[R] convergence problem in boolean logit

[ckang2 at gmail.com]

Hello, folks

when I ran my boolean logit model in R, I got an error message as

> answer <- boolean (bp, link = "logit", method = "BFGS", data=pr2)
27061 observations dropped due to missing data.
Warning message:
fitted probabilities numerically 0 or 1 occurred in: glm.fit(x = X, y
= Y, weights = weights, start = start, etastart = etastart,

Could you anyone advise me how I can solve this problem?
I also tried with other method options ("method="nlm")

It is my first time to run boolean model, in addition, I'm a very
beginer user of R.

Thank you very much for your advice in advance.

Sincerely,
Charlie