Date: Fri, 13 Jul 2007 22:37:56 +0200 From: "Frank Dittrich" <frank_dittrich@...mail.com> To: john-users@...ts.openwall.com Subject: Re: Strange bug when testing --format=ssha I wrote: > >Testing --format=krb5 on x86_64 also produces either >failing tests or segfaults, depending on which individual test cases I >comment out or activate. [...] >(I don't care whether or not this get fixed.) I still don't care. Nevertheless, here are some of the bugs. I ran valgrind --leak-check=yes ./john --test --format=krb5 on my core duo laptop. Valgrind discovered two memory leaks in the krb5 code. One of them in str2key (KRB5_std.c:254): text = (char*) malloc(strlen(user) + strlen(realm) + strlen(passwd)); This could be fixed by free(text); at the end of str2key. A better solution would probably be to use a static buffer which is large enough for the max. possible length. The other is less critical, because it leaks memory just once for each salt, in krb5_salt (KRB5_fmt.c:198) krb5_salt should be changed to use a static buffer, since it is impossible to free the memory if you have to return the salt to the caller. In addition to those two memory leaks, valgrind also reported a large number of these messages: Use of uninitialised value of size 4 at 0x40A1972: DES_encrypt2 (in /usr/lib/libcrypto.so.0.9.8) The address varies (from 0x40A1247 to 0x40A2064) This one could also be a false positive. I didn't bother to spend more effort on this. But there must be more bugs, because fixing the memory leaks doesn't make it work on my x86_64 system. At least it doesn't consume that much memory now. Frank _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ -- To unsubscribe, e-mail john-users-unsubscribe@...ts.openwall.com and reply to the automated confirmation request that will be sent to you.
Powered by blists - more mailing lists
Powered by Openwall GNU/*/Linux - Powered by OpenVZ