In OpenStack Glance through 32.0.0, the /v2/tasks API accepts type=import tasks that bypass importfilteringopts, allowing an admin to fetch internal URLs from the Glance service network (aka SSRF), as long as https:// or http:// is used. This API has been available only to admins since Xena, and it has been deprecated for several releases.

In OpenStack Neutron before 28.0.1, a project manager can create or update a port on a shared network owned by another project and set deviceowner to a value that has "network:" at the beginning ("network:dhcp" for example). The default port RBAC policies incorrectly included PROJECTMANAGER without requiring network ownership, allowing any project manager to obtain trusted network-service port behavior on shared networks. Depending on backend and deployment, this can bypass anti-spoofing and security group protections, enabling DHCP, MAC, or IP spoofing against other tenants on the shared network. This is a regression of CVE-2015-5240 (OSSA-2015-018).

In OpenStack Ironic before 35.0.2 (in a certain non-default configuration), instanceinfo['kstemplate'] is rendered without sandboxing.

Maxence Bornecque from Orange Cyberdefense CERT Vulnerability Intelligence Watch Team reported a vulnerability in Keystone's EC2 credential creation endpoint. By using a restricted application credential to call the EC2 credential creation API, an authenticated user with only a reader role may obtain an EC2/S3 credential that carries the full set of the parent user's S3 permissions, effectively bypassing the role restrictions imposed on the application credential. Only deployments that use restricted application credentials in combination with the EC2/S3 compatibility API (swift3 / s3api) are affected.

OpenStack Ironic before 29.0.1 can write unintended files to a target node disk during image handling (if a deployment was performed via the API). A malicious project assigned as a node owner can provide a path to any local file (readable by ironic-conductor), which may then be written to the target node disk. This is difficult to exploit in practice, because a node deployed in this manner should never reach the ACTIVE state, but it still represents a danger in environments running with non-default, insecure configurations such as with automated cleaning disabled. The fixed versions are 24.1.3, 26.1.1, and 29.0.1.

A flaw was found in openstack-glance. This issue could allow a remote, authenticated attacker to tamper with images, compromising the integrity of virtual machines created using these modified images.

An issue was discovered in OpenStack Nova before 23.2.2, 24.x before 24.1.2, and 25.x before 25.0.2. By creating a neutron port with the direct vnictype, creating an instance bound to that port, and then changing the vnictype of the bound port to macvtap, an authenticated user may cause the compute service to fail to restart, resulting in a possible denial of service. Only Nova deployments configured with SR-IOV are affected.

An issue was discovered in OpenStack Nova before 18.2.4, 19.x before 19.1.0, and 20.x before 20.1.0. It can leak consoleauth tokens into log files. An attacker with read access to the service's logs may obtain tokens used for console access. All Nova setups using novncproxy are affected. This is related to NovaProxyRequestHandlerBase.newwebsocketclient in console/websocketproxy.py.

OpenStack Horizon allows remote authenticated administrators to conduct XSS attacks via a crafted federation mapping.

Upstream bug:

https://bugs.launchpad.net/horizon/+bug/1667086

A vulnerability in Neutron anti-spoof protection. By forging DHCP discovery messages or non-IP traffic, such as ARP or ICMPv6, an instance may spoof IP or MAC source addresses on attached networks resulting in denial of services and/or traffic interception. Moreover when L2population isn't used, other tenants attached to a shared network are also vulnerable. Neutron setups using the IPTables firewall driver are affected.

Upstream bug:

https://bugs.launchpad.net/bugs/1502933

References:

http://seclists.org/oss-sec/2016/q2/519

A vulnerability in Neutron anti-spoof protection. By forging DHCP discovery messages or non-IP traffic, such as ARP or ICMPv6, an instance may spoof IP or MAC source addresses on attached networks resulting in denial of services and/or traffic interception. Moreover when L2population isn't used, other tenants attached to a shared network are also vulnerable. Neutron setups using the IPTables firewall driver are affected.

Upstream bug:

https://bugs.launchpad.net/bugs/1558658

References:

http://seclists.org/oss-sec/2016/q2/519

A vulnerability in Neutron anti-spoof protection. By forging DHCP discovery messages or non-IP traffic, such as ARP or ICMPv6, an instance may spoof IP or MAC source addresses on attached networks resulting in denial of services and/or traffic interception. Moreover when L2population isn't used, other tenants attached to a shared network are also vulnerable. Neutron setups using the IPTables firewall driver are affected.

Upstream bug:

https://bugs.launchpad.net/bugs/1558658

References:

http://seclists.org/oss-sec/2016/q2/519

OpenStack Compute (Nova) before 2015.1.3 (kilo) and 12.0.x before 12.0.1 (liberty), when using libvirt to spawn instances and usecowimages is set to false, allow remote authenticated users to read arbitrary files by overwriting an instance disk with a crafted image and requesting a snapshot.

It was reported that a vulnerability was found in Neutron. By changing the device owner of an instance's port right after it is created, an authenticated user may prevent application of firewall rules and so avoid IP anti-spoofing controls. All Neutron setups using the ML2 plugin or a plugin that relies on the security groups AMQP API are affected. All Neutron setups using the ML2 plugin or a plugin that relies on the security groups AMQP API are affected.

Vulnerability affects versions through 2014.2.3 and 2015.1 versions through 2015.1.1

Acknowledgements:

Red Hat would like to thank the OpenStack project for reporting this issue. Upstream acknowledges Kevin Benton from Mirantis as the original reporter.

It was discovered that the Designate component in OpenStack would enter an infinite loop when processing an internal zone file transfer if a managed DNS zone included a resource record set whose size exceeded the limitations of the DNS protocol, leading to a denial of service. Only authenticated users with access to the Designate component can add such resource record sets.

Acknowledgements:

This issue was discovered by Florian Weimer of Red Hat Product Security.

Multiple cross-site scripting (XSS) vulnerabilities in OpenStack Dashboard (Horizon) 2015.1.0 allow remote authenticated users to inject arbitrary web script or HTML via the metadata to a (1) Glance image, (2) Nova flavor or (3) Host Aggregate.

Michael Scherer of Red Hat reported multiple issues in OpenStack Trove where temporary files are used in an insecure way in different modules:

trove: MongoDB datastore module in guestagent insecure /tmp file usage in writeconfig() https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/experimental/mongodb/service.py#L176

trove: PostgreSQL datastore module in guestagent insecure /tmp file usage in resetconfiguration() https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/experimental/postgresql/service/config.py#L70

trove: Redis datastore module in guestagent insecure /tmp file usage in writeconfig() https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/experimental/redis/service.py#L236

trove: MySQL datastore module in guestagent insecure /tmp file usage in writemycnf() https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/mysql/service.py#L790

trove: MySQL restore strategy in guestagent insecure /tmp file usage in InnoBackupEx::runprepare() https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/restore/mysqlimpl.py#L194

trove: MySQL backup strategy in guestagent insecure /tmp file usage in InnoBackupEx::cmd() https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/mysqlimpl.py#L55

trove: MySQL backup strategy in guestagent insecure /tmp file usage in MySQLDump::cmd() https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/mysqlimpl.py#L36

trove: MySQL backup strategy in guestagent insecure /tmp file usage in InnoBackupExIncremental::cmd() https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/mysqlimpl.py#L110

trove: Cassandra datastore module insecure /tmp file usage in getactualdbstatus() https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/experimental/cassandra/system.py#L33 https://github.com/openstack/trove/blob/master/trove/guestagent/datastore/experimental/cassandra/service.py#L230

trove: Couchbase backup strategy insecure /tmp file usage in multiple class CbBackup methods https://github.com/openstack/trove/blob/master/trove/guestagent/strategies/backup/experimental/couchbaseimpl.py#L30

It was reported that under certain conditions (when live migrations fails), an attacker can access other VMs volumes, which under normal conditions he should not be able to access: https://bugs.launchpad.net/nova/+bug/1419577

CVE has been assigned here: http://seclists.org/oss-sec/2015/q1/990 No patches are available at the time of writing.

Cross-site scripting (XSS) vulnerability in horizon/static/horizon/js/horizon.instances.js in the Launch Instance menu in OpenStack Dashboard (Horizon) before 2013.2.4, 2014.1 before 2014.1.2, and Juno before Juno-2 allows remote authenticated users to inject arbitrary web script or HTML via a network name.

Cross-site scripting (XSS) vulnerability in the Groups panel in OpenStack Dashboard (Horizon) before 2013.2.4, 2014.1 before 2014.1.2, and Juno before Juno-2 allows remote administrators to inject arbitrary web script or HTML via a user email address, a different vulnerability than CVE-2014-3475.

Cross-site scripting (XSS) vulnerability in the Users panel (admin/users/) in OpenStack Dashboard (Horizon) before 2013.2.4, 2014.1 before 2014.1.2, and Juno before Juno-2 allows remote administrators to inject arbitrary web script or HTML via a user email address, a different vulnerability than CVE-2014-8578.

The strutils.maskpassword function in the OpenStack Oslo utility library, Cinder, Nova, and Trove before 2013.2.4 and 2014.1 before 2014.1.3 does not properly mask passwords when logging commands, which allows local users to obtain passwords by reading the log.

The processutils.execute function in OpenStack oslo-incubator, Cinder, Nova, and Trove before 2013.2.4 and 2014.1 before 2014.1.3 allows local users to obtain passwords from commands that cause a ProcessExecutionError by reading the log.

The VMWare driver in OpenStack Compute (Nova) before 2014.1.3 allows remote authenticated users to bypass the quota limit and cause a denial of service (resource consumption) by putting the VM into the rescue state, suspending it, which puts into an ERROR state, and then deleting the image. NOTE: this vulnerability exists because of an incomplete fix for CVE-2014-2573.

Cross-site scripting (XSS) vulnerability in the Host Aggregates interface in OpenStack Dashboard (Horizon) before 2013.2.4, 2014.1 before 2014.1.2, and Juno before Juno-3 allows remote administrators to inject arbitrary web script or HTML via a new host aggregate name.

The L3-agent in OpenStack Neutron before 2013.2.4, 2014.x before 2014.1.2, and Juno before Juno-2 allows remote authenticated users to cause a denial of service (IPv4 address attachment outage) by attaching an IPv6 private subnet to a L3 router.

OpenStack Orchestration API (Heat) 2013.2 through 2013.2.3 and 2014.1, when creating the stack for a template using a provider template, allows remote authenticated users to obtain the provider template URL via the resource-type-list.

The l3-agent in OpenStack Neutron 2012.2 before 2013.2.3 does not check the tenant id when creating ports, which allows remote authenticated users to plug ports into the routers of arbitrary tenants via the device id in a port-create command.

The instance rescue mode in OpenStack Compute (Nova) 2013.2 before 2013.2.3 and Icehouse before 2014.1, when using libvirt to spawn images and usecowimages is set to false, allows remote authenticated users to read certain compute host files by overwriting an instance disk with a crafted image.

It was reported that, if an Horizon user were tricked into using a malicious template in the Orchestration/Stack section of Horizon, it would be possible for an attacker to conduct cross-site scripting (XSS) attacks.

The original report notes "2013.2.1 version up to 2013.2.2" are affected.

Acknowledgements:

Red Hat would like to thank the OpenStack project for reporting this issue. Upstream acknowledges Cristian Fiorentino from Intel as the original reporter.