[sysmon-help] What am I doing wrong ?

[Mark Pace Balzan]

Hi,

After some more head-scratching, and switching on debugs, it looks like
the objects monitored via ICMP are found to be down when the object
monitored via TCP is found down.  Now the ICMP objects are surely NOT
down, but the TCP object is really down.  So it looks like one is
trigerring the other.

The moment I remove the TCP monitored object, then all works flawlessly.
And other objects monitored via TCP which are surely UP and don't go
down are not falsely reported as down, so its seems to be related to the
combination of TCP and PING.


The debugs show:

Dec 21 2006-14:44:18 : service_this:Servicing entry in queue of
212.56.128.197:ping
Dec 21 2006-14:44:18 : icmp.c:service_test_ping:While pinging
212.56.128.197, packetsent = 7
Dec 21 2006-14:44:18 : icmp.c: 212.56.128.197 unpingable after 7
attempts
Dec 21 2006-14:44:18 : service_this:Servicing entry in queue of
212.56.128.99:ping
Dec 21 2006-14:44:18 : icmp.c:service_test_ping:While pinging
212.56.128.99, packetsent = 7
Dec 21 2006-14:44:18 : icmp.c: 212.56.128.99 unpingable after 7 attempts
Dec 21 2006-14:44:18 : service_this:Servicing entry in queue of
127.0.0.1:ping
Dec 21 2006-14:44:18 : icmp.c:service_test_ping:While pinging 127.0.0.1,
packetsent = 7
Dec 21 2006-14:44:18 : icmp.c: 127.0.0.1 unpingable after 7 attempts
Dec 21 2006-14:44:18 : service_this:Servicing entry in queue of
212.56.128.205:tcp
Dec 21 2006-14:44:18 : doing service_test_tcp() of 212.56.128.205:25



Anyone else having issues (or success for that matter) with a config
which uses both ICMP and TCP in the same config file ?


Thanks


Mark



> -----Original Message-----
> From: Noel Kelly [mailto:nkelly at citrusnetworks.net] 
> Sent: 21 December 2006 09:36
> To: Mark Pace Balzan
> Cc: sysmon-help at puck.nether.net
> Subject: Re: [sysmon-help] What am I doing wrong ?
> 
> Hi Mark,
> 
> I am no expert but it looks to me like your root dependency is not up 
> and so nothing else will be checked.
> 
> Can you ping 10.0.0.215 from the command line?  What I do is make my 
> root dependency localhost 127.0.0.1 to ensure that networking 
> is up and 
> available.  So you could do that and then make 'internal' a 
> dependency 
> on that and the other dependencies on 'internal' (as you have).
> 
> What does the log file say when you start sysmond?
> 
> HTH
> Noel
> 
> Mark Pace Balzan wrote:
> > Hi All,
> > 
> > 
> > Im having a hard time getting sysmon to work in a reliable fashion,
> > although it looks like other have had success.
> > 
> > Sysmon seems to be reporting a number of devices as not 
> reachable via
> > ping, when they really are up. 
> > 
> > Ive stripped down my config to rather basic now, but still 
> cant figure
> > out why this is happening.
> > 
> > Below is the output text file (I chose textfile output not 
> html in the
> > config), as well as the relevant config file options
> > 
> > 
> > Hopefully someone can shed some light
> > 
> > OS: FreeBSD tantalum.melitacable.com 5.5-RELEASE FreeBSD 
> 5.5-RELEASE #0:
> > Tue May 23 14:58:27 UTC 2006 
> > 
> > 
> > Thanks
> > 
> > Mark
> > 
> > 
> > 
> > Network Summary     sysmon v0.93-pre3
> > Hostname             Type  Port DownN UpN   NotifiedStat    
>        Time
> > Failed
> > 10.0.0.215           ping  0    0    43    No    up         
>     Dec 20
> > 20:42:39
> > 10.0.0.212           ping  0    0    18    No    up         
>     Dec 20
> > 22:50:29
> > 10.0.0.99            ping  0    0    18    No    up         
>     Dec 20
> > 22:50:29
> > 10.0.0.197           ping  0    0    18    No    up         
>     Dec 20
> > 22:50:29
> > 10.0.0.213           ping  0    0    18    No    up         
>     Dec 20
> > 22:50:29
> > 
> > 
> > Config:
> > 
> > root = "internal";
> > ;
> > #
> > # Comments can look like this
> > ; or this
> > 
> > # create a file for web page users
> > config statusfile text "/usr/local/share/serversstatus3.txt";
> > #
> > # show up and down for hosts/services
> > config showupalso;
> > #
> > # Log devices that are up/down
> > config logging file "/var/log/sysmon-servers3.log";
> > #
> > # Configure an authentication key -- Default is none
> > #config authkey "Debug";
> > #
> > # Set the sender to be sysmon at localhost
> > config sender "sysmon";
> > #
> > # Expire in-process dns cache this often
> > config dnsexpire 7200;
> > #
> > # Do not queue more than this many objects to be checked at once
> > config maxqueued 50;
> > #
> > # service frequency
> > config queuetime 300;
> > #
> > # how many times host is found down before email or sms is sent
> > config numfailures 2;
> > #
> > # send reminder email/sms for event ... value in minutes
> > config pageinterval 10;
> > 
> > # Not implemented yet
> > #config snmp-trap;              # catch snmp-traps
> > 
> > object internal {
> >         ip "10.0.0.215";
> >         type ping;
> >         desc "internal";
> > };
> > 
> > object srvneon {
> >         ip "10.0.0.212";
> >         type ping;
> >         desc "srvneon";
> >         dep "internal";
> > };
> > 
> > Same for other objects
> > 
> > 
> > 
> > _______________________________________________
> > Sysmon-help mailing list
> > Sysmon-help at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/sysmon-help
>