[comp.unix.questions] NFSSTAT's don't match

ramsey@sundance.llnl.gov (Susanne Ramsey) (09/22/90)

I have been trying to track down a network client/server problem for the last 
couple of days and have observed something that does not make sense to me.

HISTORY:
	I have a 4/330 server and 1 diskless client and 2 diskfull clients.  
	The diskless client has been getting NFS server not responding errors
	that cause the user GREAT heartburn.  Anyway we have been playing with
	such things as increasing biod's on client and nfsd's on server, 
	swapping out the clients motherboard  and a list of others that I 
	won't go into here..

	While trying to fix this problem we were running nfsstat ..  The 
	following is how it looked on the SERVER system 

Server rpc:
calls      badcalls   nullrecv   badlen     xdrcall
2055       0          0          0          0          

Server nfs:
calls      badcalls
2055       0          
null       getattr    setattr    root       lookup     readlink   read       
0  0%      395 19%    11  0%     0  0%      297 14%    72  3%     453 22%    
wrcache    write      create     remove     rename     link       symlink    
0  0%      757 36%    23  1%     7  0%      1  0%      0  0%      0  0%      
mkdir      rmdir      readdir    fsstat     
0  0%      0  0%      24  1%     15  0%     

Client rpc:
calls      badcalls   retrans    badxid     timeout    wait       newcred
1344       0          0          0          0          0          0          

Client nfs:
calls      badcalls   nclget     nclsleep
1344       0          1344       0          
null       getattr    setattr    root       lookup     readlink   read       
0  0%      59  4%     0  0%      0  0%      81  6%     0  0%      129  9%    
wrcache    write      create     remove     rename     link       symlink    
0  0%      1066 79%   4  0%      0  0%      0  0%      0  0%      0  0%      
mkdir      rmdir      readdir    fsstat     
0  0%      0  0%      5  0%      0  0%      

======>	This is the output from the client system ...

Client rpc:
calls      badcalls   retrans    badxid     timeout    wait       newcred
6618       0          150        0          150        0          0          

Client nfs:
calls      badcalls   nclget     nclsleep
6618       0          6618       0          
null       getattr    setattr    root       lookup     readlink   read       
0  0%      1028 15%   24  0%     0  0%      780 11%    199  3%    1906 28%   
wrcache    write      create     remove     rename     link       symlink    
0  0%      2475 37%   44  0%     19  0%     4  0%      1  0%      0  0%      
mkdir      rmdir      readdir    fsstat     
0  0%      0  0%      86  1%     52  0%     
  
MY QUESTION:
	Shouldn't the number of retrans and timeouts on the client be showing
	up on the server...??? Who is lying the CLIENT, the SERVER or my EYES??

	Any and all comments are welcome...  
 
Thanks - 
	Susanne
 
*****************************************************************************
Susanne Barbera-Ramsey				ramsey@sundance.llnl.gov
Lawrence Livermore National Laboratory		(415)423-9530
PO Box 808 L-630
Livermore CA 94550  
******************************************************************************
	"What's that thing?"
        "Well, it's a highly technical, sensitive instrument we use in
computer repair.  Being a layman, you probably can't grasp exactly what
it does.  We call it a two-by-four."
                -- Jeff MacNelly, "Shoe"
******************************************************************************