AMD Problem (I think):
We are having a problem with our mail server running DU 4.0D. Or rather,
we recenty upgraded the system to DU 4.0D and have since had problems. The
machine in question serves its /local/spool/mail directory via amd (am
utils 6.0a6; see amd.var below) to all hosts requesting it at
/var/spool/mail (including itself) and mounts the home directories of each
user appearing as /home/#host#/username in the nis passwd file in order to
read any .forwards that may exist - as well as some other filesystems
(/usr/local/bin /lib etc...).
The message "NFSx server xxx not responding still trying" constantly
appears since the install and most systems cannot access
host:/var/spool/mail directory. Interrestingly enough, the machine that
serves /local/spool/mail cannot access it either to mount the dir as
/var/spool/mail (causing some rather annoying problems with pop, people
trying to read their mail etc..). The catch is, we have been using am
utils 6.0a6 automounter for a while now (with other 4.0D machines) and have
not had any problems like this before. Could this be due to the nfs access
requests that this machine is under?
The machine is not loaded (load ~0.03, however it is running slow as a
result of this constant mounting and umounting) as regular users are not
allowed on the server. It is configued as an nis slave. Other env configs
are shown below in the included rc.config file.
I am assuming that something is causing amd to be a problem and have
checked the osf archives but the closeset problem was due to a full duplex
switch (not the problem).
(
http://www.ornl.gov/its/archives/mailing-lists/alpha-osf-managers/1997/10/m
sg00270.html)
Can anyone help or suggest things I should be looking at?
Thanks,
JS Diaz
ANY responses *please*, *please* copy:
mmoran_at_mediaone.net
Exerpt of /syslog.dated/01-Oct-10:33/kern.log
Oct 1 17:11:24 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:26 harpo vmunix: NFS2 server amd not responding still trying
Oct 1 17:11:26 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:26 harpo vmunix: NFS2 server amd not responding still trying
Oct 1 17:11:26 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:27 harpo vmunix: NFS2 server amd not responding still trying
Oct 1 17:11:27 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:27 harpo vmunix: NFS2 server amd not responding still trying
Oct 1 17:11:27 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:27 harpo vmunix: NFS2 server amd not responding still trying
Oct 1 17:11:27 harpo vmunix: NFS2 server amd ok
Oct 1 17:11:28 harpo vmunix: NFS2 server amd not responding still trying
output of "ypcat amd.master"
amd.var -type:=direct
amd.local -cache:=all
amd.home
output of "ypcat amd.var"
# This must be mounted with the -type:=direct option
/defaults
opts:=rw,grpid,intr,wsize=8192,rsize=8192,intr,nodevs,timeo=12;type:=nfs
var/spool/mail type:=nfs;rhost:=harpo;rfs:=/local/spool/mail
most of /etc/rc.config (edited a tad to make it a little shorter)
DISPLAYTYPE=
HOSTNAME="harpo.wi.mit.edu"
NUM_NETCONFIG="1"
MAX_NETDEVS="16"
NETDEV_0="tu0"
edited/NETDEV_x=
IFCONFIG_0="18.157.0.35 netmask 255.255.0.0"
edited/FCONFIG_x=
RWHOD="yes"
ROUTED="no"
ROUTED_FLAGS=
ROUTER="no"
GATED="no"
GATED_FLAGS=
JOIND=
JOIND_FLAGS=
PAGERAW="1"
PAGEFILE="/dev/rz0b"
PARTITION="rz0b"
PARTITIONTYPE="RZ26F"
PAGEMINSZ=0
PAGEMAXSZ=0
QUOTA_CONFIG=
NETDEV_7
XLOGIN="cde"
LAT_SETUP="1"
BIND_CONF="YES"
BIND_SERVERTYPE="CLIENT"
BIND_SERVERARGS="NONE"
NIS_CONF="YES"
NIS_TYPE="SLAVE"
NIS_DOMAIN="nisdomain"
NIS_ARGS=" -S nisdomain,harpo.wi.mit.edu,nishost3,nishost4,nishost5"
NFS_CONFIGURED="1"
NFSSERVING="1"
NFSLOCKING="1"
PCNFSD="0"
NUM_NFSIOD="7"
NUM_NFSD="16"
XNTPD_CONF="YES"
XNTP_SERV1="ntp.wi.mit.edu"
XNTP_SERV2=""
XNTP_SERV3=""
XNTPD_OPTS="-g"
AMD="1"
AMD_ARGS="-x noinfo -p -r -l syslog -- `ypcat -k amd.master`"
NUM_TCPD="32"
NUM_UDPD="24"
NUM_PROPLISTD="0"
NONROOTMOUNTS="1"
AUTOMOUNT="0"
MOUNTOPTS="-i -n"
------
J S Diaz
jsd_at_genome.wi.mit.edu
Systems Group 617.252.1988
MIT Whitehead Institute - Genome Center
Received on Thu Oct 01 1998 - 22:14:16 NZST