[Dxspider-support] DxSpider 1.42 crash

Dirk Koopman djk at tobit.co.uk
Wed Jun 21 10:29:12 BST 2000


On Tue, 20 Jun 2000, you wrote:
> Just a few minutes ago, DxSpider crashed and stopped with the
> execution of cluster.pl
> 
> It was on this spot:
> 
> <- I PA4AB-7 PC11^21275.0^CE6TBN^20-Jun-2000^2011Z^ ^OK2WED^OK0DXC^H95^~
> 
> Illegal division by zero at /spider/perl/Spot.pm line 212.
> at /spider/perl/DXProt.pm line 231
> 

Humm...

This PC11 arrived safely in the UK (I have checked), the line it banged out
on says:- 

	$freq = sprintf "%.1f", $freq;       # normalise frequency
->	$d /= 60;                                 # to the nearest minute
	chomp $text;

I cannot, for the life of me, see how you would get a divide by zero error -
ever - never mind on the perfectly reasonable date that is in the spot (the
date is converted by library routine into unix time_t seconds - it would
never get to this piece of code if it wasn't within -60/+15 minutes of "now").

If you are experiencing frequent stoppages I wonder whether you could
run the cluster in debug mode when you are about? In theory I have caused all
the debug messages to go to the debug log, but it appears that I still have
failed to do this successfully. Running it in debug mode will give me more
information.

To do that:-

1. disable any automatic restarting of cluster.pl you may have (eg in
/etc/inittab [don't forget to telinit q after editting the file]).
2. cd /spider/perl
3. perl -d cluster.pl
4. r

wait....

If it fails, please cut and paste the screen to an email and send it to me.

Is this a sparc problem? Has anyone else with a sparc had frequent failures
with 1.42?

Dirk
-- 
Dirk-Jan Koopman, Tobit Computer Co Ltd 
At the source of every error which is blamed on the computer you will find
at least two human errors, including the error of blaming it on the computer.







More information about the Dxspider-support mailing list