[Solved] Problem after adding new controller with one disc

[Eisi2005]

Hi all,

 

i have added today a new 4 port controller from silicon image 3124, with one new 1,5 tb disk. I´m running unraid 4.7.

 

After Starting server , i must add the new disk and all was fine, after formatting the server share comes back and i could see all my files.

 

Then i want to start my cache dirs and see in my /boot/custom folder many dirs and files with crypted signs. And my sh files i have put there where gone.

 

In the log file i have massiv errors with :

 

Nov  1 14:38:02 Server kernel:    fat_bmap_cluster: request beyond EOF (i_pos 1793675)

Nov  1 14:38:02 Server kernel: FAT: Filesystem error (dev sdb1) (Errors)

 

Could someone help what to do ? Could i delete these files ?

Could i install cache_dirs new ?

 

These are the files in the custom folfer:

 

 

 

Greets

Eisi

syslog-2012-11-01.zip

[Eisi2005]

Something that i noticed. I dont have any rights on the flash dir. With root i could not delte anythin in these dir.

 

Greets

Eisi

 

[JonathanM]

I'd try to back up the /boot/config folder in total over the network first, then shut down the machine, pull the flash drive and run a file system check on it in another machine. The screen shot you posted looks like file system corruption on your flash drive.

[Eisi2005]

@jonathanm

 

Thanks for your answer. Could i try the file system check with a normal win 7 pc ?

 

Greets

Eisi

 

[JonathanM]

Could i try the file system check with a normal win 7 pc ?

Yes.

[Eisi2005]

@jonathanm

 

first the parity check have finished so i could test your hint in the afternoon.

 

But now is parity finished and i got these error msg in log.

Nov  2 08:45:18 Server kernel: md: parity incorrect: 2930275568 (Errors)

Nov  2 08:45:18 Server kernel: md: parity incorrect: 2930276080 (Errors)

Nov  2 08:45:18 Server kernel: md: parity incorrect: 2930276592 (Errors)

Nov  2 08:45:18 Server kernel: md: sync done. time=58307sec rate=25128K/sec (unRAID engine)

Nov  2 08:45:18 Server kernel: write_file: error 30 opening /boot/config/super.dat (Errors)

Nov  2 08:45:18 Server kernel: md: could not write superblock from /boot/config/super.dat (unRAID engine)

Nov  2 08:45:18 Server kernel: md: recovery thread sync completion status: 0 (unRAID engine)

 

Is this important ?

 

Greets

Eisi

 

[JonathanM]

Nov  2 08:45:18 Server kernel: write_file: error 30 opening /boot/config/super.dat (Errors)

Nov  2 08:45:18 Server kernel: md: could not write superblock from /boot/config/super.dat (unRAID engine)

Nov  2 08:45:18 Server kernel: md: recovery thread sync completion status: 0 (unRAID engine)

 

Is this important ?

Yes. Your flash drive did not save the status of your array because it is mounted as read only, or worse. If you can save the entire contents of your config folder over the network NOW, BEFORE YOU SHUT DOWN, I would. I was hoping you would attempt to back up your config folder immediately.

[Eisi2005]

@jonathanm

 

yes i have a backup of my config folder.

 

I have now make the filecheck on my win7 pc. The check deleted the complete custom folder and correct some files. Now i have boot the unraid again, i dont have anymore the fat filesystem errors.

 

Nov  1 14:38:02 Server kernel:    fat_bmap_cluster: request beyond EOF (i_pos 1793675)

Nov  1 14:38:02 Server kernel: FAT: Filesystem error (dev sdb1) (Errors)

 

Now a new parity check has started. Is this ok ? If the parrity check writes the super.dat. Could i then use the version of unraid that is on my stick now ?

 

What should i do with the 3 errors from last parrity check ? Should i do a smart check ?

 

Greets and thanks for your help

 

Eisi

 

[JonathanM]

yes i have a backup of my config folder.

Good! At least if your USB drive dies, it won't be so hard to put things back on a new stick.

Now a new parity check has started. Is this ok ? If the parrity check writes the super.dat.

That is to be expected, when the super.dat file isn't updated, unraid starts a correcting parity check automatically on the next boot. Assuming no more syslog errors, you are most likely ok to keep using the stick as is. I'd probably not put any add-ons back for a few days just to make sure the array is stable.

[Eisi2005]

@jonathanm

 

the parity check has finished and the super.dat has a new time stamp. No error message with write access on super.dat. This is the good news.

 

The parity check says the following:

Last parity check < 1 day ago .  Parity updated 3 times to address sync errors

The parity check start from alone after rebooting the server, so i don´t know if the check correct the parity or not ?

I have make a short smart test for all my disc, the reports and the syslog are in the attachment.

All disc seems to be fine what i could see, only Datadisc 1 seems to have a problem. The smart status report for disc 1 says that the check was aborted after 90%. I have started the check 4 times and every time the same message.

 

Is this a problem ? Must i do anything ? Could you give me a hint again ?

 

Greets and many many thanks

 

Eisi

Logfiles.zip

[dgaschk]

The automatic check is the correcting type. Run the check again and there should be zero errors.

[Eisi2005]

@dgaschk

 

you are rigth, no new errors after checking again, thanks.

 

Greets

Eisi