Open System Services NFS Management and Operations Guide

Table 2 Quickstart for Troubleshooting OSS NFS (continued)
Recommended ProcedureSymptom
All EDIT files show up with a
zero EOF
1. Verify that the Expand file system servers are running.
2. Check in the ZEFSCONF file that the disk volume in question is configured for the
Expand subsystem.
3. Make sure the Expand servers are running with the correct process names.
An OSS NFS server will not
start
1. Verify that the underlying OSS fileset is defined; the name of the fileset must match
the value of the ROOT server attribute.
2. Make sure that the fileset is started: Issue an SCF STATUS FILESET command. If the
fileset is not started, use the SCF START FILESET command to start it.
Typical Server-Failure Symptoms
This topic discusses the common symptoms an NFS client experiences when an OSS NFS server
fails. Although the error messages shown in this topic are specific to the SunOS version of the
UNIX operating system, they should give you an idea of how the messages might look on other
platforms. For specific information on NFS client error messages, see the NFS client’s documentation.
When the OSS NFS server experiences problems, programs that access hard-mounted remote files
fail differently from those that access soft-mounted remote files. Hard-mounted remote file hierarchies
cause the client’s system to retry the requests until the file server responds. Soft-mounted remote
file hierarchies cause the client’s system calls to return an error after retrying a given number of
times.
When a file hierarchy is hard-mounted, a program that tries to access it waits in a retry loop if the
server fails to respond. In this case, this message is posted on the client’s console:
NFS server hostname not responding, still trying
When the server finally responds, this message appears on the client’s console:
NFS server hostname ok
A program accessing a soft-mounted file hierarchy whose server does not respond receives this
message on the client’s console:
... hostname server not responding: RPC:Timed out
Recommended Troubleshooting Strategy
This topic describes a troubleshooting strategy that is applicable to most of the OSS NFS problems
you might encounter. As you gain more experience in troubleshooting the OSS NFS subsystem,
you can check the subsystem components in a different order than the order recommended here.
Problem Diagnosis Tools
• The TACL STATUS command displays whether OSS NFS components are, in fact, running.
• The SCF TRACE commands can log OSS NFS commands and responses to assist in problem
diagnosis. You might be asked, by a service provider, to use the TRACE facility to gather
diagnostic data.
• The EMS logs contain EMS event messages for the related subsystems and software. EMS is
the only way that OSS NFS servers can report problems after the servers are started.
• Third-party line-tracing products can also be helpful for troubleshooting, especially those that
can interpret and format NFS and RPC messages.
Components to Troubleshoot
Figure 7 shows the components examined in this topic. The steps listed in this topic start with the
leftmost component in Figure 7 and proceed to the right, one component at a time.
Typical Server-Failure Symptoms 73