On Tue, Jun 20, 2006 at 05:23:38PM -0700, Chuck Lever wrote: > On 6/20/06, Steve Kargl <sgk_at_troutmask.apl.washington.edu> wrote: > > > >I've seen the problem described by David on RELENG_6. > > > >You can see the network topology at > >http://troutmask.apl.washington.edu/~kargl/hpc.html > >The NFS server is on node10 (bge1) and /home is exported to node11 > >through node15. If I mount node10:/usr/src on node15 a "make installworld" > >will result in a never ending stream of "server not responding messages". > > Has anyone who has seen this issue tried using the dumb timer mount option? > I did not try that option in that the description suggested to me that -d is applicable to slow networks. Before I tried -d, I decided to see what happens. Interesting. node11:kargl[214] rsync -av kargl_at_node10:prj /tmp ... sent 11970 bytes received 34071110 bytes 22722053.33 bytes/sec total size is 34032098 speedup is 1.00 node11:kargl[215] rm -rf /tmp/prj node11:kargl[216] time cp -R ~/prj /tmp nfs server master:/home: not responding nfs server master:/home: not responding nfs server master:/home: not responding /tmp is a memory filesystem and /home/kargl is NFS exported. -- SteveReceived on Tue Jun 20 2006 - 22:39:09 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:57 UTC