're doing a packet capture now.

Hi Mark,

Thank you.

If there is an issue with MAC table we should be able to see in packet capture some unicast which is not meant for our PC, packets should be simply copied on all interfaces in the same broadcast domain.

RAM logs are erased after reboot, flash logs are kept however default settings for flash are limited. I wonder if we can be successful with external syslog server.

Also is there any chance to downgrade any unit to 1.3 to confirm the thesis?

Regards,

Aleksandra

 

DDavis

 

Are/were you running etherchannel from those SG500's? That's the only other thing we have implemented as well as the firmware upgrade.

 

Aleksandra - I don't think it produces any logs. We did try with a sys log server but I don't think we ever captured anything. We haven't had a chance to downgrade either but when we do we will let you know what happens.

 

Mark

rmalson
Beginner

We are experiencing the exact same issues here.  We run about 30 SG500-52P switches throughout the organization, and since we upgraded the firmware to 1.4.0.02/1.4.1.3 we have had several of the switches become "unmanageable"  resetting (power cycling) the switch will bring it back, but after several weeks of normal operation the same issue occurs.  The switches are running in L2 mode and are not stacked.  The issue is the same as above, cannot get to management interface/cli (http/https/ssh) and cannot ping management vlan ip.  The switch will still pass traffic to/from connected hosts, but a newly connected host will not communicate.  The issue did not happen on older firmware versions, prior to 1.4 releases.