Grr - some things are out of sync with my code here.
Index: /home/dirkx/private/new-ab-xmass-2000/ab.c
===================================================================
RCS file: /home/cvs/apache-1.3/src/support/ab.c,v
retrieving revision 1.48
diff -r1.48 ab.c
900c900
< fprintf(stderr, "SSL fd init failed ")l
---
> fprintf(stderr, "SSL fd init failed ");
953c953
< if ((done) && (!(done % heartbeatres))) {
---
> if ((done) && (heartbeatres) && (!(done % heartbeatres))) {
On 10 May 2001, Jeff Trawick wrote:
> Joe Orton <[EMAIL PROTECTED]> writes:
>
> > Index: ab.c
> > ===================================================================
> > RCS file: /home/joe/lib/cvsroot/apache-1.3/src/support/ab.c,v
> > retrieving revision 1.49
> > diff -u -r1.49 ab.c
> > --- ab.c 11 Apr 2001 21:23:54 -0000 1.49
> > +++ ab.c 2 May 2001 10:01:33 -0000
> > @@ -897,7 +897,7 @@
> > };
> > SSL_set_connect_state(c->ssl);
> > if ((e = SSL_set_fd(c->ssl, c->fd)) == -1) {
> > - fprintf(stderr, "SSL fd init failed ")l
> > + fprintf(stderr, "SSL fd init failed ");
>
> yep :)
>
> > ERR_print_errors_fp(stderr);
> > goto bad;
> > };
> > @@ -1638,8 +1638,6 @@
> > heartbeatres = 100; /* but never more often than once every 100
> > * connections. */
> > }
> > - else
> > - heartbeatres = 0;
>
> Why? (but the variable heartbeatres is pretty goofy; it is set to 100
> at startup but is mostly a flag that says we want heartbeat messages
> until we then recalculate what we think it should be)
>
> Can't we just back up to the prior level of ab? It was a big
> distraction once the new function was committed to 2.0 to actually get
> it to compile properly and stop dumping core. Obviously nobody has
> had the time/inclination yet to do the same stuff with 1.3 ab.
>
> (nothing against your patch, Joe; I just wonder what is left untested)
>
>
