Andy, In the latest release (20050623-0016) I found a serious bug in src/ XrdOlb/XrdOlbConfig.cc which needs to be addressed fairly quickly, if you haven't already fixed it. I also found a pair of unnecessary casts that can be removed from src/ XrdNet/XrdNetDNS.cc in the process of figuring out the bug. BTW, I did the diffs backwards below. Sorry... The BUG: You go past the end of the InetAddr array when resolving the hostname of the master. The solution is to move the i--; line to before the reference to InetAddr[i]. I also threw in a warning message. There is probably a better way to log the warning, but I'll leave that to you. diff src/XrdOlb/XrdOlbConfig.cc ../xrootd.sol/src/XrdOlb/ 2033,2038c2033 < {i--; < char *mp = XrdNetDNS::getHostName(InetAddr[i]); < if (strcmp(mp, mbuff) != 0) { < eDest->Emsg("Config: WARNING reverse lookup mismatch: host =", mbuff, < " rev = ", mp); < } --- > {char *mp = XrdNetDNS::getHostName(InetAddr[i]); 2040a2036 > i--; The casts: ********** diff src/XrdNet/XrdNetDNS.cc ../xrootd.sol/src/XrdNet/XrdNetDNS.cc 467c467 < sz = snprintf(buff,blen,"%s", inet_ntoa(in)); --- > sz = snprintf(buff,blen,"%s", inet_ntoa((const struct in_addr)in)); 469c469 < sz = snprintf(buff,blen,"%s:%d",inet_ntoa(in),port); --- > sz = snprintf(buff,blen,"%s:%d",inet_ntoa((const struct in_addr)in),port); -- Gregory J. Sharp email: [log in to unmask] Wilson Synchrotron Laboratory url: http://www.lepp.cornell.edu/ ~gregor Dryden Rd ph: +1 607 255 4882 Ithaca, NY 14853 fax: +1 607 255 8062