Synopsis
Low: pacemaker security, bug fix, and enhancement update
Type/Severity
Security Advisory: Low
Topic
Updated pacemaker packages that fix one security issue, several bugs, and
add various enhancements are now available for Red Hat Enterprise Linux 6.
The Red Hat Security Response Team has rated this update as having low
security impact. A Common Vulnerability Scoring System (CVSS) base score,
which gives a detailed severity rating, is available from the CVE link in
the References section.
Description
Pacemaker is a high-availability cluster resource manager with a powerful
policy engine.
A denial of service flaw was found in the way Pacemaker performed
authentication and processing of remote connections in certain
circumstances. When Pacemaker was configured to allow remote Cluster
Information Base (CIB) configuration or resource management, a remote
attacker could use this flaw to cause Pacemaker to block indefinitely
(preventing it from serving other requests). (CVE-2013-0281)
Note: The default Pacemaker configuration in Red Hat Enterprise Linux 6 has
the remote CIB management functionality disabled.
The pacemaker package has been upgraded to upstream version 1.1.10, which
provides a number of bug fixes and enhancements over the previous version:
- Pacemaker no longer assumes unknown cman nodes are safely stopped.
- The core dump file now converts all exit codes into positive 'errno'
values.
- Pacemaker ensures a return to a stable state after too many fencing
failures, and initiates a shutdown if a node claimed to be fenced is still
active.
- The crm_error tool adds the ability to list and print error symbols.
- The crm_resource command allows individual resources to be reprobed, and
implements the "--ban" option for moving resources away from nodes.
The "--clear" option has replaced the "--unmove" option. Also, crm_resource
now supports OCF tracing when using the "--force" option.
- The IPC mechanism restores the ability for members of the haclient group
to connect to the cluster.
- The Policy Engine daemon allows active nodes in the current membership to
be fenced without quorum.
- Policy Engine now suppresses meaningless IDs when displaying anonymous
clone status, supports maintenance mode for a single node, and correctly
handles the recovered resources before they are operated on.
- XML configuration files are now checked for non-printing characters and
replaced with their octal equivalent when exporting XML text. Also, a more
reliable buffer allocation strategy has been implemented to prevent
lockups.
(BZ#987355)
Additional bug fixes:
- The "crm_resource --move" command was designed for atomic resources and
could not handle resources on clones, masters, or slaves present on
multiple nodes. Consequently, crm_resource could not obtain enough
information to move a resource and did not perform any action. The "--ban"
and "--clear" options have been added to allow the administrator to
instruct the cluster unambiguously. Clone, master, and slave resources can
now be navigated within the cluster as expected. (BZ#902407)
- The hacluster user account did not have a user identification (UID) or
group identification (GID) number reserved on the system. Thus, UID and GID
values were picked randomly during the installation process. The UID and
GID number 189 was reserved for hacluster and is now used consistently for
all installations. (BZ#908450)
- Certain clusters used node host names that did not match the output of
the "uname -n" command. Thus, the default node name used by the crm_standby
and crm_failcount commands was incorrect and caused the cluster to ignore
the update by the administrator. The crm_node command is now used instead
of the uname utility in helper scripts. As a result, the cluster behaves as
expected. (BZ#913093)
- Due to incorrect return code handling, internal recovery logic of the
crm_mon utility was not executed when a configuration updated failed to
apply, leading to an assertion failure. Return codes are now checked
correctly, and the recovery of an expected error state is now handled
transparently. (BZ#951371)
- cman's automatic unfencing feature failed when combined with Pacemaker.
Support for automated unfencing in Pacemaker has been added, and the
unwanted behavior no longer occurs. (BZ#996850)
All pacemaker users are advised to upgrade to these updated packages, which
contain backported patches to correct these issues and add these
enhancements.
Solution
Before applying this update, make sure all previously-released errata
relevant to your system have been applied.
This update is available via the Red Hat Network. Details on how to
use the Red Hat Network to apply this update are available at
https://access.redhat.com/site/articles/11258
Affected Products
-
Red Hat Enterprise Linux Server 6 x86_64
-
Red Hat Enterprise Linux Server 6 i386
-
Red Hat Enterprise Linux High Availability (for RHEL Server) 6 x86_64
-
Red Hat Enterprise Linux High Availability (for RHEL Server) 6 i386
-
Red Hat Enterprise Linux Resilient Storage (for RHEL Server) 6 x86_64
-
Red Hat Enterprise Linux Resilient Storage (for RHEL Server) 6 i386
-
Red Hat Enterprise Linux Server from RHUI 6 x86_64
-
Red Hat Enterprise Linux Server from RHUI 6 i386
-
Red Hat Enterprise Linux High Availability (for RHEL Server) from RHUI 6 x86_64
-
Red Hat Enterprise Linux High Availability (for RHEL Server) from RHUI 6 i386
-
Red Hat Enterprise Linux Resilient Storage (for RHEL Server) from RHUI 6 x86_64
-
Red Hat Enterprise Linux Resilient Storage (for RHEL Server) from RHUI 6 i386
Fixes
- BZ - 891922 - CVE-2013-0281 pacemaker: remote DoS when CIB management is enabled caused by use of blocking sockets
- BZ - 902407 - Different results when moving Master/Slave resources
- BZ - 902459 - Persistent resource-related data after the resource is deleted
- BZ - 996850 - Unfence at cluster startup with fence_scsi
- BZ - 997346 - pacemaker enables itself on boot during installation
- BZ - 1011618 - Slave roles inconsistent in pcs status xml (or crm_mon)
CVEs
References
Vulmon