RHBA-2014:0020 - Bug Fix Advisory

Topic

OpenShift Enterprise by Red Hat release 2.0.1 is now available with updates to
packages that fix several bugs and introduce feature enhancements.

Description

OpenShift Enterprise by Red Hat is the company's cloud computing
Platform-as-a-Service (PaaS) solution designed for on-premise or private cloud
deployments.

This update introduces an upgrade path from OpenShift Enterprise 1.2 to the
latest release of OpenShift Enterprise 2.0. The OpenShift Enterprise 2.0
Deployment Guide at https://access.redhat.com/site/documentation now includes a
chapter that describes available upgrade paths using the ose-upgrade tool. Note
that the ose-upgrade tool procedures are separate from the procedure required to
apply this 2.0.1 update to a previously installed 2.0 deployment; for the latter
procedure, see the Solution section of this advisory.

This update addresses the following bug fixes and enhancements on broker hosts:

• Recent changes to the default location of MCollective log files resulted in

the oo-diagnostics tool providing inaccurate suggested fixes for related file
permission problems. The tool now provides accurate suggestions based on the
location of the file causing the problem. (BZ#1046202)

• The oo-diagnostics tool previously issued a stack trace when interrupted by a

CTRL+C key combination; the tool now exits cleanly. The tool also now reports
when the configured cloud domain does not resolve correctly. This enhancement
helps identify issues related to configured cloud domains that are not being
handled by a DNS server. (BZ#1034517)

• Previously, the Applications page of the Management Console displayed the

overall gear usage of a user account multiple times for each domain owned by
that user. This was potentially confusing for users because gear usage is
calculated across all domains created on a user account. Gear usage for a user
account now appears only once on the Applications page to avoid confusion.
(BZ#1038523)

• Previously, when adding a community cartridge to a scaled application, the

mouse hover text for the gear cost icon was potentially inaccurate. The text is
now more distinct about the expected number or range of gears that the cartridge
will add. (BZ#1032982)

This update also addresses the following bug fixes on node hosts:

• If an application was deployed using the artifact_url method, and the

application deployment method was later changed to git, a subsequent git push
would fail. The deployment process has been updated and these failures no longer
occur as a result of changing the deployment method. (BZ#1033113)

• Previously, restoring an application that contained a PostgreSQL cartridge

would succeed, but it would display the message "ERROR: database <db_name>
already exists". The restoration process has been updated and the message no
longer appears when restoring PostgreSQL databases that retain the original
database name. (BZ#1035237)

• The Display-Name field in the JBoss EAP cartridge manifest has been updated to

be more accurate in consideration of future JBoss EAP 6 releases. (BZ#1038796)

• Previously, it was possible for scaled applications to have a number of

deployment directories on its gears that totaled greater than the configured
deployment record limit. This was due to the deployment distribution process
creating new deployment directories on child gears without pruning older
directories. As a result, a gear could also reach its file quota limit
prematurely. The distribution process has been updated and gears now retain the
correct number of deployment directories. (BZ#1038753)

All OpenShift Enterprise by Red Hat users are advised to upgrade to these
updated packages which fix these bugs and add these enhancements.

Solution

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

To apply this update to previously installed OpenShift Enterprise 2.0 hosts,
install the updated packages on all hosts:

# yum update

Restart the MCollective service on each node host:

# service ruby193-mcollective restart

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 OpenShift Enterprise Infrastructure 2.0 x86_64
• Red Hat OpenShift Enterprise Application Node 2.0 x86_64
• Red Hat OpenShift Enterprise JBoss EAP add-on 2.0 x86_64

Fixes

• BZ - 1032982 - Correct gear cost info should display when adding cartridge using url for a scalable app
• BZ - 1033113 - [origin_node_331]Failed to build app by git push after deploying by artifact_url
• BZ - 1034517 - oo-diagnostics OSE 2.0 cherry-pick
• BZ - 1035237 - An warning message should be thrown out instead of error message when restore a app with postgresql-8.4 (and postgresl-9.2) cartridge
• BZ - 1038523 - The gear cost infomation on my applications page should be displayed for only 1 time when user have multiple domains.
• BZ - 1038753 - Scaled app, deployments retained over configured limit on a single gear [Upstream]
• BZ - 1038796 - EAP Cartridge manifest mentions 6.1.0 [Upstream]
• BZ - 1046202 - oo-diagnostics gives out the wrong path of the file when the log file "mcollective-client.log" has wrong permission

CVEs

(none)

References

(none)