Noob questions.

[sully]

It's been a bumpy few months with UnRAID.  Finally pulled the trigger to build my first server, I had everything setup and going well, when the old re-purposed MB I was using failed and I decided to get a replacement.  The new MB wouldn't boot UnRAID and I posted a few questions to aid in the trouble shooting (which didn't get any response), but in the end I figured it out.

 

The upgraded system fired to life about two weeks ago and everything has been great, until last night, when the power went out. 

 

The UPS kept the system alive, but I wasn't able to remote access UnRAID to stop the array.  Not knowing how long the power would be out, I figured I should shut the system down, so I invoked shutdown from the command line.  I'm thinking this wasn't the correct procedure, because when I went to start the machine back up this afternoon, none of the Disks show up in the configuration.  It seems to have decided that two of the Disks are my Cache, but it's asking me to repopulate the list of disks in the array.

 

When the power went out, I wasn't transferring data to or from the array, but the cache disks had about 500 gb of data set for the next 'mover' operation.

 

From what I gather from the FAQ/Documentation, this may not be a big deal, but I'm just curious:  What's my next step? 

 

1. Do I just assign the Disk I think is the Parity and the remaining to the array, hit Start and see what happens?

2. Is this a time when I would start the array in Maintenance Mode?

3. If the array disks end up in different positions, how does this affect the shares I've setup and their inclusion/exclusion rules?

 

I'm running 6.1, so just the one parity disk to locate, then assigning the remaining 12 data disks.

 

I've not run across this issue before, as after solving the boot issue, UnRAID figured it all out on it's own, (i.e., populated the disks in the array where they were with the original [failed] MB) and after starting the array and letting the parity check complete, I figured it would be fine moving forward.

 

I'm not sure if I'm using the right search terms, but I didn't find this question in the forums or with a google search.

 

Any help appreciated!

 

Sully.

 

P.S. I know that UnRAID can work with the UPS to shut the system down, but the machine isn't in its "final resting place" and so I haven't fully configured the rig.  Baby steps.  I also haven't got any dockers or any of the other fun stuff going yet.  I'd like to get comfortable / reacquainted with linux/UNIX before I dive in too deep.

[Frank1940]

NO!!!!  DO NOT ASSIGN ANY DISK AS A PARITY DISK UNLESS YOU KNOW IT IS THE PARITY DISK!!!!!!  (Assigning a data disk as a parity disk will cause data loss!)

 

First thing.  Take two minutes and do the setup for the UPS.  (90% of the time, just setting the 'Start APC UPS daemon:' prompt to "Yes" will get things running!  The rest is 'fine tuning'.)

 

Second thing.  Shut the server down again and run chkdsk on the Flash Drive.  (Reason being that something has screwed up the configuration files in the config folder.  I believe that super.dat contains the disk assignment table.) 

 

Third thing.  You assign all of the remaining disks as data disks.  One (and only one) of them will show up as unformatted.  That is the parity disk. 

 

Shares will be uneffected by any changes in disk order.

 

You will have to setup the include/excluded rules again. 

[tdallen]

Once everything is setup again it's worth noting the last 4 digits of each drive's serial number and where it sits within the array.  Notebook, yellow sticky, or the Server Layout plugin - it's information you want to have handy.

[Frank1940]

Once everything is setup again it's worth noting the last 4 digits of each drive's serial number and where it sits within the array.  Notebook, yellow sticky, or the Server Layout plugin - it's information you want to have handy.

 

An even easier way is to print out the 'Main' tab page from your browser!

 

And once you get everything running again, stop the array and backup your Flash Drive by copying its contents to a folder on your PC.  You should do this periodically and always after you add/change a drive in the array! 

[sully]

Thanks for the advice.

 

Ran CHKDSK on the flash and no errors.  The super.dat file is zero bytes for some reason, so that's the likely culprit.

 

I've identified the Parity.  It was the one I thought it was.  Assuming the cache drives are correctly identified (I think they are, b/c they show the 500 GB that Mover hadn't run thru yet).

 

Now I have an invalid configuration and I'm supposing that I just rest the array under New Config?

 

If so, I should be back in business.  Please confirm!  I would suppose the shares are maintained under a New Config as they are a part of the array?

 

 

 

      Found this in relation to the above question on 'New Config'.

 

      https://lime-technology.com/forum/index.php?topic=47504.msg479489#msg479489

 

      Array is back up and running.  I'll be doing a parity check but I still have the below question on command like shutdown.

 

 

BTW, I have labels on the physical drive trays for each disk, with the SN and the logical address... But for the old system!  To be honest, I was just happy that everything was working and hadn't gotten around to doing new ones...  So I'll be using the screenshot/print out to create new labels for the new MB/Configuration.

 

Last question... While I'll add the routers/switches to the UPS side of the power equation (which will allow me to remote in to shutdown the array and power down -- assuming that I don't just let the UPS Settings deal with it)...  What is the proper command line to shutdown a running UnRAID array.

 

For example, I login as Root and I'm at the command prompt?

 

What I did to apparently mess this up was invoke >Shutdown -h -P now

 

What should have been done?

 

Thanks!

 

Sully.

[Frank1940]

Here is the output from shutdown -?  :

 

shutdown: invalid option -- '?'
Usage:    shutdown [-akrhPHfFnc] [-t sec] time [warning message]
                  -a:      use /etc/shutdown.allow
                  -k:      don't really shutdown, only warn.
                  -r:      reboot after shutdown.
                  -h:      halt after shutdown.
                  -P:      halt action is to turn off power.
                  -H:      halt action is to just halt.
                  -f:      do a 'fast' reboot (skip fsck).
                  -F:      Force fsck on reboot.
                  -n:      do not go through "init" but go down real fast.
                  -c:      cancel a running shutdown.
                  -t secs: delay between warning and kill signal.
                  ** the "time" argument is mandatory! (try "now") **

 

As I recall, a straight shutdown should work.

 

 

 

[sully]

 

As I recall, a straight shutdown should work.

 

 

 

 

I think it needs the -t time (for which I picked 'now') at minimum...  I recall typing shutdown and just got the same as if I'd typed Shutdown -?...

 

I'm just wondering if I did something else wrong.  With the UnRAID array started, do I need to do something more than just 'shutdown -t time -h -P' to ensure that the array is stopped properly, so I can hit the power switch?

 

As I said, this would be for an emergency more than anything, I'll setup the network infrastructure to run from the powered ports on the nearest UPS so I can at least login and shutdown (assuming the power goes out when I'm home/awake) and I'll let the UnRAID UPS interface deal with it otherwise... (You were right about UnRAID just 'finding' the UPS, I was worried b/c it's a CyberPower rather than an APC )

 

Anyway, thanks for all the help!

 

Sully.