Peter Maydell wrote:
>Peter Maydell wrote:
>>Anyway, for SASL, I wonder if this is the same problem as mentioned
>>here for mutt:
>>
>>http://osdir.com/ml/security.cyrus.sasl/2006-09/msg00010.html
>
>I set up a spare machine with an SMTP server with SASL enabled so
>I could test this, and reproduced the problem. In fact it turns out
>to be something different: sasl_decode64() could return SASL_CONTINUE
>as well as SASL_OK on success. I've fixed this in CVS (and will put it
>in the 1.3 release too). The patch below may or may not apply on
>the very elderly 1.2 the original submitter has.
...only of course I forgot to include the patch...
===begin===
Index: mts/smtp/smtp.c
===================================================================
RCS file: /cvsroot/nmh/nmh/mts/smtp/smtp.c,v
retrieving revision 1.19
retrieving revision 1.20
diff -u -r1.19 -r1.20
--- mts/smtp/smtp.c 13 Apr 2007 11:53:08 -0000 1.19
+++ mts/smtp/smtp.c 29 Apr 2008 17:04:38 -0000 1.20
@@ -1246,7 +1246,7 @@
result = sasl_decode64(sm_reply.text, sm_reply.length,
outbuf, sizeof(outbuf), &outlen);
- if (result != SASL_OK) {
+ if (result != SASL_OK && result != SASL_CONTINUE) {
smtalk(SM_AUTH, "*");
sm_ierror("SASL base64 decode failed: %s",
sasl_errstring(result, NULL, NULL));
===endit===
-- PMM
_______________________________________________
Nmh-workers mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/nmh-workers
