I have exactly the same problem. It's very annoying. The only thing you can do to bypass this error is to do a (very time consuming) cold reboot of the box, and at the same time power OFF and then ON the external hard drive. Then the recording is possible again.

My box is the DM800 PVR and my image is the latest Newnigma2 3.2.1. My Hard drive is a WD MyBook 750 and is connected to the Upper USB Slot and reported in the /etc/fstab: /dev/stb1 media/hdd auto defaults 0 0

At the time this error occurs, the drive's contents cannot be explored in any way, although the drive is reported as mounted in Device Manager (Partition sdb1 is mounted on: /media/hdd , The partition has a linux filesystem). Furthermore, I've noticed that this error occurs particularly after a long HDD standby period. It can sometimes happen though after a short standby period as well. I can also happen after a cold reboot but without powering OFF and then On the hard drive.

Any ideas anyone?

Thanks!

As I did some digging, I'm continuing the above post...
I've followed a possible solution from another similar thread that they assumed the drive could be lacking write permissions and they suggested changing chmod permissions at /hdd/media.
I've tried it but in my case the telnet command returned "Read-only file system" error and the permissions could not be changed. This explains why the recording cannot start but does not explain why the drive is sometimes read-only and sometimes not.
So I browsed at a few linux forums to read about this error and I 've found a thread that suggested to check the system log because filesystems are sometimes automatically re-mounted as read only if problems with the device are detected. So I went and checked the /var/log and "boom!" I've found the below messages:

Jun 2 18:31:09 dm800 user.info kernel: [ 2739.492000] sd 2:0:0:0: Device not ready: <6>: Current: sense key=0x2
Jun 2 18:31:09 dm800 user.info kernel: [ 2739.492000] ASC=0x4 ASCQ=0x2
Jun 2 18:31:09 dm800 user.warn kernel: [ 2739.492000] end_request: I/O error, dev sdb, sector 1311105
Jun 2 18:31:09 dm800 user.crit kernel: [ 2739.494000] EXT3-fs error (device sdb1): ext3_get_inode_loc: unable to read inode block - inode=675, block=163888
Jun 2 18:31:09 dm800 user.err kernel: [ 2739.494000] Aborting journal on device sdb1.
Jun 2 18:31:09 dm800 user.info kernel: [ 2739.533000] sd 2:0:0:0: Device not ready: <6>: Current: sense key=0x2
Jun 2 18:31:09 dm800 user.info kernel: [ 2739.533000] ASC=0x4 ASCQ=0x2
Jun 2 18:31:09 dm800 user.warn kernel: [ 2739.533000] end_request: I/O error, dev sdb, sector 449
Jun 2 18:31:09 dm800 user.err kernel: [ 2739.533000] Buffer I/O error on device sdb1, logical block 56
Jun 2 18:31:09 dm800 user.warn kernel: [ 2739.533000] lost page write due to I/O error on sdb1
Jun 2 18:31:09 dm800 user.info kernel: [ 2739.578000] sd 2:0:0:0: Device not ready: <6>: Current: sense key=0x2
Jun 2 18:31:09 dm800 user.info kernel: [ 2739.578000] ASC=0x4 ASCQ=0x2
Jun 2 18:31:09 dm800 user.warn kernel: [ 2739.579000] end_request: I/O error, dev sdb, sector 1
Jun 2 18:31:09 dm800 user.err kernel: [ 2739.579000] Buffer I/O error on device sdb1, logical block 0
Jun 2 18:31:09 dm800 user.warn kernel: [ 2739.579000] lost page write due to I/O error on sdb1
Jun 2 18:31:19 dm800 user.crit kernel: [ 2749.485000] ext3_abort called.
Jun 2 18:31:19 dm800 user.crit kernel: [ 2749.485000] EXT3-fs error (device sdb1): ext3_journal_start_sb: Detected aborted journal
Jun 2 18:31:19 dm800 user.crit kernel: [ 2749.485000] Remounting filesystem read-only

As I'm not a linux expert, can someone enlighten me of what exactly is wrong with the drive?

Thanks!

I first tried to set write and read access to 777 directory hdd, but it did not help so I created the folder and movie then set the read and write to permissions to 777 on this too. Then it would record, hope this is of help to you too