fit_counts error

I get this error during a call to fit_counts() (shown with Fit_Verbose=1). 
Using ISIS 1.3.0.  This is part of a set of automated fit routines, so I'd like
to be able to predict that this error is coming and circumvent it before it
halts the fit (thus halting a series of fits that may have been going on for a
long time).

Have I done something wrong?  Is there a way to predict the error will occur?

Thanks,
Rob

---------------------------------

... [stuff] ...
chisqr= 6.0208e+01:
 voigt(4).norm=  3.07185834e+07 voigt(4).energy=  4.48790498e-06
 voigt(4).fwhm=  1.57040638e-01 voigt(4).vtherm=  4.62048978e+03

chisqr= 6.0208e+01:
 voigt(4).norm=  3.07185834e+07 voigt(4).energy=  4.48342156e-06
 voigt(4).fwhm=  1.57197678e-01 voigt(4).vtherm=  4.62048978e+03

chisqr= 6.0208e+01:
 voigt(4).norm=  3.07185834e+07 voigt(4).energy=  4.48342156e-06
 voigt(4).fwhm=  1.57040638e-01 voigt(4).vtherm=  4.62511027e+03

*** Error:  function evaluation failed
S-Lang Error: Intrinsic Error: Error while executing __isis_tmp_ffname__
lmdif1: algorithm estimates that the relative error in the sum of squares is at
most tol.
 Parameters[Variable] = 23[4]
            Data bins = 35
           Chi-square = 40.32
   Reduced chi-square = 1.301

----
You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
unsubscribe