RHBA-2015:2354 - Bug Fix Advisory

Topic

Updated corosync packages that fix multiple bugs and add one enhancement are now available for Red Hat Enterprise Linux 7.

Description

The corosync packages provide the Corosync Cluster Engine and C APIs for Red Hat Enterprise Linux cluster software.

This update fixes the following bugs:

• In earlier versions of Corosync, realtime (RT) scheduling priority was not enabled by default. This had the potential to increase the possibility of token losses when systems were under heavy CPU load. This change enables RT scheduling for Corosync by default, which reduces the chance of token losses. (BZ#1170347)
• In the votequorum subsystem of Corosync, the auto_tie_breaker_node option was implemented in such a way that the option was incorrectly required to be present for the auto_tie_breaker option to function. As a consequence, older systems with auto_tie_breaker enabled had the auto_tie_breaker_node option disabled when the systems were upgraded to a newer version, and new systems with auto_tie_breaker enabled did not use the "Auto Tie Breaker" algorithm. This update ensures that auto_tie_breaker_node is optional, which makes auto_tie_breaker independent of auto_tie_breaker_node and also compatible with earlier versions. (BZ#1197091)
• Previously, Corosync allowed the two_node and auto_tie_breaker options to be present in the corosync.conf file at the same time. As a consequence, if both options were included, auto_tie_breaker was silently ignored and the two_node fence race decided which node would survive in the event of a communication break. With this update, if both two_node and auto_tie_breaker are detected in corosync.conf, Corosync issues a message at startup and disables two_node mode. In addition, the pcs daemon has been fixed so that it does not produce corosync.conf files with both options included. As a result, auto_tie_breaker effectively overrides two_node mode if both options are specified. (BZ#1197671)
• Prior to this update, in clusters with an odd number of nodes that had the auto_tie_breaker option enabled, when one of the nodes failed, the remaining nodes were split 50:50. Consequently, auto_tie_breaker was not invoked, and a random half of the cluster was fenced, rather than the half that did not contain the tie breaker node. With this update, the wait_for_all option is required for clusters with an odd number of nodes. As a result, the cluster half that does not contain the tie breaker node is now fenced in the described scenario. (BZ#1229194)

In addition, this update adds the following enhancement:

• The maximum size of a message that could be transferred using Corosync Closed Process Groups (CPG) messaging facility was previously limited to 1 MB. This limit has now been removed, allowing the Pacemaker resource manager to send data packets of unlimited size to communicate changes in the cluster state. To enable this change, adjust the Pacemaker configuration file located in the /etc/sysconfig/ directory. (BZ#682771)

Users of corosync are advised to upgrade to these updated packages, which fix these bugs and add this enhancement.

Solution

Before applying this update, make sure all previously released errata
relevant to your system have been applied.

For details on how to apply this update, refer to:

https://access.redhat.com/articles/11258

Affected Products

• Red Hat Enterprise Linux High Availability for x86_64 7 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Update Support 7.7 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Update Support 7.6 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Update Support 7.5 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Update Support 7.4 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Update Support 7.3 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 7 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Update Support 7.7 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Update Support 7.6 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Update Support 7.5 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Update Support 7.4 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Update Support 7.3 x86_64
• Red Hat Enterprise Linux High Availability (for RHEL Server) from RHUI 7 x86_64
• Red Hat Enterprise Linux Resilient Storage (for RHEL Server) from RHUI 7 x86_64
• Red Hat Enterprise Linux Resilient Storage for IBM z Systems 7 s390x
• Red Hat Enterprise Linux High Availability for IBM z Systems 7 s390x
• Red Hat Enterprise Linux High Availability for x86_64 - Update Services for SAP Solutions 7.7 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Update Services for SAP Solutions 7.6 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Update Services for SAP Solutions 7.4 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Update Services for SAP Solutions 7.3 x86_64
• Red Hat Enterprise Linux High Availability for x86_64 - Extended Life Cycle Support 7 x86_64
• Red Hat Enterprise Linux Resilient Storage for x86_64 - Extended Life Cycle Support 7 x86_64

Fixes

• BZ - 682771 - RFE: remove 1M message size limit
• BZ - 1225441 - corosync does not rotate its logs in /var/log/cluster
• BZ - 1226842 - RRP man page
• BZ - 1229194 - auto_tie_breaker can create two quorate clusters
• BZ - 1234261 - RRP Active: Reset timer_problem_decrementer on fault
• BZ - 1234266 - Use safer permission when storing ringid

CVEs

(none)

References

• http://www.redhat.com/security/updates/classification/#normal