The 4.5.6 version of unRAID introduced a new error message that looks like:

 

Jul 26 00:25:46 saturn kernel: md: parity incorrect: 1477474488
Jul 26 00:30:21 saturn kernel: md: parity incorrect: 1510045800
Jul 26 00:30:25 saturn kernel: md: parity incorrect: 1510600792

 

How is the number (such as 1477474488) interpreted?  It is a byte counter, or number of a disk block or number of a raid stripe?  I'm wondering if this could be used to figure out where on the disks the problem might be, or based on this size perhaps I can eliminate some of my smaller disks from consideration.

 

I am also wondering if there is a way to run the parity check command by hand, but to tell it to just check a range of blocks as I have been observing a number of parity errors (and subsequent corrections) at similar addresses:

 

Jul 19 00:50:35 saturn kernel: md: parity incorrect: 1485297976
Jul 19 00:57:47 saturn kernel: md: parity incorrect: 1536676720
Jul 19 01:12:25 saturn kernel: md: parity incorrect: 1636175680
---
Jul 21 00:30:09 saturn kernel: md: parity incorrect: 1485297976
Jul 21 00:52:02 saturn kernel: md: parity incorrect: 1636175680
---
Jul 23 00:34:04 saturn kernel: md: parity incorrect: 1472497824
Jul 23 00:39:36 saturn kernel: md: parity incorrect: 1512072552
---
Jul 24 00:12:09 saturn kernel: md: parity incorrect: 1472497824
Jul 24 00:17:42 saturn kernel: md: parity incorrect: 1512072552
---
Jul 25 00:56:25 saturn kernel: md: parity incorrect: 1477474488
Jul 25 01:00:59 saturn kernel: md: parity incorrect: 1510045800
Jul 25 01:01:04 saturn kernel: md: parity incorrect: 1510600792
---
Jul 26 00:25:46 saturn kernel: md: parity incorrect: 1477474488
Jul 26 00:30:21 saturn kernel: md: parity incorrect: 1510045800
Jul 26 00:30:25 saturn kernel: md: parity incorrect: 1510600792

 

Regards,

 

Stephen

It's the sector.

 

/* called on a sync error */
void md_sync_error(mddev_t *mddev, sector_t sector)
{
        mdp_super_t *sb = &mddev->sb;

        sb->sync_errs++;
        if (sb->sync_errs <= 20)
                printk("md: parity incorrect: %llu\n", sector);

}

 

It's the sector.

 

/* called on a sync error */
void md_sync_error(mddev_t *mddev, sector_t sector)
{
        mdp_super_t *sb = &mddev->sb;

        sb->sync_errs++;
        if (sb->sync_errs <= 20)
                printk("md: parity incorrect: %llu\n", sector);

}

 

 

And a sector is 512 bytes?  In which case this issue is happening at about the 756MByte point.

 

Thanks,

 

Stephen