Tuesday Oct 20, 2015

ANNOUNCING: Oracle OpenStack for Oracle Linux Release 2

Today we are announcing Oracle OpenStack for Oracle Linux Release 2. It is the first commercially available OpenStack deployment based on Docker containers without the need to install components individually with RPMs.

Based on the upstream Kilo release,  Oracle OpenStack for Oracle Linux R2 includes support for new modules such as Heat and Murano. The new release also uses MySQL Cluster to provide mission-critical Active/Active High Availability (HA), high performance and scalability for all OpenStack services.

Highlights of Oracle OpenStack for Oracle Linux Release 2:

  • Deployment and operation  are simplified compared to other vendors’ releases, by leveraging Docker-based containers for deployment and upgrades.
  • Based on upstream Kilo release
  • With integrated high availability features customers can now deploy MySQL in an Active/Active configuration, bringing enterprise database functionality to OpenStack cloud deployments.
  • Cinder drivers for Oracle All Flash FS Storage System and Oracle ZFS Storage Appliance are supported. Ceph Storage solutions are offered as a technology preview with Oracle Linux, utilizing the Ceph Object Gateway.
  • Integrated enterprise OpenStack cloud solution with a single line of support.
  • Oracle OpenStack for Oracle Linux is free to download and use. 
Read more:

If you are attending OOW15:

  • Visit Oracle OpenStack for Oracle Linux Demopod located at Oracle Linux, Oracle VM and OpenStack Showcase, Moscone South, Booth #121, Kiosk: SLX-010 to talk to product manager and engineers
  • Attend sessions: 
    • General Session: Oracle Linux-the State of the Penguin [GEN9479], Tuesday, Oct 27, 11:00 a.m. | Park Central-Metropolitan II
    • Maximize Your Private Cloud Investment with Oracle OpenStack for Oracle Linux [CON9574], Wednesday, Oct 28, 4:15 p.m. | Park Central-Metropolitan II
    • Rapid Private Cloud with Oracle VM and Oracle OpenStack for Oracle Linux [CON9576], Wednesday, Oct 28, 1:45 p.m. | Park Central-Metropolitan II
    • HOL: Getting Started with Oracle OpenStack for Oracle Linux [HOL10473], Monday, Oct 26, 12:30 p.m. | Hotel Nikko—Nikko Ballroom II (3rd Floor)

Connect with Oracle OpenStack for Oracle Linux via FacebookBlogYouTubeSlideShare

Tuesday Oct 06, 2015

Oracle OpenStack at Oracle OpenWorld 2015

Oracle OpenWorld 2015 is nearly here. We've got a great line up of OpenStack related sessions and hands on labs. If you're coming to the event and want to get up to speed on the benefits of OpenStack and the work that Oracle is doing across its product line to integrate with this cloud platform, make sure to check out the sessions below:

General Sessions

  • General Session: Oracle Linux-the State of the Penguin [GEN9479]
    Wim Coekaerts, Senior Vice President, Oracle
    Chris Kawalek, Sr. Principal Product Director, Oracle
    Tuesday, Oct 27, 11:00 a.m. | Park Central-Metropolitan II
  • General Session: Security, Speed, Simplicity - Hybrid Cloud Present & Future with Oracle Solaris [GEN8606]
    Markus Flierl, Vice President, Oracle
    Chris Riggin, Chief IT Architect, Verizon
    Tuesday, Oct 27, 11:00 a.m. | Intercontinental—Intercontinental C (5th Floor)

Conference Sessions

  • OpenStack and MySQL [CON2625]
    Matthew Lord, MySQL Product Manager, Oracle
    Thursday, Oct 29, 1:15 p.m. | Moscone South-250
  • Secure Private Cloud Done Right with Oracle and OpenStack [CON8313]
    Glynn Foster, Oracle Solaris Product Manager, Oracle
    Jeffrey Kiely, Principal Product Manager, Oracle
    Monday, Oct 26, 5:15 p.m. | Intercontinental-Intercontinental C (5th Floor)
  • Developer Cloud Made Simple: How to Build an OpenStack Developer Cloud [CON8337]
    Deepankar Bairagi, Principal Software Engineer, Oracle
    Liang Chen, Architect, Oracle Developer Studio, Oracle
    Nasser Nouri, Principal Software Engineer, Oracle
    Thursday, Oct 29, 9:30 a.m. | Intercontinental-Intercontinental B (5th Floor)
  • Maximize Your Private Cloud Investment with Oracle OpenStack for Oracle Linux [CON9574]
    Chris Kawalek, Sr. Principal Product Director, Oracle
    Dilip Modi, Principal Product Manager, Oracle OpenStack, Oracle
    Wednesday, Oct 28, 4:15 p.m. | Park Central-Metropolitan II
  • The DBaaS You’ve Been Waiting for-Oracle Database, Oracle Solaris, SPARC, and OpenStack [CON8354]
    Mehmet Kurtoglu, TT Group - BroadBand Database Operations Manager, TTNET
    Onder Ozbek, Product Manager, Oracle
    Eric Saxe, Director of Engineering, Oracle
    Thursday, Oct 29, 1:15 p.m. | Intercontinental-Intercontinental B (5th Floor)
  • Rapid Private Cloud with Oracle VM and Oracle OpenStack for Oracle Linux [CON9576]
    Michael Glasgow, Technical Product Manager, Oracle
    John Priest, Director, Oracle VM Product Management, Oracle
    Wednesday, Oct 28, 1:45 p.m. | Park Central-Metropolitan II
  • The Cutting Edge of Technology: Deploying a Secure Cloud with OpenStack [CON3225]
    Detlef Drewanz, Master Principal Sales Consultant, Oracle
    Eric Saxe, Director of Engineering, Oracle
    Thursday, Oct 29, 2:30 p.m. | Intercontinental-Intercontinental B (5th Floor)
  • Oracle Enterprise Manager? OpenStack? VSphere? Have It Your Way! [CON8059]
    Shrikanth Krupanandan, Director, Database administration, Fidelity Management & Research Company
    Scott Meadows, Director, Oracle
    Nirant Puntambekar, Senior Manager, Software Development, Oracle
    Thursday, Oct 29, 1:15 p.m. | Intercontinental-Sutter (5th Floor)
  • DevOps Done Right: Secure Virtualization with Oracle Solaris [CON8468]
    Duncan Hardie, Principal Product Manager, Oracle
    Fritz Wittwer, Service Engineer, Swisscom Schweiz AG
    Wednesday, Oct 28, 12:15 p.m. | Intercontinental-Intercontinental B (5th Floor)
  • MySQL Backup and Recovery-Use Cases and Solutions [CON2494]
    Mike Frank, Product Management Director, Oracle
    Monday, Oct 26, 12:15 p.m. | Moscone South-262

Hands on Labs

  • How to Build a Hadoop Cluster Using OpenStack [HOL1598]
    Ekine Akuiyibo, Software Engineer, Oracle
    Thursday, Oct 29, 9:30 a.m. | Hotel Nikko-Monterey
  • Oracle OpenStack for Oracle Solaris-Fast, Secure, and Compliant App Deployment [HOL10358]
    Scott Dickson, Principal Sales Engineer, Oracle
    Glynn Foster, Oracle Solaris Product Manager, Oracle
    Wednesday, Oct 28, 10:15 a.m. | Hotel Nikko-Nikko Ballroom I
  • Deploying a Multinode OpenStack Setup with a Preconfigured Oracle WebLogic Cluster [HOL6653]
    Sai spoorthy Padigi, Software Developer 1, Oracle
    Sandeep Shanbhag, Software Engineer, Oracle
    Wednesday, Oct 28, 8:45 a.m. | Hotel Nikko-Monterey
  • Oracle OpenStack for Oracle Solaris-a Complete Cloud Environment in Minutes [HOL10357]
    Scott Dickson, Principal Sales Engineer, Oracle
    Tuesday, Oct 27, 4:00 p.m. | Hotel Nikko-Nikko Ballroom I
  • In 60 Minutes: Build a Storage Cloud That Is Sustainable, Low Cost, and Secure [HOL2925]
    Donna Harland, Principal Product Manager, Oracle
    Joseph Lampitt, Solution Specialist, Oracle
    Hanli Ren, Senior Software Engineer, Oracle
    Adam Zhang, Principle Software Engineer, Oracle
    Tuesday, Oct 27, 8:45 a.m. | Hotel Nikko-Monterey
  • Build Your Own Cloud Environment with Oracle Solaris 11 RAD and REST [HOL6663]
    Gary Wang, Manager, Oracle
    Yu Wang, Software Engineer, Oracle
    Xiao-song Zhu, Principal Software Engineer, Oracle
    Monday, Oct 26, 3:30 p.m. | Hotel Nikko-Nikko Ballroom I
You can see all the session abstracts, along with all the rest of the Oracle OpenWorld 2015 content, at the content catalog. Looking forward to you joining us for a great event!

Friday Oct 02, 2015

Friday Spotlight: Oracle Linux, Virtualization, and OpenStack Showcase at OOW15

Happy Friday everyone!

Today's topic will be about our amazing showcase at Oracle OpenWorld, Oct 25-29. The Oracle Linux, Oracle VM and OpenStack showcase is located in Moscone South, booth #121, featuring Oracle product demos and Partners.  In past years, our showcase had been a great location to see demos of Oracle Linux and Oracle VM as well as solutions from our Partners. This year, it is expanded with Oracle OpenStack product demos and a theatre. Here's a list of the Oracle and Partner kiosks, don't forget to visit and talk to one of the experts that can help you out with your questions:

  • SLX-007 - Access Applications Securely with Oracle Secure Global Desktop
  • SLX-008 - Oracle VM VirtualBox
  • SLX-009 - Enhance Security and Reduce Costs Using Zero-Downtime Updates with Oracle Linux and Ksplice
  • SLX-010 - Oracle OpenStack for Oracle Linux  -Enterprise Ready
  • SLX-011 - Oracle Linux for the Cloud-Enabled Data Center
  • SLX-012  - Develop and Distribute Containerized Applications with Oracle Linux
  • SLX-013 - Oracle VM Server for x86
  • SLX-014 - Oracle VM Server for SPARC  

The table below lists the featured Partners and their solutions:

The Oracle Linux, Oracle VM, and OpenStack Showcase will also include an in-booth theatre for Partners and Oracle experts to share their solutions to customers and partners, alike. For the latest listing of theatre sessions currently confirmed please refer to the Schedule Builder

Don't forget to visit us at Moscone South #121, we will giveaway some great software (keeping it as a surprise- you need to come and see) and be in the drawing for our famous penguins and and Intel Mini PC - NUC appliance where you can use it for set top boxes to video surveillance, and home entertainment systems to digital signage, it is one appliance that can do it all.

Register today.

Tuesday Aug 11, 2015

Chat to us at OpenStack Silicon Valley 2015 Event!

Oracle is sponsoring the upcoming OpenStack Silicon Valley 2015 event in a couple of weeks time. We're looking forward to participating in the discussions, and we will have a sponsored session with Markus Flierl, VP of Solaris engineering (not currently posted on the schedule).

We've made some pretty great progress in OpenStack over the past year across all of the software and hardware portfolios that I mentioned in my recent OpenStack Silicon Valley blog post. The IT industry is moving fast and with the recent interest in containers, agile development and microservices, we're excited to see the standardization through recent efforts including the Open Container Initiative and our announcement to include Docker into Oracle Solaris. We'd love to chat to you about what we're doing with OpenStack, our Software in Silicon strategy at Oracle and some of the trends we're seeing in our customer base, at our booth E4. Come join us!

Tuesday Jul 28, 2015

Migrating Neutron Database from sqlite to MySQL for Oracle OpenStack for Oracle Solaris

Many OpenStack development environments use sqlite as a backend to store data. However in a production environment MySQL is widely used. Oracle also recommends to use MySQL for its OpenStack services. For many of the OpenStack services (nova, cinder, neutron...) sqlite is the default backend. Oracle OpenStack for Oracle Solaris users may want to migrate their backend database from sqlite to MySQL.

The general idea is to dump the sqlite database. Translate the dumped SQL statements so that they are compatible with MySQL. Stop neutron services. Create MySQL database. Replay the modified SQL statements in the MySQL database.

The details listed here are for the Juno release (integrated in Oracle Solaris 11.2 SRU 10.5 or newer) and Neutron is taken as an example use case.

Migrating neutron database from sqlite to MySQL

If not already installed, install MySQL

# pkg install --accept mysql-55 mysql-55/client python-mysql

Start the MySQL service
# svcadm enable -rs mysql

NOTE: If MySQL was already installed and running, then before running the next step double check that neutron database on MySQL is either not yet created or it is empty. The next step will drop the existing MySQL Neutron database if it exists on MySQL and create it. If the MySQL Neutron database is not empty then stop at this point. The following steps are limited to the case where MySQL neutron database and newly created/recreated.

Create Neutron database on MySQL

mysql -u root -p<<EOF
GRANT ALL PRIVILEGES ON neutron.* TO 'neutron'@'localhost' \
IDENTIFIED BY 'neutron';

Enter the root password when prompted

Identify that the Neutron services are online: # svcs -a | grep neutron | grep online | awk '{print $3}' \ > /tmp/neutron-svc
Disable the Neutron services: # for item in `cat /tmp/neutron-svc`; do svcadm disable $item; done
Make a backup of Neutron sqlite database:
# cp /var/lib/neutron/neutron.sqlite \
Get the db dump of Neutron from sqlite:
# /usr/bin/sqlite3 /var/lib/neutron/neutron.sqlite .dump \
       > /tmp/neutron-sqlite.sql

The following steps are run to create a neutron-mysql.sql file which will be compatible with MySQL database engine.

Suppress foreign key checks during create table/index
# echo 'SET foreign_key_checks = 0;' > /tmp/neutron-sqlite-schema.sql

Dump sqlite schema to a file
# /usr/bin/sqlite3 /var/lib/neutron/neutron.sqlite .dump  | \  grep -v 'INSERT INTO' >> /tmp/neutron-sqlite-schema.sql


Remove BEGIN/COMMIT/PRAGMA lines from the file.
(Oracle Solaris sed does not support -i option and hence redireciting to a new file 
 and then renaming it to original file)
# sed '/BEGIN TRANSACTION;/d; /COMMIT;/d; /PRAGMA/d' \ /tmp/neutron-sqlite-schema.sql \ > /tmp/neutron-sqlite-schema.sql.new \ && mv /tmp/neutron-sqlite-schema.sql.new \ /tmp/neutron-sqlite-schema.sql

Replace some SQL identifiers that are enclosed in double quotes, 
to be enclosed in back quotes
e.g. "limit to `limit`
# for item in binary blob group key limit type; do sed "s/\"$item\"/\`$item\`/g" \ /tmp/neutron-sqlite-schema.sql > /tmp/neutron-sqlite-schema.sql.new \ && mv /tmp/neutron-sqlite-schema.sql.new \ /tmp/neutron-sqlite-schema.sql; done

Enable foreign key checks at the end of the file

# echo 'SET foreign_key_checks = 1;' >> /tmp/neutron-sqlite-schema.sql 
Dump the data alone (INSERT statements) into another file

# /usr/bin/sqlite3 /var/lib/neutron/neutron.sqlite .dump \
| grep 'INSERT INTO' > /tmp/neutron-sqlite-data.sql
In INSERT statements table names are in double quotes in sqlite,
 but in mysql there should not be double quotes

# sed 's/INSERT INTO \"\(.*\)\"/INSERT INTO \1/g' \
/tmp/neutron-sqlite-data.sql > /tmp/neutron-sqlite-data.sql.new \
 && mv /tmp/neutron-sqlite-data.sql.new /tmp/neutron-sqlite-data.sql

Concat schema and data files to neutron-mysql.sql

# cat /tmp/neutron-sqlite-schema.sql \
/tmp/neutron-sqlite-data.sql > /tmp/neutron-mysql.sql 
Populate Neutron database in MySQL: # mysql neutron < /tmp/neutron-mysql.sql

Specify the connection under [database] section of /etc/neutron/neutron.conf file:

The connection string format is as follows:
connection = mysql://%SERVICE_USER%:%SERVICE_PASSWORD%@hostname/neutron 
For example:
connection = mysql://neutron:neutron@localhost/neutron
Enable the Neutron services:
# for item in `cat /tmp/neutron-svc`; do svcadm enable -rs $item; done 
# rm -f /var/lib/neutron/neutron.sqlite.ORIG \ /tmp/neutron-sqlite-schema.sql \ /tmp/neutron-sqlite-data.sql \   /tmp/neutron-mysql.sql 

Details about translating SQL statements to be compatible with MySQL

NOTE: /tmp/neutron-sqlite-schema.sql will have the Neutron sqlite database schema as SQL statements and /tmp/neutron-sqlite-data.sql will have the data in Neutron sqlite database which can be replayed to recreate the database. The sql statements in neutron-sqlite-schema.sql and neutron-sqlite-data.sql are to be MySQL compatible so that it can be replayed on MySQL Neutron database. A set of sed commands as listed above are used to create MySQL compatible SQL statements. The following text provides detailed information about the differences between sqlite and MySQL that are to be dealt with.

There are some differences in the way sqlite and MySQL expect the SQL statements to be which are as shown in the table below:

Reserved words are in double quotes: 
e.g "blob", "type", "key", 
"group", "binary", "limit"
Reserved words are in back quotes: 
e.g `blob`, `type`, `key`, 
`group`, `binary`, `limit`
Table name in Insert Statement 
are in quotes 
INSERT INTO "alembic_version"
Table name in Insert Statement 
are without quotes 
INSERT INTO alembic_version

Apart from the above the following requirements are to be met before running neutron.sql on MySQL:

The lines containing PRAGMA, 'BEGIN TRANSACTION', 'COMMIT' are to be removed from the file.


The CREATE TABLE statements with FOREIGN KEY references are to be rearranged (or ordered) in such a way that the TABLE name that is REFERENCED has to be created earlier than the table that is REFERRING it. The Indices on tables which are referenced by FOREIGN KEY statements are created soon after those tables are created. The last two requirements are not necessary if FOREIGN KEY check is disabled. Hence foreign_key_checks is SET to 0 at the beginning of neutron-mysql.sql and enabled again by setting foreign_key_checks to 1 before the INSERT statements in neutron-mysql.sql file.

New Oracle University course for Oracle OpenStack!

A new Oracle University course is now available: OpenStack Administration Using Oracle Solaris (Ed 1). This is a great way to get yourself up to speed on OpenStack, especially if you're thinking about getting a proof of concept, development or test, or even production environments online!

The course is based on OpenStack Juno in Oracle Solaris 11.2 SRU 10.5. Through a series of guided hands-on labs you will learn to:

  • Describe the OpenStack Framework.
  • Configure a Single-Node OpenStack Setup.
  • Configure a Multi-Node OpenStack Setup.
  • Administer OpenStack Resources Using the Horizon UI.
  • Manage Virtual Machine Instances.
  • Troubleshoot OpenStack.

The course is 3 days long and we recommend that you have taken a previous Oracle Solaris 11 administration course. This is an excellent introduction to OpenStack that you'll not want to miss!

Monday Jul 20, 2015

OpenStack and Hadoop

It's always interesting to see how technologies get tied together in the industry. Orgad Kimchi from the Oracle Solaris ISV engineering group has blogged about the combination of OpenStack and Hadoop. Hadoop is an open source project run by the Apache Foundation that provided distributed storage and compute for large data sets - in essence, the very heart of big data. In this technical How To, Orgad shows how to set up a multi-node Hadoop cluster using OpenStack by creating a pre-configured Unified Archives that can be uploaded to the Glance Image Repository for deployment across VMs created with Nova.

Check out: How to Build a Hadoop 2.6 Cluster Using Oracle OpenStack

Thursday Jul 09, 2015

PRESENTATION: Oracle OpenStack for Oracle Linux at OpenStack Summit Session

In this blog, we wanted to share a presentation given at OpenStack Summit in Vancouver early in May. We have just setup our SlideShare.net account and published our first presentation there.  

If you want to see more of these presentations, follow us at our Oracle OpenStack SlideShare space.

Tuesday Jul 07, 2015

What's New in Solaris OpenStack Juno Neutron

The  current update  of Oracle  OpenStack  for Oracle  Solaris updates  existing
features to the OpenStack Juno release and adds the following new features:

  1. Complete IPv6 Support for Tenant Networks
  2. Support for Source NAT
  3. Support for Metadata Services
  4. Support for Flat (untagged) Layer-2 Network Type
  5. Support for New Neutron Subcommands

1. Complete IPv6 Support for Tenant Networks

Finally, Juno  provides feature parity between  IPv4 and IPv6. The  Juno release
allows tenants to create networks and associate IPv6 subnets with these networks
such that  the VM instances  that connect to these  networks can get  their IPv6
addresses in either of the following ways:

- Stateful address configuration
- Stateless address configuration

Stateful or DHCPv6  address configuration is facilitated  through the dnsmasq(8)

Stateless address configuration is facilitated in either of the following ways:
- Through the provider (physical) router in the data center networks.
- Through  the  Neutron  router  and Solaris  IPv6  neighbor  discovery  daemon
(in.ndpd(1M)). The Neutron L3 agent  sets the AdvSendAdvertisements parameter to
true in ndpd.conf(4) for  an interface that hosts the IPv6  subnet of the tenant
and refreshes the SMF service (svc:/network/routing/ndp:default) associated with
the daemon.

This  IPv6 support  adds the  following two  new attributes  to Neutron  Subnet:
ipv6_address_mode and  ipv6_ra_mode. Possible  values for these  attributes are:
slaac, dhcpv6-stateful, and  dhcpv6-stateless. The two Neutron  agents - Neutron
DHCP agent and Neutron L3 agent - work together to provide IPv6 support.

For most  cases, these new  attributes are  set to the  same value. For  one use
case,  only  ipv6_address_mode  is  set.   The  following  table  provides  more

2. Support for Source NAT

The floating IPs feature in  OpenStack Neutron provides external connectivity to
VMs by performing a  one-to-one NAT of a the internal IP address  of a VM to the
external floating IP address.

The SNAT  feature provides external connectivity  to all of the  VMs through the
gateway public IP. The  gateway public IP is the IP address  of the gateway port
that   gets  created   when   you  execute   the   following  command:   neutron
router-gateway-set router_uuid external_network_uuid

This external  connectivity setup is similar  to wireless network setup  at home
where you have a single public IP from the ISP configured on the router, and all
our personal devices  are behind this IP on an  internal network. These internal
devices can  reach out to  anywhere on the  internet through SNAT;  however, the
external entities cannot reach these internal devices.


- Create a Public network

# neutron net-create --router:external=True --provider:network_type=flat public_net
neutron Created a new network:
| Field                 | Value                                |
| admin_state_up        | True                                 |
| network_id            | 3c9c4bdf-2d6d-40a2-883b-a86076def1fb |
| name                  | public_net                           |
| provider:network_type | flat                                 |
| router:external       | True                                 |
| shared                | False                                |
| status                | ACTIVE                               |
| subnets               |                                      |
| tenant_id             | dab8af7f10504d3db582ce54a0ce6baa     |

# neutron subnet-create --name public_subnet --disable-dhcp \
--allocation-pool start=,end= \
--allocation-pool start=,end= \
Created a new subnet:
| Field             | Value                                              |
| allocation_pools  | {"start": "", "end": ""} |
|                   | {"start": "", "end": ""} |
| cidr              |                                     |
| dns_nameservers   |                                                    |
| enable_dhcp       | False                                              |
| gateway_ip        |                                        |
| host_routes       |                                                    |
| subnet_id         | 6063613c-1008-4826-ae17-ce6a58511b2f               |
| ip_version        | 4                                                  |
| ipv6_address_mode |                                                    |
| ipv6_ra_mode      |                                                    |
| name              | public_subnet                                      |
| network_id        | 3c9c4bdf-2d6d-40a2-883b-a86076def1fb               |
| tenant_id         | dab8af7f10504d3db582ce54a0ce6baa                   |

- Create a private network

# neutron net-create private_net
# neutron subnet-create --name private_sunbet private_net

- Create a router

# neutron router-create provider_router
Created a new router:
| Field                 | Value                                |
| admin_state_up        | True                                 |
| external_gateway_info |                                      |
| router_id             | b48fd525-2519-4501-99d9-9c2d51a543f1 |
| name                  | provider_router                      |
| status                | ACTIVE                               |
| tenant_id             | dab8af7f10504d3db582ce54a0ce6baa     |

Note: Update /etc/neutron/l3_agent.ini file with following entry and
restart neutron-l3-agent SMF service (svcadm restart neutron-l3-agent)
router_id = b48fd525-2519-4501-99d9-9c2d51a543f1

- Add external network to router

# neutron router-gateway-set provider_router public_net
Set gateway for router provider_router
# neutron router-show provider_router
| Field                 | Value                                                                        |
| admin_state_up        | True                                                                         |
| external_gateway_info | {"network_id": "3c9c4bdf-2d6d-40a2-883b-a86076def1fb",                       |
|                       | "enable_snat": true,                                                         |
|                       | "external_fixed_ips": [{"subnet_id": "6063613c-1008-4826-ae17-ce6a58511b2f", |
|                       | "ip_address": ""}]}                                             |
| router_id             | b48fd525-2519-4501-99d9-9c2d51a543f1                                         |
| name                  | provider_router                                                              |
| status                | ACTIVE                                                                       |
| tenant_id             | dab8af7f10504d3db582ce54a0ce6baa                                             |

Note: By default, SNAT is  enabled on the gateway interface of  the Neutron router. To
disable  this  feature,  specify  the   --disable-snat  option  to  the  neutron
router-gateway-set subcommand.

- Add internal network to router

# neutron router-interface-add provider_router private_subnet
Added interface 9bcfd21a-c751-40bb-99b0-d9274523e151 to router provider_router.

# neutron router-port-list provider_router
| id                                   | mac_address       | fixed_ips                               |
| 4b2f5e3d-0608-4627-b93d-f48afa86c347 | fa:16:3e:84:30:e4 | {"subnet_id":                           |
|                                      |                   | "6063613c-1008-4826-ae17-ce6a58511b2f", |
|                                      |                   | "ip_address": ""}          |
|                                      |                   |                                         |
| 9bcfd21a-c751-40bb-99b0-d9274523e151 | fa:16:3e:df:c1:0f | {"subnet_id":                           |
|                                      |                   | "c7f99141-25f0-47af-8efb-f5639bcf6181", |
|                                      |                   | "ip_address": ""}          |
Now all of the VMs that are in the internal network can reach outside through SNAT through

3. Support for Metadata Services

A metadata service provides an OpenStack VM instance with information such as the

 -- The public IP/hostname
 -- A random seed
 -- The metadata that the tenant provided at install time
 -- and much more

The metadata requests are  made by the VM instance at the  well known address of, port  80. All  such requests  arrive at  the Neutron  L3 agent,
which forwards the requests to a Neutron  proxy server running at port 9697. The
proxy  server was  spawned by  the Neutron  L3 agent.  The Neutron  proxy server
forwards the requests  to the Neutron metadata agent through  a UNIX socket. The
Neutron metadata  agent interacts with  the Neutron Server service  to determine
the instance UUID that is making  the requests. After the Neutron metadata agent
gets the  instance UUID, it makes  a call into  the Nova API metadata  server to
fetch  the information  for the  VM instance.  The fetched  information is  then
passed back to the instance that made the request.

4. Support for Flat (untagged) Layer-2 Network Type

Flat OpenStack Network is used to place all the VM instances on the same segment
without VLAN  or VXLAN.  This  means that the VM  instances will share  the same

In the flat l2-type there is no VLAN tagging or other network segregation taking
place,  i.e., all  the VNICs  (and thus  VM instances)  that connect  to a  flat
l2-type network are created with VLAN ID set to 0.  It follows that flat l2-type
cannot be used to achieve multi-tenancy. Instead, it will be used by data center
admins to map directly to the existing physical networks in the data center.

One  use of  Flat network  type is  in the  configuration of  floating IPs.   If
available floating IPs are subset of  the existing physical network's IP subnet,
then you would need to create flat  network with subnet set to physical networks
IP  subnet and  allocation pool  set to  available floating  IPs.  So,  the flat
network contains  part of  the existing  physical network's  IP subnet.  See the
examples in previous section.

5. Support for New Neutron Subcommands

With this  Solaris OpenStack  Juno release,  you can  run multiple  instances of
neutron-dhcp-agent, each instance running on  a separate network node. Using the
dhcp-agent-network-add neutron  subcommand, you can manually  select which agent
should serve a DHCP enabled subnet. By default, the Neutron server automatically
load balances the work among various DHCP agents.

The following table  shows the new subcommands  that have been added  as part of
the Solaris OpenStack Juno release.
| neutron subcommands       | Comments                                |
| agent-delete              | Delete a given agent.                   |
| agent-list                | List agents.                            |
| agent-show                | Show information for a specified agent. |
| agent-update              | Update the admin status and             |
|                           | description for a specified agent.      |
| dhcp-agent-list-          | List DHCP agents hosting a network.     |
| hosting-net               |                                         |
| net-list-on-dhcp-agent    | List the networks on a DHCP agent.      |
| dhcp-agent-network-add    | Add a network to a DHCP agent.          |
| dhcp-agent-network-remove | Remove a network from a DHCP agent.     |

OpenStack Juno in Solaris 11.3 Beta

It's been less than year since we announced availability of the Havana version of the OpenStack cloud infrastructure software as part of Solaris 11.2 and we've since continued to see what can only be described as a startling amount of momentum build in the OpenStack community. It's an incredibly exciting space for us, and for Oracle as a whole, as we watch the benefits of cloud based infrastructure and service management transform the way in which our customers run their Enterprises. 

Fully automated self-service provisioning and orchestration of compute, network, and storage is a beautiful thing...empowering developers to self-provision in minutes the infrastructure needed to build, test or deploy applications without having to waste time trying to file tickets, procure systems, or wait on others. Administrators are able to view, and manage what would otherwise be a sprawl of compute, networking, and storage as an actual system. Rather than wasting time repeatedly servicing individual requests, they can instead focus their attention on managing the cloud 's resources as a pool, and ensuring smooth operation of services provided by the cloud.

We've watched this transformation happen internally in Solaris Engineering as we've shifted from ad-hoc management of the test and development systems used, to managing that infrastructure as an OpenStack cloud. Utilization efficiency of our infrastructure has dramatically improved as Engineers who formerly "camped" on systems to ensure those environments would be available when needed no longer need to, since they can easily save and later re-deploy images of their development environment in minutes. Wasted time formerly spent hunting through lists of systems trying to find one that's free, working, and sufficient, is now spent getting actual work done, or better yet, drinking coffee!

If you've been thinking you would like to get started learning about OpenStack, perhaps by experimenting and building yourself a small private cloud, there's really never been a better time. Especially since today we're very excited to announce that OpenStack Juno is now available to you as part of Oracle Solaris 11.3 Beta. You can start small, and in about 10 minutes install a Solaris Unified Archive that essentially is a fully configured OpenStack Cloud-In-A-Box. Deploy the OpenStack Unified Archive to a system, perform a few configuration steps (specific to your environment, e.g. SSH keys and such), and voila you have a functional OpenStack cloud that you can start learning how to operate.

If you are more experienced with OpenStack and are looking to build a cloud system for your Enterprise that is powered by best of breed Solaris technologies, such as Solaris Zones, the ZFS file system, and Solaris SDN...and that leverages SPARC systems, x86 systems (or both) you'll appreciate how well we've integrated the worlds most popular open source cloud infrastructure software with the Solaris technologies you've come to know and trust.

Within Solaris 11.3 Beta, we've integrated the Juno versions of the core OpenStack Cloud Infrastructure services: Nova, Neutron, Cinder, Swift, Keystone, Glance, Heat, and Horizon, along with the drivers enabling OpenStack to drive Solaris virtualization, and ZFS backed shared storage over iSCSI or FC (both from Solaris natively or via the ZFS Storage Appliance). Within OpenStack Horizon, you'll find an integrated Zones Console interface, and you can upgrade your 11.2 Havana based OpenStack cloud via IPS to Juno based Solaris 11.3 Beta.

Post 11.3 Beta, we'll be very excited to introduce bare metal provisioning support for SPARC and x86 systems through OpenStack Ironic. In addition to being able to offer virtualized environments of varying sizes/configs (e.g. flavors) to cloud tenants, Ironic enables bare metal flavors to also be provided. We'll probably also have a few more exciting features to talk about as well. :) But in the meanwhile, we hope you enjoy OpenStack Juno on Solaris 11.3 Beta, and do let us know if you have any questions and/or run into any issues as we would be more than happy to help!

Wednesday May 20, 2015

LIVE WEBINAR (May 28): How to Get Started with Oracle OpenStack for Oracle Linux

Webinar title: Oracle VM VirtualBox to Get Started with Oracle OpenStack for Oracle Linux

Date: Thursday, May 28, 2015

Time: 10:00 AM PDT

Dilip Modi, Principal Product Manager, Oracle OpenStack
Simon Coter, Principal Product Manager, Oracle VM and VirtualBox

You are invited to our webinar about how to get started with Oracle OpenStack for Oracle Linux. Built for enterprise applications and simplified for IT, Oracle OpenStack for Oracle Linux is an integrated solution that focuses on simplifying the building of a cloud foundation for enterprise applications and databases. In this webcast, Oracle experts will discuss how to use Oracle VM VirtualBox to create an Oracle OpenStack for Oracle Linux test environment and get you started learning about Oracle OpenStack for Oracle Linux.

Tuesday May 19, 2015

Join us at the Oracle OpenStack booth!

We've reached the second day of the OpenStack Summit in Vancouver and our booth is now officially open. Come by and see us and talk about some of the work that we've been doing at Oracle - whether it's integrating a complete distribution of OpenStack into Oracle Linux and Oracle Solaris, Cinder and Swift storage on the Oracle ZFS Storage Appliance, integration with Swift and our Oracle HSM tape storage product, and how to quickly provision Oracle Database 12c in an OpenStack environment. We've got a lot of demos and experts there to answer your questions.

The Oracle sponsor session is on today also. Markus Flierl will be talking about "Making OpenStack Secure and Compliant for the Enterprise" at 2:50-3:30pm Tuesday Room 116/117. Markus will talk about the challenges of deploying an OpenStack cloud while still meeting critical secure and compliance requirements, and how Oracle can help you do this.

And in case anyone asks, yes, we're hiring!

Thursday Apr 16, 2015

OpenStack Swift on Oracle Solaris

Jim Kremer has written a blog about the OpenStack object storage service Swift and how to set it up on Oracle Solaris. For Swift on Solaris we use the ZFS file system as the underlying storage, which means we can take advantage of things like snapshots and clones, data encryption and compression, and the underlying redundancy that the ZFS architecture provides with storage pools and mirroring.

Read Jim's blog on How to get Swift up and running on Solaris.

-- Glynn Foster

Thursday Apr 09, 2015

Oracle at OpenStack Summit in Vancouver - May 18-22

Oracle is premier sponsor at OpenStack Summit in Vancouver, May 18-22. This year we will have experts from all of Oracle's OpenStack technologies including Oracle Linux and Oracle VM, Oracle Solaris, Oracle ZFS Storage Appliance, and Oracle Tape Storage Solutions. We will have informative sessions and booth to visit. Here's one of the Oracle sessions:

Title:Making OpenStack secure and compliant for the enterprise

Many Enterprises deploying OpenStack also need to meet Security and Compliance requirements. In this talk, you will learn how Oracle can help you address these requirements with OpenStack Cloud Infrastructure solutions designed to meet the needs of the Enterprise. Come learn how Oracle can help you deploy OpenStack solutions that you can trust to meet the needs of your enterprise, your customers, and the demands of mission-critical cloud services.

Tuesday, May 19 from 2:50 p.m. to 3:30 p.m., Room 116 / 117

We encourage you to visit the Oracle Booth # P9 for discussion with our OpenStack experts on your requirements and how best to adress your issues for smooth deployment. Marketplace hours and demos will be done on: 

  • Monday, May 18: 6:00pm – 7:30pm
  • Tuesday, May 19: 10:45am – 6:00pm
  • Wednesday, May 20: 9:00am – 6:00pm
  • Thursday, May 21: 9:00am – 4:10pm 

Hope to meet you at OpenStack Summit!  

Monday Mar 16, 2015

OpenStack Summit Vancouver - May 18-22

The next OpenStack developers and users summit will be in Vancouver. Oracle will again be a sponsor of this event, and we'll have a bunch of our team present from Oracle Solaris, Oracle Linux, ZFS Storage Appliance and more. The summit is a great opportunity to sync up on the latest happenings in OpenStack. By this stage the 'Kilo' release will be out and the community will be in full plan mode for 'Liberty'. Join us there and see what the Oracle teams have been up to recently!

-- Glynn Foster


Oracle OpenStack is cloud management software that provides customers an enterprise-grade solution to deploy and manage their entire IT environment. Customers can rapidly deploy Oracle and third-party applications across shared compute, network, and storage resources with ease, with end-to-end enterprise-class support. For more information, see here.


« November 2015