bind-9.10.0-P2/bind-utilities-9.10.0-P2

[Fernando de Oliveira]

​ftp://ftp.isc.org/isc/bind9/9.10.0-P2/bind-9.10.0-P2.tar.gz

​ftp://ftp.isc.org/isc/bind9/9.10.0-P2/RELEASE-NOTES-BIND-9.10.0-P2.txt

Introduction

   BIND 9.10.0-P2 is a security fix release of BIND 9.10 which also
   includes changes to address GCC optimization issues described in
   ISC Operational Notification https://kb.isc.org/article/AA-01167.

   This document summarizes features added or significantly changed
   since the previous major release, BIND 9.9. Bug fixes since BIND
   9.10.0 are also summarized. Changes marked with '**' have been added
   since the previous release (BIND 9.10.0-P1). Please see the CHANGES
   file in the source code release for a complete list of all changes.
...
Security

   A query specially crafted to exploit a defect in EDNS option
   processing can cause named to terminate with an assertion failure.
   This fixes a missing isc_buffer_availablelength check when
   printing out a packet.  [CVE-2014-3859] [RT #36078] **

   A programming error in the prefetch feature could cause named
   to crash with a "REQUIRE" assertion failure in name.c [CVE-2014-3214]
   [RT #35899]
...

[Fernando de Oliveira]

I have never done this. Will try and give back, if I fail to get it working.

[Fernando de Oliveira]

I don't understand this:

[ $(uname -m) = x86_64 ] && ln -sv lib usr/lib64

Shouldn't be:

[ $(uname -m) = x86_64 ] && ln -sv lib /usr/lib64

?

[bdubbs@…]

No, what's there is correct.

cd /srv/named 
mkdir -p usr/lib/engines 
...
[ $(uname -m) = x86_64 ] && ln -sv lib usr/lib64

We are creating /srv/named/usr/lib64 -> lib, but only for x86_64 systems.

[Fernando de Oliveira]

Thanks.

I have 4 tests failures, 60 passes and 4 skips.

Failures: resolver, rpz, statistics and xfer (this was already in the book). So, will include three lines in the sed, but modify the text: "... some tests fail for unknown reasons."

Will include a sentence with a command to find which test failed.

[Fernando de Oliveira]

I didn't like the results of the tests and could not find anything about the new failures.

Also, Cannot understand why I had 68 tests, while Bruce had 185.

Will consider as fixed at r13235 and close, but if Bruce or somebody wants to investigate further the tests, please do, I will try to learn better, then.

[Fernando de Oliveira]

Correction: forgot to run

grep "R:PASS" check.log | wc -l

Result is 191. Thus, it is greater than the 185 for Bruce.

I cannot remeber anymore why, but I knew that the number of tests had increased, so the failures I was getting were new. Forgot about that and reado only the tail of the log, which was giving me the much smaller number of tests.

I am happier, now.