$FreeBSD: src/release/doc/en_US.ISO8859-1/errata/article.sgml,v
-1.73.2.27 2006/06/17 04:31:40 bmah Exp $
+1.73.2.29 2006/08/30 18:03:44 bmah Exp $
Inoperative access controls in +href="http://www.FreeBSD.org/cgi/man.cgi?query=ypserv&sektion=8&manpath=FreeBSD+6.0-stable"> ypserv(8)
Incorrect multipart message handling in Sendmail
+ +Buffer overflow in +sppp(4)
+(2006/05/13) A change in the -ata(4) driver in +href="http://www.FreeBSD.org/cgi/man.cgi?query=ata&sektion=4&manpath=FreeBSD+6.0-stable">ata(4) driver in FreeBSD 6.1-RELEASE modified a software interface visible from userland programs. Programs that use the IOCATAREQUEST +href="http://www.FreeBSD.org/cgi/man.cgi?query=ioctl&sektion=2&manpath=FreeBSD+6.0-stable"> ioctl(2) interface will need to be recompiled. Typically, these programs communicate directly with ATA and ATAPI drives, such as CDROM burning or ripping tools, or monitoring utilities. @@ -177,21 +188,21 @@ href="http://www.FreeBSD.org/cgi/url.cgi?ports/sysutils/smartmontools/pkg-descr" class="FILENAME">sysutils/smartmontools port.
(2006/05/27, updated 2006/06/01) On FreeBSD/alpha, +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) sometimes does not allow a disk to be labeled. The disk is presented in +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) but actions like Auto do not result in a disklabel being presented. One workaround is to boot the install CD, then select Fixit. From the Fixit# prompt, run +href="http://www.FreeBSD.org/cgi/man.cgi?query=disklabel&sektion=8&manpath=FreeBSD+6.0-stable"> disklabel(8) manually. For example: disklabel -rw da1 auto ensures a valid disklabel is written onto da1. Reboot from the install media. +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) will now detect a valid disklabel on the disk and allow it to be partitioned to your preference.
@@ -202,49 +213,55 @@ source-level compatibility with externally-maintained 10Gb Ethernet network driv written for FreeBSD 6.0-RELEASE. This change should have been documented in the release notes. -(2006/06/01, updated 2006/06/07) A bug in the rc.d/jail +
(2006/06/01, updated 2006/07/07) A bug in the rc.d/jail startup script could cause various problems for users attempting to configure +href="http://www.FreeBSD.org/cgi/man.cgi?query=jail&sektion=2&manpath=FreeBSD+6.0-stable"> jail(2) -environments at system startup time. A fix has been committed to HEAD and RELENG_6; it -will be merged to the RELENG_6_1 errata fix branch after more testing.
+environments at system startup time. A fix has been committed to HEAD, RELENG_6, and +RELENG_6_1. More details can be found in errata notice FreeBSD-EN-06:01.jail. -(2006/06/01) Deadlocks have been observed (and can be reproduced) in some cases where -snapshots and quotas are both being used on a file system. This problem has been fixed on -the HEAD and RELENG_6 branches; more testing is planned before a merge to the RELENG_6_1 -errata fix branch.
+(2006/06/01, updated 2006/08/30) Deadlocks have been observed (and can be reproduced) +in some cases where snapshots and quotas are both being used on a file system. This +problem has been fixed on the HEAD and RELENG_6 branches.
-(2006/06/01, updated 2006/06/16) A bug in the Neighbor Discovery Protocol (NDP) +
(2006/06/01, updated 2006/08/28) A bug in the Neighbor Discovery Protocol (NDP) implementation causes spurious error messages for point-to-point IPv6 links on FreeBSD 6.1-RELEASE. This behavior is a regression from FreeBSD 6.0-RELEASE. A solution for this -problem has been committed to HEAD and is planned for merging to RELENG_6.
+problem has been committed to HEAD, RELENG_6, and RELENG_6_1. Further details can be +found in errata notice FreeBSD-EN-06:02.net.(2006/06/01) Problems have been observed when trying to load packages from multiple CDROMs inside the +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) environment. Specifically, the disk-switching functionality in +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) seems to become confused about the contents of a disk that has just been requested. This situation has been seen primarily when installing the base system without Xorg (which lives on the first CDROM of a multi-disk set), and then later using +href="http://www.FreeBSD.org/cgi/man.cgi?query=sysinstall&sektion=8&manpath=FreeBSD+6.0-stable"> sysinstall(8) to load some packages (such as the GNOME or KDE desktop environments) from a subsequent CDROM that depend on Xorg. As a workaround during a new install, one should try to install Xorg along with the base system, or at least before any other packages that depend on it.
-(2006/06/01, updated 2006/06/07) An error in the statistics-keeping logic in the UMA +
(2006/06/01, updated 2006/08/28) An error in the statistics-keeping logic in the UMA kernel memory allocator has led to the number of requests for mbufs denied (as reported by netstat -m) to increase erroneously. -A fix for this problem has been committed to HEAD and RELENG_6 and will be merged to the -RELENG_6_1 branch after additional testing.
+A fix for this problem has been committed to HEAD, RELENG_6, and RELENG_6_1. More +information is contained in errata notice FreeBSD-EN-06:02.net.(2006/06/01) A longstanding bug in the i386 bootloader can cause root file system corruption when the +href="http://www.FreeBSD.org/cgi/man.cgi?query=nextboot&sektion=8&manpath=FreeBSD+6.0-stable"> nextboot(8) utility is used and the /boot/nextboot.conf configuration file happens to be located after cylinder 1023 on the boot disk. A bugfix has been committed @@ -253,6 +270,12 @@ to HEAD and planned for merging after additional testing.
(2006/06/16) A locking-related bug has been observed to cause noticeable slowdowns in NFS server performance. A fix has been committed to HEAD and RELENG_6, and is a candidate for a merge to RELENG_6_1 after additional testing.
+ +(2006/08/28) A local kernel panic in the network stack could be caused by not checking +the validity of a pointer before use. This problem has been fixed on HEAD, RELENG_6, and +RELENG_6_1, as described in FreeBSD-EN-06:02.net.