[Lustre-discuss] (no subject)

Jan H. Julian julian at arsc.edu
Mon Jul 7 16:58:46 PDT 2008


While running cat /proc/sys/lnet/peers across all of our client node  
a consistently see all nodes marking two NID's with state "down" and  
many NIDS with a min in the negative range.  Seem as though this  
could not be a good state, but we are running???
We're still in 1.4.11.  Does this indcate a problem???
...
mg30
----------------
  nid                      refs state   max   rtr   min    tx   min  
queue
  10.10.7.238 at vib             1    up     8     8     8     8     2 0
  10.10.7.239 at vib             1  down     8     8     8     8     2 0
  10.10.7.240 at vib             1  down     8     8     8     8     2 0
  10.10.7.241 at vib             1    up     8     8     8     8     7 0
  10.10.7.235 at vib             1    up     8     8     8     8     2 0
  10.10.7.236 at vib             1    up     8     8     8     8     2 0
  10.10.7.237 at vib             1    up     8     8     8     8     2 0
----------------
mt023
----------------
  nid                      refs state   max   rtr   min    tx   min  
queue
  10.10.7.238 at vib             1    up     8     8     8     8   -16 0
  10.10.7.239 at vib             1  down     8     8     8     8   -66 0
  10.10.7.240 at vib             1  down     8     8     8     8   -63 0
  10.10.7.241 at vib             1    up     8     8     8     8     4 0
  10.10.7.235 at vib             1    up     8     8     8     8    -8 0
  10.10.7.236 at vib             1    up     8     8     8     8    -8 0
  10.10.7.237 at vib             1    up     8     8     8     8    -8 0
...

Jan H. Julian
System Administrator
Midnight Cluster Activity Lead
julian at arsc.edu
907-450-8641






More information about the lustre-discuss mailing list