I've placed this in the kernel forum because I believe my error message is presented by the kernel.

Has anyone ever seen anything like this before or any idea what it could mean?

For a short period one of my RedHat 6.2 (2.2 kernel) machines would report "Power failure" after typing cli commands. The box had stopped doing this when I came back again the next day.

I was connecting remotely via ssh;


After a netstat -anc:

............ detail removed.
Active UNIX domain sockets (servers and established)
Proto RefCnt Flags Type State I-Node Path
unix 3 [ ] DGRAM 357 /dev/log
unix 1 [ W ] STREAM CONNECTED 427694
unix 1 [ ] STREAM CONNECTED 427693
unix 0 [ ] DGRAM 522
unix 0 [ ] DGRAM 424
unix 0 [ ] DGRAM 370
Power failure
[xxxxx@xxxx06 folder]$

Also after a 'top' command:

[xxxxx@xxxx06 folder]$ top
Power failure
[xxxxx@xxxx06 folder]$

Is your system connected to a UPS?

Do you have a UPS monitoring daemon running?

Are there any power-related messages in the logs?

No, no and no I'm afraid.

Any hit of power loss would have sent the system offline.

The system is in a standard atx type case with old hardware, the only thing different from a desktop PC is a couple of eicon x25 pci cards.

Well, something on your system believes that a power failure of some sort has occurred. Are you running lm_sensors?

Have you checked /var/log/messages and /var/log/syslog? If this is a kernel message, it should be visible in the output from dmesg.

There seems to a few of these errors in the messages log from around the time of the problem.

kernel: EtCM1: Dropped NCB from completed request list pid=0 cmd=0x92 retcode=0x0
kernel: EtCM0: Dropped NCB from complete request list pid=0 cmd=0x91 retcode=0x0

There's very little about this on Google. Is NCB "network control block" ?