June 20, 201016 yr I am in Brazil and my home lost power for nearly 24 hours yesterday and today due to storms. When power came back I started getting hourly Unraid Re-Sync emails until it reached 99.8% (with one minute left). About an hour before that I started also getting emails I have never seen before with contents like this: /bin/sh: line 1: 6084 Killed /usr/bin/run-parts /etc/cron.hourly >/dev/null I use OpenVPN and logged into the server using Telnet, and can get in ok. But the system seems to keep killing processes I try to run. In fact none of the web services like emhttp are running and when I try to run them manually they seem to be killed off automatically. Anyone know what's happening? Is there a command-line command to do a reboot (not just a safe powerdown) as no one is around to restart the server. Thanks for any insight anyone can provide. G P.S. And yes, there is a good reason I am worried about unraid when I am in Brazil, there are files I need on Monday for a job I am doing down here for work.
June 20, 201016 yr I am in Brazil and my home lost power for nearly 24 hours yesterday and today due to storms. When power came back I started getting hourly Unraid Re-Sync emails until it reached 99.8% (with one minute left). About an hour before that I started also getting emails I have never seen before with contents like this: /bin/sh: line 1: 6084 Killed /usr/bin/run-parts /etc/cron.hourly >/dev/null I use OpenVPN and logged into the server using Telnet, and can get in ok. But the system seems to keep killing processes I try to run. In fact none of the web services like emhttp are running and when I try to run them manually they seem to be killed off automatically. Anyone know what's happening? Is there a command-line command to do a reboot (not just a safe powerdown) as no one is around to restart the server. Thanks for any insight anyone can provide. G P.S. And yes, there is a good reason I am worried about unraid when I am in Brazil, there are files I need on Monday for a job I am doing down here for work. Reboot once more, you've basically run out of RAM. The out-of-memory kernel process is killing off processes trying to free enough to run. First you want to stop the server. (if you can) Then, you can issue the reboot command. To stop the server, if you can log on, try: killall smbd killall nmbd umount /dev/md1 umount /dev/md2 umount /dev/md3 up to umount /dev/md19 If a drive fails to un-mount type fuser -k /dev/mdX where mdX = the drive that will not un-mount Once you get all the disks un-mounted, type: /root/mdcmd stop finally type reboot Joe L.
June 20, 201016 yr Author Joe to the rescue again! Thanks man. So based on your assessment I went in and tried to kill a few processes to free up memory and got emhttp and uu started. With emhttp it would not unmount the cache drive (I assumed due to the swapfile), I started unmenu and disabled swap and it stopped unraid properly. Then I chose to reboot and it started up perfectly! Thanks again!
June 20, 201016 yr Author Yeah, I'll have to see. I have 4GB in there... I do use sickbeard and sabnzbd and when it started up I see that I had a huge backlog it was processing. I'll have to watch it more closely and see if it could have cause this. I'll look at the logs and see. I also need to know if the server shutdown properly. I do use APCUPS package (sorry don't remember spelling) with my APC UPS. I have tested it and it has worked, don't know if it did this time.
June 20, 201016 yr I think the later revs of unRAID are more sensitive to third party programs and heavy memory utilization. Even with swap space I ran out of memory. Swap was totally unused. I had to turn off vm.cache_pressure. I kept running out of memory just dong a find down the filesystem (I have 15 disks and tons of small files).
June 20, 201016 yr Author What's the max addressable storage in Unraid? Looked it up and it says after 4.4 you can address more than 4GB. Maybe I should upgrade, but I do have 4GB now... This is the only time I can recall this happening... EDIT: I guess 4GB is the max the C2SEE board supports...
Archived
This topic is now archived and is closed to further replies.