I have seen several people on this list complain about ypbind not
being able to bind to a host. I'm interested to find out whether this
is an isolated occurence (i.e. likely a configuration error) or whether
there is a pattern to this.
My set of circumstances:
Debian distribution: 2.1
Kernel 2.2.1
ypbind: 1.3 (doesn't works)
1.4 (partially working)
ypbind host: Sun4/50 (Sparc IPX)
NIS master: Sun4/65 (Sparc 1+), SunOS 4.1.2
Symptoms of problem
ypbind will start, but won't bind and won't terminate until
interrupted. Here's a sample output of ypbind -d (v. 1.3)
Trying entry: domain ocgy.ubc.ca server nessie.ocgy.ubc.ca
parsed domain 'ocgy.ubc.ca' server 'nessie.ocgy.ubc.ca'
add_server() domain: ocgy.ubc.ca, host: nessie.ocgy.ubc.ca,
nobroadcast, slot: 0
[Welcome to ypbind-mt, version 1.3]
ping host 'nessie.ocgy.ubc.ca', domain 'ocgy.ubc.ca'
Pinging all active server.
ping host 'nessie.ocgy.ubc.ca', domain 'ocgy.ubc.ca'
Pinging all active server.
ping host 'nessie.ocgy.ubc.ca', domain 'ocgy.ubc.ca'
... repeats every 20 seconds ...
Any attempt to run any yp utilities (e.g. ypwhich) will produce
ypbind
ypbindproc_domain_2_svc (ocgy.ubc.ca)
trylock = success
ping host 'nessie.ocgy.ubc.ca', domain 'ocgy.ubc.ca'
Status: YPBIND_FAIL_VAL
ypwhich
can't yp_bind: Reason: Domain not bound
Workaround
I was finally able to ypbind to work by
1) Upgrading to v. 1.4
2) Running by binding via broadcast (-broadcast) and under
debug (-d"). This produces lots of verbiage to
console.
Maybe it's the threading code that's breaking. I'll try an older an
older non-threading ypbind and see whether that fixes it.
Joseph Tam <[EMAIL PROTECTED]>
