I'm currently running nntpcached 1.0.3 under bsdi and i'm *STILL* having this problem. Both with trn 3.6 and trn 4.0-test54. I've check nntpcached and 1.0.3 has this line patched. Any ideas? > Hi, > I've just been trying out nntpcache on a Solaris 2.5 system and it seems to > be working well except that trn would sometimes hang after displaying part of > a message for some messages. > > I traced it down to the following bug in the handling of the body command for > articles that have short bodies and are in the cache. > > *** article.c.orig Tue Oct 15 21:43:05 1996 > --- article.c Wed Feb 5 08:49:07 1997 > *************** > *** 811,817 **** > fflush(clientout); /* keep fd and fh in-sync */ > write (fileno(clientout), body, buf+len-body); /* avoid buffering, push > it out in one hit */ > } else > ! fwrite (body, body-buf, 1, clientout); > break; > case c_stat: > break; > --- 811,817 ---- > fflush(clientout); /* keep fd and fh in-sync */ > write (fileno(clientout), body, buf+len-body); /* avoid buffering, push > it out in one hit */ > } else > ! fwrite (body, buf+len-body, 1, clientout); > break; > case c_stat: > break;
