RHSA-2017:0515 - Security Advisory
Important: ansible and ceph-ansible security, bug fix, and enhancement update
Security Advisory: Important
An update for ansible and ceph-ansible is now available for Red Hat Storage Console 2.
Red Hat Product Security has rated this update as having a security impact of Important. A Common Vulnerability Scoring System (CVSS) base score, which gives a detailed severity rating, is available for each vulnerability from the CVE link(s) in the References section.
The ceph-ansible package provides Ansible playbooks for installing, maintaining, and upgrading Red Hat Ceph Storage.
Ansible is a simple model-driven configuration management, multi-node deployment, and remote task execution system. Ansible works over SSH and does not require any software or daemons to be installed on remote nodes. Extension modules can be written in any language and are transferred to managed machines automatically.
The following packages have been upgraded to a later upstream version: ceph-installer (1.2.2), ansible (188.8.131.52), python-passlib (1.6.5), ceph-ansible (2.1.9). (BZ#1405630)
- An input validation vulnerability was found in Ansible's handling of data sent from client systems. An attacker with control over a client system being managed by Ansible and the ability to send facts back to the Ansible server could use this flaw to execute arbitrary code on the Ansible server using the Ansible server privileges. (CVE-2016-9587)
- Previously, the ceph-ansible utility permanently disabled the swap partition. With this update, ceph-ansible can no longer disable swap. (BZ#1364167)
- Previously, the ceph-ansible utility did not support adding encrypted OSD nodes. As a consequence, an attempt to upgrade to a newer, minor, or major version failed on encrypted OSD nodes. In addition, Ansible returned the following error message during the disk activation task:
mount: unknown filesystem type 'crypto_LUKS'
With this update, ceph-ansible supports adding encrypted OSD nodes, and upgrading works as expected. (BZ#1366808)
- Due to a bug in the underlying source code, the ceph-ansible utility in some cases failed on the copy roundep task. Consequently, the installation process was unsuccessful. This bug has been fixed, and the installation now proceeds as expected. (BZ#1382878)
- Previously, installation using the ceph-ansible utility failed on the "fix partitions gpt header or labels of the journal devices" task in the ceph-osd role because of an empty variable. The underlying source code has been modified, and the installation no longer fails in this case. (BZ#1400967)
- Previously, Red Hat Console Agent setup performed by the ceph-ansible utility only supported installations by using the Content Delivery Network (CDN). Installations with an ISO file or local Yum repository failed. With this update, all installations are successful. (BZ#1403576)
- Previously, the ceph-ansible utility was unable to purge a cluster with encrypted OSD devices because the underlying ceph-disk utility was unable to destroy the partition table on an encrypted device by using the "--zap-disk" option. The underlying source code has been fixed allowing ceph-disk to use the "--zap-disk" option on encrypted devices. As a result, ceph-ansible can purge clusters with encrypted OSD devices as expected. (BZ#1414647)
- Previously, during the creation of Ceph clusters with nodes that use IPv6 addressing, ceph-ansible added the "ms bind ipv6" key to the Ceph configuration file, but it did not assign any value to it. This behavior caused an error when parsing the configuration file after the cluster creation. With this update, the "ms bind ipv6" key is properly set in the Ceph configuration file allowing for proper configuration file parsing on clusters that use IPv6 addressing. (BZ#1419814)
- The ceph-ansible utility now supports the client role. This new role enables you to install Ceph clients by using Ansible and deploy nodes to run tests against the Ceph cluster. (BZ#1384622)
- The ceph-installer API now supports installation of OSD nodes that have journals collocated on the same devices. (BZ#1412867)
For details on how to apply this update, which includes the changes described in this advisory, refer to:
- Red Hat Storage Console 2 x86_64
- BZ - 1312443 - restrict access to endpoints to localhost only
- BZ - 1339576 - [ceph-ansible] purge-cluster fails with "grep: write error"
- BZ - 1364167 - [ceph-ansible] role ceph-common disables swap by default
- BZ - 1366808 - Ansible: upgrade nodes with encrypted OSDs and support for dm-crypt
- BZ - 1382878 - [ansible 1.0.8-1] fails during copy rundep - with_items expects a list or a set
- BZ - 1384622 - Add the client role to ceph-ansible
- BZ - 1400967 - [ceph-ansible] ceph-ansible failing to check raw journal devices
- BZ - 1403576 - Agent installation fails in ceph nodes upgraded from 1.3 to 2.0 and connected to local yum repository
- BZ - 1404378 - CVE-2016-9587 Ansible: Compromised remote hosts can lead to running commands on the Ansible controller
- BZ - 1404515 - ceph-ansible should set set apt pinning for deb packages
- BZ - 1405630 - resync with ceph-ansible upstream v2.1 release
- BZ - 1405985 - [ceph-ansible]: enable CDN repository to allow installation of ceph-mds
- BZ - 1406556 - Upgrade of Ansible packages (184.108.40.206)
- BZ - 1412867 - Allow OSDs to be configured/installed with journal colocation
- BZ - 1413985 - [ceph-ansible]: purge-cluster.yml fails with ""Partition number 2 out of range!" on RHEL
- BZ - 1414647 - [ceph-ansible]: purge cluster fails to zap OSD disks having encrypted OSDs
- BZ - 1416010 - [ceph-ansible] config file generation fails on ipv6 nodes with message dict object' has no attribute 'address'
- BZ - 1418980 - [ceph-ansible]: purge cluster on ubuntu fails to purge client node.
- BZ - 1419814 - [ceph-ansible] no value is set to the key 'ms bind ipv6' in config file during ipv6 cluster creation
- BZ - 1422134 - RHSCON-2: purge-cluster does not remove calamari-server packages
Red Hat Storage Console 2