V3 broke after power outage...

[chrisparker2000]

So I've been running pretty good for 2-3 months now, when the power went out. During the outage I removed the USB drives so when the power came on it would boot off my stick.

When the power came on, the pogo booted, but no solid green light. So I killed the power and stuck the stick in my linux box, and ran fsck on it. It found a few problems, fixed it, then I unmounted it an stuck it back in the pogo and booted. No solid green light again. The ethernet link and traffic lights are active.

The Installation page for the V3 says this..
If your Pogoplug boots and the front LED blinks wildly for about 15-20 seconds and then TURNS OFF, you're in business when it comes to booting, but there is likely something that has stopped the network. It's not a brick, but be ready to troubleshoot.

This seems to describe my situation, but I've never had to troubleshoot networking not coming up on a box that the only way I could get to it is via networking. All I know is that I can't ping it, and of course I can't ssh into it.

Any ideas? TIA

[pepedog]

Has the IP address changed?

[chrisparker2000]

I don't know what it was, but I am using the address associated with its MAC address according to the DHCP server. I don't know when it got that address...

[WarheadsSE]

My suggestion would be to put the drive into another linux computer,
- edit /etc/rc.local to execute halt
- place the usb drive back in the plug and let it start up/shutdown
- re-attach to pc, and review logs/dmesg.

[chrisparker2000]

What information I have gleaned from the logs:

-These lines keep filling up my logs:
localhost init: open(/dev/console): No such file or directory
localhost udevd[361]: unable to receive ctrl connection: Function not implemented

-Networking does start, get a DHCP address, and successfully sets the clock based on an ntp request

- minilogd invokes oom killer, which killed minilogd. Perhaps I am reading this one wrong, but I thought that init did this when I looked at this before.

I've seen posts for those lines filling up logs, but they seem to be after a pacman update, which did not happen.

My /dev/console permissions are crw-------. I found somewhere that someone said it should be crw--w--w-. I tried chmoding it and that didn't work, which is probably not surprising. Is this the culprit?

Thanks

[WarheadsSE]

Bingo.
initscripts & you didn't replace udev with udev-oxnas before rebooting.

[chrisparker2000]

But I had that problem months ago. I backed out initscripts then added IgnorePkg=initscripts to pacman.conf and I did replace udev with udev-oxnas. And I haven't ran pacman -Syu in months. Well not since you helped me get rsync working a few months ago. And when I had that problem I could at least ssh into the box. Now I cannot.

[WarheadsSE]

At this point, I would consider copying config files and re-working the rootfs.