cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1560
Views
0
Helpful
2
Replies

Found 4948-10G in rommon

rtannertwc
Level 1
Level 1

Today one of the 4948s I am responsible for lost communication.  When I arrived on-site and consoled in, I found the switch in rommon mode.  After checking the bootflash, I power cycled and it booted back up normally and is now running.  Every time I have seen a switch go into rommon it was either put there intentionally or the bootflash failed on powerup.  Is there any other reason that it would have gone into rommon once and recovered with a single power-cycle?  It is dual powered with each power supply on seperate building wide UPSs so power failure isn't likely.  register setting is x2102 and the boot system commands match the filename in bootflash.  Wouldn't normally have problem with equipment that is now running but it seemed weird and I can't decide on a cause of failure

1 Accepted Solution

Accepted Solutions

Shashank Singh
Cisco Employee
Cisco Employee

Hi Ryan,

As you already mentioned. a switch may fall into ROMMON either because of the config-register setting (which is correct in your case), or because the switch is unable to load the image from the location specified by the boot variable.

In most of the cases, this happens because of flash memory corruption (especially when the switch has remained up for years). Hence, it is safer to have more than one location for the image (say bootdisk: and disk0:) and specify both of them in boot variable.

In rare cases or when the problem happens repeatedly, there can be a software bug or a hardware fault involved with the switch.

Hope this helps,

Shashank

Please rate if you found the content helpful

View solution in original post

2 Replies 2

Shashank Singh
Cisco Employee
Cisco Employee

Hi Ryan,

As you already mentioned. a switch may fall into ROMMON either because of the config-register setting (which is correct in your case), or because the switch is unable to load the image from the location specified by the boot variable.

In most of the cases, this happens because of flash memory corruption (especially when the switch has remained up for years). Hence, it is safer to have more than one location for the image (say bootdisk: and disk0:) and specify both of them in boot variable.

In rare cases or when the problem happens repeatedly, there can be a software bug or a hardware fault involved with the switch.

Hope this helps,

Shashank

Please rate if you found the content helpful

I can accept the flash corruption explination.  The other 4948s that are doing the same job in other buildings have uptimes of almost two years.  I'll keep an eye on this switch and hopefully it will run for another two years.  Might get these terminal server connections to avoid the drive though.  Thanks for your insight.

Ryan

Review Cisco Networking products for a $25 gift card