[exim-dev] [Bug 1773] New: GETSHORT reading 65535 instead of…

トップ ページ
このメッセージを削除
このメッセージに返信
著者: admin
日付:  
To: exim-dev
題目: [exim-dev] [Bug 1773] New: GETSHORT reading 65535 instead of 4 when running Exim test suite
https://bugs.exim.org/show_bug.cgi?id=1773

            Bug ID: 1773
           Summary: GETSHORT reading 65535 instead of 4 when running Exim
                    test suite
           Product: Exim
           Version: 4.86+ HEAD
          Hardware: x86-64
                OS: OpenBSD
            Status: NEW
          Severity: bug
          Priority: medium
         Component: Networking
          Assignee: nigel@???
          Reporter: km@???
                CC: exim-dev@???


While setting up a new animal for the buildfarm, I discovered that test 0020
segfaulted.

The issue was traced to some code at the and of dns_next_rr() in src/dns.c:

/* Move the pointer past the name and fill in the rest of the data structure
from the following bytes. */

dnss->aptr += namelen;
GETSHORT(dnss->srr.type, dnss->aptr); /* Record type */
dnss->aptr += 2;                      /* Don't want class */
GETLONG(dnss->srr.ttl, dnss->aptr);   /* TTL */
GETSHORT(dnss->srr.size, dnss->aptr); /* Size of data portion */
dnss->srr.data = dnss->aptr;          /* The record's data follows */
dnss->aptr += dnss->srr.size;         /* Advance to next RR */


Now, what happens is that in some circumstances dnss->srr.size is set to 65535
by GETSHORT (instead of 4) and this causes the process to segfault. Adding this
check after GETSHORT() works around the issue:

if (dnss->srr.size == 65535) dnss->srr.size = 4;

This happens when resolving manyhome.test.ex.

--
You are receiving this mail because:
You are on the CC list for the bug.