So far unRaid has been a fun project. But my wife has also started to get used to having the "media library" accessible to her as well. When the unRaid server goes down, my life gets more difficult. Especially when I want to justify blowing more money on computer shit (which I love)!

 

My goal is to make this unRaid box survive for two months without any interaction from root.

 

Currently, I suspectect I'm missim an essential piece of hardware - a ups. Now my house doesn't have rolling brownouts or anything but we do get our fair share of electrical issues from time to time. I suspect this is a source of some of the problems I've been experiencing so far.

 

 

1. Back up UPS

This also needs a plugin to perform a clean power down if the power is out for XX seconds.

 

2. Plugin stability

When I go to unmount the array so the server can be rebooted, invariably, some plugin hangs up the whole process and I have to kill-9 it from the terminal. Those plugins either need to be removed (    ) or made to work properly.

 

3. Auto-on

Is this possible? Maybe is a wake-on-lan thing or bios setting. When the power comes back on, I meed the server to come back up, start all services and resume normal operations.

 

This winter I will start chipping away at the problems to ensure the server can survive the whole two months I away. I will update this post as I move along and mark it as solved when I believe I've addressed all the issues.

 

 

 

Sent from my SAMSUNG-SGH-I727 using Tapatalk 2

 

 

I used to have plugins hang up the shutdown process, but since I have stepped over to influencers version of these plugins I have not experienced these, his versions seem very clean to that effect..

There is a clean powerdown you can install that will shut your array and power down and it can be used in conjunction with the apcupsd plugin.  I bought a APC SmartUPS which has a serial cable that connects to your unraid.  When the ups detects it is on battery, it tells the server which can then shut itself down after a certain amount of time (I think thats how it works).  This link has some info for you.  I have taken it a step further and am using my unraid server to alert and shut down a windows box in the event it is on battery.

 

http://lime-technology.com/forum/index.php?topic=19418.msg178503#msg178503

 

Not sure about an auto power on.  That might be cool, but have done no research into it as of yet.

Auto-start on AC power application for most motherboards is normally a simple BIOS option in the power settings (nothing to do with networks or wake-on-lan).

Good to know.  I'll have to take a look.  Although..How would that work i wonder.  As far as the hardware is concerned, it still has power since its being supplied from the UPS.

 

And i have to correct myself.  Its not a serial cable (thats old school).  Its a USB cable that connects my UPS to my server.

on my MB i have it set to powerup upon power - the ups tells the server to shutdown, it does a shutdown/halt i think  - then X min later the ups shuts down.  When it gets power again, it supplies power to the MB and then it starts back up as well.

 

Has been perfect the last 3 times I have been home when power has gone out......

 

Myk

 

I'm glad for where I live, only time I lose power is during big storms... Think around 3 inches of ice before I even have to worry.

Thanks for the info MyKro!

Good to know.  I'll have to take a look.  Although..How would that work i wonder.  As far as the hardware is concerned, it still has power since its being supplied from the UPS.

It is one reason why the option to shut down the UPS soon after the shutdown of unRAID exists on the plugin.  That way, when commercial power returns the auto-power-on option in the BIOS will detect the power being re-applied and boot the server.

And i have to correct myself.  Its not a serial cable (thats old school).  Its a USB cable that connects my UPS to my server.

Either can work with "apsupsd".  Most older UPS were serial, newer ones are USB.

Transmission is mysteriously running.

root@Dumbo:/boot/packages# ps -ef | grep trans
nobody   25562     1  3 17:37 ?        00:00:21 /usr/bin/transmission-daemon -g /mnt/cache/.transmission/conf --pid-file /var/run/transmission/transmission-daemon.pid
root     28164 25964  0 17:47 pts/0    00:00:00 grep trans

 

There is no transmission nothing anywhere in /boot/configure/plugins

There is a bunch of transmission stuff in /boot/packages

 

root@Dumbo:/boot/packages# ls -l | grep trans
-rwxrwxrwx  1 root root   957704 2011-11-04 05:42 transmission-2.42-i486-1sl.txz*
-rwxrwxrwx  1 root root   581624 2012-06-27 13:54 transmission-2.52p-i486-1_W-W.txz*
-rwxrwxrwx  1 root root   618664 2012-09-11 17:52 transmission-2.61-i486-1_PTr.txz*
-rwxrwxrwx  1 root root    33383 2012-04-15 08:27 transmission-unmenu-package.conf*
-rwxrwxrwx  1 root root    33367 2011-11-20 20:30 transmission-unmenu-package.conf-2011-11-20-203030.bak*
-rwxrwxrwx  1 root root    33368 2011-11-20 20:44 transmission-unmenu-package.conf-2011-11-20-204400.bak*
-rwxrwxrwx  1 root root    33376 2011-12-05 19:37 transmission-unmenu-package.conf-2011-12-05-193703.bak*
-rwxrwxrwx  1 root root    33376 2011-12-17 19:58 transmission-unmenu-package.conf-2011-12-17-195837.bak*
-rwxrwxrwx  1 root root    33376 2011-12-20 18:03 transmission-unmenu-package.conf-2011-12-20-180341.bak*
-rwxrwxrwx  1 root root    33376 2011-12-20 20:05 transmission-unmenu-package.conf-2011-12-20-200506.bak*
-rwxrwxrwx  1 root root    33378 2011-12-20 23:14 transmission-unmenu-package.conf-2011-12-20-231434.bak*
-rwxrwxrwx  1 root root    33383 2012-01-14 16:15 transmission-unmenu-package.conf-2012-01-14-161540.bak*
-rwxrwxrwx  1 root root    33383 2012-03-10 19:28 transmission-unmenu-package.conf-2012-03-10-192857.bak*
-rwxrwxrwx  1 root root    33383 2012-03-30 15:59 transmission-unmenu-package.conf-2012-03-30-155951.bak*
-rwxrwxrwx  1 root root    33383 2012-04-15 08:27 transmission-unmenu-package.conf-2012-04-15-082732.bak*
-rwxrwxrwx  1 root root   975096 2011-10-05 19:41 transmission_unraid-2.33-i486-11kh.txz*
-rwxrwxrwx  1 root root    18598 2012-04-15 08:34 transmission_unraid-2.33-i486-11kh.txz.auto_install*
-rwxrwxrwx  1 root root    18598 2012-04-15 08:27 transmission_unraid-2.33-i486-11kh.txz.manual_install*

 

I am not running the old unmenu script any longer out of my go script.

Whole /boot/config/go script

#!/bin/bash
# Start the Management Utility
/usr/local/sbin/emhttp &



# wait for all disks to be mounted, or until timed out...
max_wait=300
total_wait=0
sleep_delay=5
num_configured=`grep "disk[0-9]*=" /boot/config/disk.cfg | wc -l`
while [ : ]
do
md_disk_mounted=`mount | grep "/dev/md[0-9]*" | wc -l`
[ "$num_configured" = "$md_disk_mounted" ] && break
  [ $total_wait -gt $max_wait ] && break
sleep $sleep_delay
(( total_wait += $sleep_delay ))
done

# Run Plex server at boot
echo "/boot/custom/plex/init.sh" | at now

# Run unMenu at boot
# /boot/unmenu/uu

cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c
/usr/local/sbin/cache_dirs -w -B -u -m1 -M10 -d9999

 

Transmission is one of the things that's been holding up my shutdowns of the server. Should I just delete all the transmission* files from /boot/packages ?

I have a hunch this is also connected to the go script.

/boot/config/go script:

#!/bin/bash
# Start the Management Utility
/usr/local/sbin/emhttp &



# wait for all disks to be mounted, or until timed out...
max_wait=300
total_wait=0
sleep_delay=5
num_configured=`grep "disk[0-9]*=" /boot/config/disk.cfg | wc -l`
while [ : ]
do
md_disk_mounted=`mount | grep "/dev/md[0-9]*" | wc -l`
[ "$num_configured" = "$md_disk_mounted" ] && break
  [ $total_wait -gt $max_wait ] && break
sleep $sleep_delay
(( total_wait += $sleep_delay ))
done

# Run Plex server at boot
echo "/boot/custom/plex/init.sh" | at now

# Run unMenu at boot
# /boot/unmenu/uu

cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c
/usr/local/sbin/cache_dirs -w -B -u -m1 -M10 -d9999

 

/boot/custom/plex/init.sh:

#!/bin/sh

# Start Plex using runas user from configuration
su root -s /bin/sh -c "/boot/custom/plex/plex.sh &> /dev/null &"

 

In the latest iteration of the Plex server from here http://forums.plexapp.com/index.php/topic/42513-0963-for-unraid/, no where does it say anything about modifying the go script. Perhaps the file /boot/custom/plex/init.sh shouldn't be there? along with the plex line in the go script. Should I remove all those things?

 

as the server is normally running along:

root@Dumbo:/boot/custom/plex# ps -ef | grep plex
999      23775 23741  0 17:30 ?        00:00:10 Plex Plug-in [com.plexapp.system] /mnt/disk1/plexapp/Application Support/Plex Media Server/Plug-ins/Framework.bundle/Contents/Resources/Versions/2/Python/bootstrap.py --server-version 0.9.7.7.339-5ec3b53 /mnt/disk1/plexapp/Application Support/Plex Media Server/Plug-ins/System.bundle
999      23913 23741  0 17:31 ?        00:00:00 /usr/local/plexmediaserver/Plex DLNA Server
999      24096 23741  0 17:31 ?        00:00:01 Plex Plug-in [com.plexapp.plugins.spotify] /mnt/disk1/plexapp/Application Support/Plex Media Server/Plug-ins/Framework.bundle/Contents/Resources/Versions/2/Python/bootstrap.py --server-version 0.9.7.7.339-5ec3b53 /mnt/disk1/plexapp/Application Support/Plex Media Server/Plug-ins/Spotify.bundle
root     26813 25688  4 17:41 ?        00:00:35 find /mnt/disk1/plexapp -noleaf
root     30172 29154  0 17:55 pts/0    00:00:00 grep plex

 

After quitting Plex through the unRaid GUI:

root@Dumbo:/boot/custom/plex# ps -ef | grep plex
root     26813 25688  4 17:41 ?        00:00:38 find /mnt/disk1/plexapp -noleaf
root     30769 29154  0 17:57 pts/0    00:00:00 grep plex

 

There is still that lingering find command. Not sure why it's there, but it keeps the array from spinning down while it's running there insistently. What gives!?

PLEX PROBLEM

As I outlined above, there was a lingering process even after quitting Plex through the unRaid GUI. This was caused by a single line of code that was left over from an old Plex install in my /boot/config/go script. After commenting out that line, the lingering process did not run after a system reboot, and Plex was out of the way.

# Run Plex server at boot
# echo "/boot/custom/plex/init.sh" | at now

 

TRANSMISSION PROBLEM

Also a lingering problem there. I updated to an "unplugged Influencer" version of this program. I had many lingering files in /boot/packages from old Transmission installs. I think this was because unMenu, the "old" way of using Transmission, and put them there. I've long since stopped running the umenu (commented it out of the go script), but the lingering packages were still being executed due to this line in the go script:

cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c

I removed all the transmission* packages from the /boot/packages directory

root@Tower:/boot/packages# rm tranmission*

and the problem did not persist after a reboot.

 

POWER PROBLEM

I do live a little far out of town. The fastest DSL I can get is 300KB/sec down. Cable is not offered in my area. There are occasional power surges and drops. From time to time, I would get home, and TOWER would be running, but it would be all weird. It's like the power stopped for just a moment. Enough to knock out some parts of the server, but not others. Very strange. And it didn't simply restart - which would've been fine. I bought a UPS from Newegg, it arrived, I plugged that in, along with the USB<>ethernet cable that came along with it, installed the plugin "APC UPS Daemon," and all seems to be well on that front as well.

 

The test I kept using was to go to the unRaid GUI, I would hit Power Down. So long as the array goes offline without any further complaints, I trust that APC UPS Daemon will also be able to shut down the server if the power goes out for any appreciable amount of time.

 

I think this thread is now closed. Thanks so much for all your help Joe L., althoralthor, Influencer, MyKroFt, S80_UK, and Helmonder. You guys are what make unRaid work.