Hi Kelsey, On Mon, 2012-12-03 at 17:59 +0000, Prantis, Kelsey wrote: > > On 11/20/12 12:36 PM, "Prantis, Kelsey" <kelsey.prantis@xxxxxxxxx> wrote: > > > >I applied this patch to our setup, but we don't seem to have gotten any > >messages involving "vbr bad magic". I wonder if perhaps we are seeing a > >different oops than Nicholas had investigated? Or perhaps due to a > >difference in versions? > > > >Debug output we did get here: > >ftp://ftp.whamcloud.com/uploads/lio-debug-5.txt.bz2 > > > >... > > Hello folks, > > It's been nearly two weeks since I sent my last response here, wondering > if everyone is sitting around assuming the ball is in someone else's > court, or it has just fallen off the radar in general. We didn't seem to > hit the debug code provided in our last oops - not sure if its because we > hit a different oops than we were trying to troubleshoot or if we are > barking up the wrong tree in our troubleshooting, but would appreciate if > we could figure out why as our target server is still unusable. > Unfortunately, I've been busy with some external items recently and have not been able to get back to help identifying this bug. AFAICT it's still something specific to virtio-blk backends with at least 1 MB requests, and is confirmed independent of target fabric driver (eg: it's reproducible with tcm_loop as well). As mentioned, you can still use a lower max_sectors_kb on your iSCSI initiator as a work-around until the bug can be identified. Thank you, --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
