Andriy Gapon
2017-Mar-16 07:51 UTC
moutnroot failing on zpools in Azure after upgrade from 10 to 11 due to lack of waiting for da0
On 13/03/2017 21:07, Edward Tomasz Napiera?a wrote:> Are you sure the above transcript is right? There are three reasons > I'm asking. First, you'll see the "Root mount waiting" message, > which means the root mount code is, well, waiting for storvsc, exactly > as expected. Second - there is no "Trying to mount root". But most > of all - for some reason the "Mounting failed" is shown _before_ the > "Root mount waiting", and I have no idea how this could ever happen.Edward, your observation is not completely correct. https://www.twisted.org.uk/~pete/914893a3-249e-4a91-851c-f467fc185eec.txt We have: Trying to mount root from zfs:rpool/ROOT/default []... <==vmbus0: version 3.0 ... storvsc0: <Hyper-V IDE Storage Interface> on vmbus0 Solaris: NOTICE: Cannot find the pool label for 'rpool' Mounting from zfs:rpool/ROOT/default failed with error 5. <==Root mount waiting for: storvsc <==... So, the kernel attempted to mount the root even before vmbus was attached and, thus, before storvsc appeared and informed the kernel that it might be holding the root. How ZFS was supposed to know that vmbus is ever going to appear? To me this sounds more like a problem with the Hyper-V drivers. -- Andriy Gapon
Pete French
2017-Mar-16 11:18 UTC
moutnroot failing on zpools in Azure after upgrade from 10 to 11 due to lack of waiting for da0
> So, the kernel attempted to mount the root even before vmbus was attached and, > thus, before storvsc appeared and informed the kernel that it might be holding > the root. > How ZFS was supposed to know that vmbus is ever going to appear? > To me this sounds more like a problem with the Hyper-V drivers.I am currently running with the patch which waits for a number fo seconds and retries the mount, and that appears t fix it. However I dont really like rnning a patched OS. How would I set about reporting this to Microsoft and getting it fixed, or getting the timeoutpatch commited ? Preferably both, as the timeout patch is generally a useful thing to have working for ZFS I think. -pete.