PowerConnect 3448P crashes/reboots when using web interface

Question

I have 3 PowerConnect 3448P's stacked. They seem to work fine but whenever I try to use the web interface I have problems. Sometimes just the act of trying to go to the page will cause problems, sometimes it'll lock up as soon as I enter my login info. Or most of the time it locks up whenever I try to do stuff such as check the health, monitor PoE status, etc. In the Log File I get errors like this: '%SYSLOG-F-OSFATAL: EXCEPTION OCCURRED in task MUXT: PROGRAM exception ***** FATAL ERROR ***** Instruction 0x0 Accessed data 0x0 Stack pointer 0x1493838 Machine state register 0x8b030 Condition register 0x20000048 General purpose R0 0x0 General purpose R1 0x1493838 General purpose R2 0x0 General purpose R3 0x595c398 General purpose R4 0x0 General purpose R5 0x1 Last calls stack 00000000 002b4450 000a9cf8 ***** END OF FATAL ERROR *****' When it has problems sometimes the master reboots and then the slave takes over, sometimes it just hangs where all 3 switches work fine but the web interface is useless until I reset the switch. I did some searching about that error and saw a previous thread where it was said that this is due to mismatched firmware. I checked all 3 switches and they all have 1.0.0.13 (dated Feb of 2006.. is this a beta/alpha firmware because the 1.0.0.13 on the site says its for april 2006?). I've tried switching the stack roles so unit 1 (normally master) is slave and unit 2 is master, the same thing happens. I also run 4 stacked 3448's (non PoE) and I'm not having any problems, it's just these 3 PoE models for my phone LAN that's having the problem.

rdefusco · Answer

Just some more info...

I tried upgrading the firmware just to make sure (the files are the same so i guess even though it was released in april its dated feb?)

I tried switching out one at a time with a spare unit, same results.

One thing I did notice though, if the backup unit is missing it works fine. i.e. if I have Unit 1 (master), Unit 3 (slave), and even Unit 4 (spare slave) the problem goes away, but as soon as I connect a Unit 2 (backup), I get crashes/reboots when using the interface.

Also it seems to crash the CLI.

robinsjt · Answer

I have noticed the same issue although ours are not stacked. When you log into the web interface and don't log out; at the top right hand corner of the interface, then it starts. The only thing to do is either reboot the switch or use the CLI. If this happens and you enter the CLI and do a show user you'll see that there are two admins logged in. From what I understand the next firmware should allow an admin to remove this hanging cached admin user from the CLI.

Networking General

Was this post helpful?