Bug Fix Advisory rgmanager bug fix and enhancement update

Advisory: RHBA-2011:0134-1
Type: Bug Fix Advisory
Severity: N/A
Issued on: 2011-01-13
Last updated on: 2011-01-13
Affected Products: RHEL Clustering (v. 5 server)


Updated rgmanager packages that provide a fix for a bug and add an enhancement
are now available for Red Hat Enterprise Linux 5.

The rgmanager packages contain the Red Hat Resource Group Manager, which
provides the ability to create and manage high-availability server applications
in the event of system downtime.

This update addresses the following bugs:

* HA-LVM now resolves volume group names correctly (BZ#506587)

* HA-LVM can now protect from metadata changes by utilizing CLVM; logging has
also been improved (BZ#509368 BZ#572311 BZ#583769 BZ#585229)

* The NetBIOS name stored in the automatically generated smb.conf files for
services now has spaces removed (BZ#531098)

* It is now possible to stop rgmanager before CMAN finishes starting (BZ#545229)

* Failover domains are now documented in the rgmanager.8 man page (BZ#557562)

* SAPInstance and SAPDatabase resource agents no longer leave extraneous
temporary files (BZ#565363)

* Rgmanager now retries the operation if it tries to relocate a service to
another node before that node has reread the cluster configuration file

* fs.sh no longer sleeps unnecessarily if the file system is already unmounted
during a stop operation (BZ#573705)

* Postgres-8 agent will now stop correctly if there are active clients connected

* Restricted failover domain boundaries are now honored when performing virtual
machine migrations (BZ#592380)

* Temporary files created when using NFS lock code are now deleted (BZ#595455)

* Live migrations now use the correct interfaces when using migration mappings

* Errors from tomcat.sh have been addressed (BZ#591003)

* Previously, clusters failed to recover from a clurgmgrd crash if the reboot()
system call failed. With this update, the rgmanager watchdog process now uses a
more robust method to reboot the machine if the main rgmanager process crashes.

* clustat no longer returns 255 if rgmanager is not running (BZ#620730)

* If openais dies, rgmanager now halts services and exits (BZ#639961)

* Handling of CIFS mount points mounted from the netfs agents has been improved

* User-specified migration URIs work correctly with the fix for BZ#569016

* The postgres-8 resource agent now correctly places IP netmask information into
the configuration files it generates (BZ#614456)

This update also contains the following enhancements:

* Previously, vm.sh only checked the status of the VM itself, not the status of
any services inside. With this update, administrators may now use a newly
provided status check program which checks the availability of services within
virtual machines running Red Hat Enterprise Virtualization Manager. Timeouts for
starting and stopping virtual machines are now configurable in cluster.conf. The
start timeout is based on the status check program. (BZ#583788)

* Independent subtrees may now be flagged as 'non-critical', meaning they may
fail and have their components manually restarted without the entire service
being affected (BZ#605733)

* The file system agent will now accept 'vxfs' as the file system type

* The file system agent will now accept 'ext4' as the file system type

* Users of Oracle database may now separate the listener from the database
instance and have multiples of each resource type in a single cluster. This
feature is offered as a Technical Preview and is not supported in production
environments. (BZ#629208)

All users of Red Hat Resource Group Manager are advised to upgrade to these
updated packages, which address these issues and add these enhancements.


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

Updated packages

RHEL Clustering (v. 5 server)

File outdated by:  RHBA-2014:1207
    MD5: ee2c8e4e6a1a1e6eb975c2be3f91305b
SHA-256: e19aeec319c7a2e388b7ac34fed4164f87b4fd2c0465184861c2d9e4442b84b3
File outdated by:  RHBA-2014:1207
    MD5: adee0fb1aa8e709295853c9d481fdf90
SHA-256: 277d425bad173eaf1e7230cf04e3909f436abe9e18fe121612d71b1507530283
File outdated by:  RHBA-2014:1207
    MD5: 4af2d9f2ae77272347af628b9ce4776d
SHA-256: 3ee988da6a7f37db56bb02868147009da9d2b07d339a98c0202d2a6c7ebeaf6e
File outdated by:  RHBA-2014:1207
    MD5: 330a95c77f77c652956c5d4178551db4
SHA-256: a8afb7e0b8538d79b0ccae284d45d81af3acf1145fdad2f37f249f70b0f20e28
File outdated by:  RHBA-2014:1207
    MD5: 96c22f2990be8ed919ee4d994faabfde
SHA-256: 8b19f4c7eef23ab4a74cd3c7037285726159b638dc5de9ea9f162b27b5a457f9
(The unlinked packages above are only available from the Red Hat Network)

Bugs fixed (see bugzilla for more information)

506587 - lvm agent incorrectly reports vg is in volume_list
531098 - samba.sh set wrong netbios name in smb.conf
531843 - fs.sh fsck on VXFS fails, needs fsck_needed="" or invoke without -p
545229 - rgmanager will hang on start if cman is not running and will not stop
557562 - failoverdomains are not documented in any man page
565363 - rgmanager creates /tmp/SAPDatabase.tmp.XXXXXX files eventhough there is no SAPDatabse resources in cluster
568126 - service configuration update occurred out of sync
573705 - fs.sh stop unnecessary waits 5 seconds when the file system is already unmounted
583769 - Update lvm.sh halvm error message when lvm.conf is newer than /boot/*.img files
583788 - status check program for vm.sh & user-controlled error tolerance
587735 - postgres-8.sh will fail to stop gracefully if there is active clients connected
591003 - Errors in resource script tomcat-5.sh, part of rgmanager
592380 - Restricted failover domains don't work with VM migrations
595455 - rgmanager does not remove nfs service temp directories when using nfslock=1
596016 - Live Migration of KVMs uses wrong interface in cluster
605733 - RFE: Critical/Non-Critical services & resources
608397 - rgmanager: fail to recover from clurgmgrd crash
614456 - Resource agent for psql does not work correctly with netmask
620730 - clustat returns 255 instead of 0
636550 - Support ext4 file system in fs.sh
639961 - Failover of service fails if non power fencing is used with qdisk
640676 - netfs.sh failed to retrieve CIFS mountpoint status
659477 - migrate_uri support broken in 5.6 beta

These packages are GPG signed by Red Hat for security. Our key and details on how to verify the signature are available from:

The Red Hat security contact is secalert@redhat.com. More contact details at http://www.redhat.com/security/team/contact/