Re: [hylafax-users] what is an acceptable rate of failure ?

["George H" <george.dma@xxxxxxxxx>]

I did notice that there were entries in xferfaxlog that were not
showing up in the recvstats. For example if I just do a simple query
to retrieve all errors for a particular TSI (SELECT * FROM xferfaxlog
WHERE REASON NOT IN('','""') AND MODEM='ttyS5' AND TSI='123456789')

I get more rows describing errors than shown in recvstats. I also
notice that recvstats shows errors that occur when the ENTRYTYPE is
RECV (where it will be logged in the database twice once as CALL and
once as RECV). I do see also errors where the ENTRYTYPE is just CALL,
those ones I don't see in recvstats.

So what I did was run some queries that got all errors for all calls.
I exclude error E102. And I just create the percentage from total
errors occur with the total calls received. So that's how I got 86.46%
success rate or rather that's the percentage of calls that did not
result in an error in xferfaxlog (excluding E102 errors).

After reading your email I started to understand why you counted on a
per-page basis. I assume this is the proper way to perform statistics
on faxes.

So can I assume it is safe to use recvstats to justify a success/error
rate on my faxserver? Even though we sometimes have certain people who
just can't manage to send a fax to us (those cases are low as well and
usually their type/rate is between 2400 and 7600).

When I run recvstats I have this summary

Numpages: 53311
Time: 3:42:04
Pg/Min: 1.8
Errs: 1524

I tend to see most (not all) errors occuring on senders who's
type/rate is usually under 14400. I have senders sending me
at 2400 1-D MH to 14400 2-D MMR and rarely I get JBIG.

So from the hylafax stats above my error rate is (1524/53311)*100 = 2.85%
Did I do this right?

Thanks for helping me out.

-- 
"Nothing is impossible for the person that doesn't have to do it"
"The probability of anything happening is in inverse ratio to its desirability"
--
George H
george.dma@xxxxxxxxx


____________________ HylaFAX(tm) Users Mailing List _______________________
  To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
 On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null
  *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*