kernel crash at zfs/zio.c:918

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Ejected UBS HDD formated with zfs.
2. Inserted HDD again.
3. Kernel crashed.

See the crash log:
https://gist.github.com/kenorb/7553888

Related:
https://groups.google.com/forum/#!topic/zfs-macos/kjvyl65XuDw

Backtrace:
panic(cpu 0 caller 0xffffff7f866c3e60): "ZFS: I/O failure (write on <unknown> 
off 0: zio 0xffffff8028049298 [L0 DMU dnode] 4000L/c00P 
DVA[0]=<0:8340035000:c00> DVA[1]=<0:80028000:c00> fletcher4 lzjb LE contiguous 
birth=392 fill=23 
cksum=d25046830d:160b700ffe45f:163c7b703a978b4:9e7159c29ffc4e1): error " 
"6"@/Users/bj/new-mac-zfs-bjoka/usr/src/uts/common/fs/zfs/zio.c:918
      Kernel Extensions in backtrace:
         org.maczfs.zfs.fs(74.3.1)[613592BF-7F84-A446-344E-C33628BE7C75]@0xffffff7f86654000->0xffffff7f866e0fff

Original issue reported on code.google.com by [email protected] on 19 Nov 2013 at 10:43

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Another crash, but this time I didn't do anything, just writing some files into 
my hdd.

panic(cpu 6 caller 0xffffff7f912c3e60): "ZFS: I/O failure (write on <unknown> 
off 0: zio 0xffffff8032f95020 [L0 DMU dnode] 4000L/1200P 
DVA[0]=<0:30c06bb000:1200> DVA[1]=<0:5c80377000:1200> fletcher4 lzjb LE 
contiguous birth=4763 fill=32 
cksum=134bcfc23b3:2e1513eb808c9:4476c73211787b7:c352c0698bfdbdf9): error " 
"6"@/Users/bj/new-mac-zfs-bjoka/usr/src/uts/common/fs/zfs/zio.c:918

Original comment by [email protected] on 21 Nov 2013 at 2:38

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

Hmm, it looks like the drives disappear from the system without prior 
notification.
ZFS is designed to panic in that case, because there is no safe way to continue 
operation.
You must export a pool before physically pulling the drive in the pool.
I'll leave the report open for now, but I don't think this is a bug.

Original comment by [email protected] on 22 Nov 2013 at 9:31

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

Thanks for your comment.
Today I had 3 the same crashes (as usual) and I didn't ejected my drive as I 
was in the middle of copying and downloading the data into my HDD.
I had 2 crashes when actually uncompressing the compressed file from the hard 
drive.
My HDD is: WD My Passport 1TB USB 1000GB (Model: WDBBEP0010BBK).
OS: 10.9

Let me know if you have any suggestion.

Original comment by [email protected] on 22 Nov 2013 at 11:15

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

The crash happens all the time when restoring my laptop from sleep time and I'm 
simply re-connecting the HDD.

Original comment by [email protected] on 25 Nov 2013 at 9:39

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

This is probably the duplicate of:
Issue 51:   USB not supported

Original comment by [email protected] on 25 Nov 2013 at 9:40

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

Again, when connecting the USB drive:
    panic(cpu 1 caller 0xffffff7f91ac3e60): "ZFS: I/O failure (write on <unknown> off 0: zio 0xffffff8065bb6d10 [L0 persistent error log] 4000L/400P DVA[0]=<0:1787033000:400> DVA[1]=<0:4380673000:400> DVA[2]=<0:9542a23c00:400> fletcher4 lzjb LE contiguous birth=28246 fill=1 cksum=5f9e620b9a:3f055d1cd0cc:16d12dbb4522fe:5e6239923165272): error " "6"@/Users/bj/new-mac-zfs-bjoka/usr/src/uts/common/fs/zfs/zio.c:918

Original comment by [email protected] on 6 Dec 2013 at 11:05

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

I had this also happen to me. I have a macbook with two thunderbolt ports that 
I use to connect two separate volumes on separate disks to (via seagate 
thunderbolt adapters). I was copying files from one disk to another when I had 
the exact same panic happen.

Original comment by [email protected] on 11 Mar 2014 at 4:41

• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

I have something similar happening - thunderbolt disk with WD green 3tb drives. 
The mirrored pool works and stays mounted on an old PPC mac - albeit with 
performance issue.

I have tried mounting pool with one of the 2 disks (status degraded) and have 
the same issue - after perhaps 30 seconds the kernel panic occurs. 

Sat Mar 15 15:20:55 2014
panic(cpu 5 caller 0xffffff7faa5efda0): "ZFS: I/O failure (write on <unknown> 
off 0: zio 0xffffff8042e16070 [L0 persistent error log] 4000L/600P 
DVA[0]=<0:900075e00:600> DVA[1]=<0:8c00058c00:600> DVA[2]=<0:20c0003fc00:600> 
fletcher4 lzjb LE contiguous birth=9128 fill=1 
cksum=5d27a6c4a2:644cd2b18c2a:3a4fb3c433c0b4:17dd77b9f6665922): error " 
"6"@/Users/bj/new-mac-zfs-bjoka/usr/src/uts/common/fs/zfs/zio.c:918

I have the crash logs and can reproduce at will if required.

Not performing any file operations, just post mounting the drive(s). 

Original comment by [email protected] on 15 Mar 2014 at 4:39

• Added labels: ****
• Removed labels: ****