[sheepdog-users] qemu-nbd and dd

Valerio Pachera sirio81 at gmail.com
Wed Feb 5 13:21:09 CET 2014


I think it's sheepdog related problem.
I used the same procedure on the same hosts but using a regular file
instead of a vdi and it didn't stop.

I also updated sheepdog and qemu
  Sheepdog daemon version 0.8.0_31_gf2a00e1
  QEMU emulator version 1.7.50

This is what I get in sheep.log when it hangs
Feb 05 12:08:52   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:08:52   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:08:57   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:08:57   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:02   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:02   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:07   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:07   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:12   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:12   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:17   WARN [gway 2981] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again
Feb 05 12:09:17   WARN [gway 3013] wait_forward_request(413) poll timeout
1, disks of some nodes or network is busy. Going to poll-wait again

I tested it with and without erasure code.
It seem ti hangs before using erasure code.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.wpkg.org/pipermail/sheepdog-users/attachments/20140205/704ba33a/attachment-0005.html>


More information about the sheepdog-users mailing list