# File:		hostmon-confg
# 	Config file for hostmon-master
#
## Regular expressions must NOT HAVE ANY '|' marks in it.
## For multiple lines describing a variable-  The LAST line should be the 
## 'default' (or empty) line. Note that the order is important, and the
## first match stops any more searches.

# You can put a '-1' in any of the thresholds to never let the value
# reach that threshold (for decreasing values). Put a very hi number
# for increasing thresholds for same functionality.

# Select the pollinterval to be larger than the cycletime of the
# hostmon.client programs, else this program will get stale data.
# About 10-15 minutes is a good value.

# List all the hosts after the 'STARTHOSTS' keyword.

Pollinterval	10	

## THRESHOLDS CAN BE IN INCREASING OR DECREASING ORDER
##
#  <variable>  <host regex>  wthres ethres cthres  [comment regex]
#
# The host-regex  of '*' is automatically converted into '.+' by
# hostmon.

# Like to know if Sybase server crashes.
Uptime		syserv	20 15 10
Uptime		*	10 9 8

Load5		*	2 5 8

FileTable	*	50  70  90
InodeTable	*	60  70  90
ProcTable	*	65  70  80

# Warn if free swapspace decreases below this many MBytes
SwapSpace	*	30 20 10

# NFS retransmission rate (BSDI has retry & timeout values)
#	retrans > 5% of calls	=> NFS server overload
#	timeout hi		=>
#	badxid = timeout	=> NFS server overload
#	timeout + retrans hi, badxid low	=> Network overload

NFStimeouts	*	5 7 10
NFSretries	*	5 7 10

# Used disk space (in MBytes available)
# Note the default with a critical threshold of 101 so that it never
# goes into critical.
DFspace_avail	hosta|hostb	10 7 5	^/$
DFspace_avail	*	5 4 3		^/$
DFspace_%used	*	80 90  100	^/usr$
DFspace_%used	*	80 90  97	^/usr/local$
DFspace_%used	*	80 90  97	^/home$
DFspace_%used	*	80 90  97	^/var/spool$
DFspace_%used	*	90 95  101

DFinodes_%used	*	70 80  90	
DFinodes_avail	*	7000 5000 1000

# The IObw is probably available only on SunOS and Solaris
IOseek		*	5  10  15
IObw		*	20 30  40

# Dont want to see SwapIn SwapOut get to Critical level. Set unreasonable
# third threshold.
# SwapIn is normal,   SwapOut is BAD
SwapIn		*	10  20 999
SwapOut		*	5   10 20

# Paging is normal
PageIn		*	10  20 999
PageOut		*	5   10 999

MemFree		*	20  15  10

CtxtSw		compu.*	500 700 999
CtxtSw		*	50  150 999

# We dont want to see critical status if CPU is very busy. Flag at Error
# for a trace record...
CPUus		*	50  60  70
CPUsy		*	50  60  70
CPUidle		*	98  99  -1

NetIErr		*	10  15  25
NetOErr		*	10  15  25
NetColl		*	10  15  25

# mail queue length monitor
#
# QLocal is total queue length on any system (warn, err, crit)
# QDest  is the total mail queue for any given destination
#
MailQLocal	*	40  60  200
MailQDest	*	10  30  100	uucp-relay.remote.domain
MailQDest	*	20  40  100

###
############## START of HOST LIST ###############

STARTHOSTS

## All the main fileservers
acctserv
techserv
dataserv

## Workhorses
# dataserv # already in domain-servers list above
compu1
compu2

## Misc
mickey
minnie





