Hey guys,

 

A few weeks ago I had a drive die (Seagate 1tb) in my unraid box and I replaced the drive with a newer model seagate 1tb. I was incredibly busy and I should have gotten onto this sooner, but I have had exams.

 

After replacing the drive, the array rebuilt the data onto it, and it is about 75% filled. But it has a red ball beside it. I never ran a preclear on the drive. I unmounted the drive from the array and telnet'd into the box and while running preclear_disl.sh I got a message the the drive will NOT be cleared, but no explanation why not.

 

I figured that it was because the drive has been part of the array and had data on it.

 

I have attached my syslog. If anyone has any ideas let me know. However, I might transfer some data off of a drive in my desktop and replace the drive in question and let that preclear over night and see how things are in the morning.

 

I will report back.

syslog.txt

if I press "Y" I get a warning

 

You need to type the word "Yes". 

 

Take notice to the capital "Y"... small "es".

If that is the case, that will be a huge face palm on my account.

 

I am trying now...

 

and that is it. I simply didn't read the message right. I am tired at the moment. Sorry for my stupidity.

 

Thanks mikejp

I started the preclear after sorting out my stupidity.

 

About 15 minutes into the clear, putty closed down by itself, the Server drives spun down a moment later. I cannot visit the web interface, I can't open any shares and I can't map the flash share either. I had speedfan running on my PC and all temps were high, but normal for a hot night. The temp of the drive being cleared last time I checked was about 35 degrees, and it is a hot night here.

 

I read that the telnet client needs t be connected the whole time to the server while preclear takes place. How bad is it do stop the preclear half way through?

From the Unraid config tututal:

Preclearing With Screen

This is the preferred method of using the preclear script. It takes a long time! (see preclear times for different drives here). The problem without using "Screen" is when you have to telnet to your server. That connection must be uninterrupted for a full 24+ hours, or it will fail - don't pass go, do not collect $200. Using "Screen" will enable you to log into your box, start up the preclear script, then log out. Set it, and forget it.

http://lime-technology.com/wiki/index.php/Configuration_Tutorial#Preclearing_With_Screen

 

 

The first time I tried to preclear a 3tb disk, the PC I was using locked up and I lost my telnet session. No harms was done, other than to have to start all over. If your preclear was halted 15 mins in, the disk was only being read, not changes had been made.

Since you have had a problem, I would suggest that you plug a monitor and keyboard into the server.  You will not be dependent on maintaining that link via Putty to the server during the preclear and, if it should crash, you will be more likely to to able see the error message.

Forgive my Linux noobness but wouldn't something like "(echo Yes | nohup preclear_disk.sh /dev/sdX) &" work? And just check nohup.out later on?

Forgive my Linux noobness but wouldn't something like "(echo Yes | nohup preclear_disk.sh /dev/sdX) &" work? And just check nohup.out later on?

Never tried it like that.  Let us know how it works for you.

Easier to use "screen" or system console.

 

Joe L.

This morning I hooked a monitor and keyboard up to it and this is what was on the screen.

 

Seeing as there was no activity on anything, I did a hard reset on the box. Everything ooks like it is back to normal. I will attempt the preclear again.

Just an update. I figured while this was all out I would put a different 1TB in its place in case that drive was faulty and was causing the red ball. while the drive boy was out of the tower I thought I would do some maintenance to the system, clean fans and put in the shorter SATA cables I had gotten but not put in yet.

 

All turned to Crap pretty fast, and the computer didn't post, I ended up pulling the whole thing apart and testing parts in another computer I have. Long story short, all good again, I booted back in, I have access to the shares again, and in the mean time I have stopped the array and with a monitor hooked up directly to the unraid box, am preclearing from there.

 

My unraid box is a little ITX system, which is why doing anything in it becomes a real pain in the ass.

 

Thanks for everyones suggestions.

 

I connected a keyboard and monitor up to the unraid box and I ran preclear from there. The 1TB drive took about 8 hours to complete fully. I could start the array up (with that drive unassigned) and connect to the shares while the preclear took place. Once it finished I was able to assign it to the missing drive in my array and the array has been rebuilding since. I tried to avoid doing anything at all with the server while this has been going (just in case).

 

I will test the other 1TB drive tomorrow and see how it is.

 

Thanks Everyone, I will mark thread as closed