Cisco 3662 w/ (1) T-1 WIC, multiple FE and E interfaces.
Continuous uptime of 1.5 years until this morning around 8:30 AM. The “System” LED on the front and back of the chassis was blinking, indicating that the router was in ROMMON mode. I did not connect via serial to confirm this.
A power-cycle of the router brought it right back up.
My question is, what causes a router to be in working mode, and suddenly switch to ROMMON? Can a command be issued from the console to do that? The config regsiter is 2102, so if it rebooted itself it should have attempted to load the saved/start-up config. This unit is connected to a very large UPS. I look at other PC’s on the same UPS and they were fine. Could a burp (switch to UPS from Main AC) cause the router to reboot, and for some reason not load its startup-config?
Thanks.
A break signal over the serial port will cause a cisco router to drop into rommon, but usually only during the boot-up sequence. Was anything left plugged into the console port?
If not, just keep an eye on it. If it happens again, just replace it.
You might just have experienced an IOS bug of some sort. There are tons of them all over the place in Cisco land. If you aren’t running a somewhat recent image, then you might want to consider upgrading.
Nope, nothing left unplugged. It is very strange.
I was trying to determine if this could be caused by unauthorized access. From what I have been told, there is no command that could be issued that would cause the router to switch to rommon.
I have seen routers that require (2) power cycles to load their startup-configs. My upstream provider was able to tell me what time the remote alarm occured, which helped.
Is there any type of log that I can view that may give me a little more information?
Thanks
if you give the command ‘sh ver’, Cisco routers gives you, immediately after the up time, the cause of the last reboot/restart.
Can you check what it says and if the time of the last reboot is the time of the problem you had?
It tells me that the “system was returned to ROM by power-on”. This is when I power cycled the unit.
Do you have a UPS on the ROUTER?
This happens if you have a power problem.
Yes I do have a UPS. I was told yesterday that it was “humming” about 2 hours after my upstream provider reported the remote T1 alarm.
So, AC was lost and it switched to batteries. Whether or not the batteries are good will have to be determined when I have time to test it.
Have you seen my specific problem happen before?
Yes,
I will not happen with Cisco Telecom products, 48V.
I happens from time to time if you have AC problem.
When was the last time you replaced the Batteries on your UPS.
It’s best to change them every 18 month.
A lot longer than 18 months…
Thanks for your help. I’m glad I am not the only one to experience this issue. I will keep your advice in mind.
Thanks Again,
Matt
This is the reason we have “syslog” enabled on our routers and have them syslog to a Linux box. We also have a “dumb” terminal plugged in so that when things happen in the router we can go back and review between the syslog and the monitor and get a better idea of what happened. That aside, you might also try “sh log” and see if there are any logging entries on the Cisco. If not routed to a syslog server they are scrolled (oldest entry dropped as each new entry is logged) so the longer the router runs, the less chance anything from the reboot is there.