I know this may be a very stupid question but can you remove the usb key from the server while the server is running?? And if so can you then access it again by re-inserting it??

 

Or should you shutdown the server whenever you want to add something to the key??

 

 

Thanks,

I know this may be a very stupid question but can you remove the usb key from the server while the server is running?? And if so can you then access it again by re-inserting it??

 

Or should you shutdown the server whenever you want to add something to the key??

 

 

Thanks,

You should NOT unplug the flash drive while the server is running.   However, you do not need to shut down the server to add something to the flash drive.   

 

Why you should not unplug the flash drive:

The superblock for the array is located on the flash drive.  It is read and written to whenever the unRAID array is started or stopped.  If it is not possible to write to the super.dat file to record a clean shut-down, it will be assumed you did not shut down properly and parity will be checked when you restart the array.   

 

The flash drive is also mounted as a file system and holds a number of server specific files created when you administer the array.   The disk configuration, share configuration, and network configuration are also stored on the flash drive.  It will not be possible to make changes to them if the flash drive is removed.

How to access the flash drive while plugged into the unRAID server:

There is an easy solution to updating the flash drive, and it does not involve unplugging it from the server.  When setting up the unRAID "Share" options, you can export the flash drive as a shared drive on the LAN.  It will then be known as \\tower\flash on your network.  You can access it from another PC and write to it, or read from it. (I configured mine to be "exported read/write as hidden."  That way, it does not show up in my media players, but from any PC I can bring up file explorer and type \\tower\flash and get to the drive.

 

If you log in via telnet, the flash drive will be found mounted at /boot.

 

You can type:

cd  /boot

and get to the directory with the flash drive contents.

 

Then, normal linux commands can access it and do whatever you would like to it.   Normally, you will do most management of the array from the web-interface, but there are exceptions.

 

There is no need to remove the flash drive from the unRAID array, even for upgrade operations.  I can't remember the last time I removed mine from the back of the server.  Probably a year or so ago, and I've upgraded many many times through various beta versions.

 

Joe L.

That make perfect sense to me now!!!!

 

I had forgotten all about seeing the "Flash" folder under my //tower folder. Cool, so I can just copy the scripts to that and run them thru telnet.

 

Thanks for answering such a basic question.

 

Have a good day!!!!

Thx Joe for that snippet.  I didn't know that \\tower\flash one.  Works a treat - with a short-cut too.

 

How do you upgrade when the system is live (needs to be live to access the flash...yes?).  Or...are the flash's contents held in RAM when it's running?

Thx Joe for that snippet.  I didn't know that \\tower\flash one.  Works a treat - with a short-cut too.

 

How do you upgrade when the system is live (needs to be live to access the flash...yes?).  Or...are the flash's contents held in RAM when it's running?

The operating system is uncompressed and read into an in-memory file system.   There are some config folder files accessed directly, but those are not touched in an upgrade. 

 

Basically, for most upgrades, you replace two files, bzroot and bzimage on the flash drive, and then reboot.  Instead of removing them, I usually rename them as bzroot433 and bzimage 433 (for this version) and then copy the new versions unzipped from the new distribution in place. That way I can rename them back if there is any need and reboot to get back to where I was.  This has almost never been needed, but then I've tested some of the alpha releases, before they got to be beta....

Ahh...more really good info.  In the past I've always switched my USB to my backup (having brought two initially).  This way I figured I could easily revert to the old one should I face issues.  However, I do worry about pulling the USB drive - seems so fragile.

Added to FAQ (link back only), here.  Feel free to add, edit, or expand.