[pve-devel] [PATCH] migrate: syncdisk : avoid scanning shared storage

[datanom.net]

On 07-16-2012 15:19, Alexandre DERUMIER wrote:
> Seem strange that user report hang, as nfs default timeout is 0.7
> second according the doc.
>
The picture is more muddy than alone looking at the NFS doc. If a NFS 
share disappears
while writing to the share and nfslock is active then nfslock on the 
client and nfsd
on the server will leave dangling file locks hanging. If this is 
combined with the
mount option hard then the service will be lock until nfsd is restartet 
on the server.

 From the client perspective this situation will be seen as a socket 
which hangs forever
and never do timeout.

Some interesting mount options which have a huge impact on NFS and NFS 
behavior:

        soft / hard    Determines the recovery behavior of the NFS 
client after
                       an NFS request times out.  If neither option  is  
speci‐
                       fied  (or if the hard option is specified), NFS 
requests
                       are retried indefinitely.  If the soft option is  
speci‐
                       fied,  then  the  NFS  client fails an NFS 
request after
                       retrans retransmissions have been sent, causing 
the  NFS
                       client to return an error to the calling 
application.

                       NB:  A  so-called  "soft"  timeout can cause 
silent data
                       corruption in certain  cases.  As  such,  use  
the  soft
                       option only when client responsiveness is more 
important
                       than data integrity.  Using NFS over TCP  or  
increasing
                       the value of the retrans option may mitigate some 
of the
                       risks of using the soft option.

        lock / nolock  Selects whether to use the NLM sideband protocol 
to lock
                       files on the server.  If neither option is 
specified (or
                       if lock is specified), NLM  locking  is  used  
for  this
                       mount point.  When using the nolock option, 
applications
                       can lock files, but such locks  provide  
exclusion  only
                       against  other  applications running on the same 
client.
                       Remote applications are not affected by these 
locks.

                       NLM locking must be disabled with the nolock 
option when
                       using NFS to mount /var because /var contains 
files used
                       by the NLM implementation on Linux.   Using  the  
nolock
                       option  is  also  required  when mounting exports 
on NFS
                       servers that do not support the NLM protocol.

        intr / nointr  Selects whether to allow signals to interrupt 
file oper‐
                       ations  on this mount point. If neither option is 
speci‐
                       fied (or if nointr is specified), signals do not  
inter‐
                       rupt  NFS  file operations. If intr is specified, 
system
                       calls return EINTR if an in-progress  NFS  
operation  is
                       interrupted by a signal.

                       Using  the  intr  option  is preferred to using 
the soft
                       option because it is significantly less likely to 
result

                       in data corruption.

                       The  intr / nointr mount option is deprecated 
after ker‐
                       nel 2.6.25.  Only SIGKILL can interrupt  a  
pending  NFS
                       operation on these kernels, and if specified, 
this mount
                       option is ignored  to  provide  backwards  
compatibility
                       with older kernels.