Oracle Linux Reviews

• Installing Oracle APEX for testing purposes
• Learning Oracle APEX
• Making some fancy web applications, including forms that load and display images

The main aim of installing Oracle Linux was to test it after years of using CentOS, and  to test Oracle APEX. After one year of testing,  I encountered problems that I was not able to solve: 
I was not able to do rotated backups; upgrading to a newer version of APEX was a nightmare; I could not enter GlassFish admin console; and I could not install SQL Developer. Then I thought that if I bought the cheapest version of Oracle Database I would get the support neccesary to get this know how. Two sales persons contacted me. But they were not interested in solving my doubts. Their only interest was selling me the cloud database, for the fantastic price of (more or less) 1000 euros per year. I explained to them that my alternative was a second hand PC, plus CentOS, plus Postgres, plus Eclipse.  They did not offer me a cheap version of Oracle database. I understoo the message, Oracle, both Linux and database, were not for me.

Ease of development of professional-looking web pages, full of functionality and with secure access.

I decided to purchase Oracle Database. I thought that if I bought the product, I would get support. I asked some easy questions to the sales agent who contacted me. He was more interested in selling me the cloud database than answering my questions. I learned that I am too small for Oracle. I continued with CentOS and Postgres. Google provides me with all the information I need. I could not get all the information I needed from Oracle.

New infrastructure for Oracle databases: deployment of several database environments with no issues and in a faster way.

Speed, simplicity, and performance. Every DBA and OS administrator like the simple fact that one RPM install can be very powerful and can configure all prereqs.

Simplicity and agility to deploy Oracle Database configurations with a single RPM and custom tweaks on Oracle Linux kernel.

Improve performance and integration with new releases without affecting the ability and stability of the system.

We are using the firewall for intrusion prevention, and it has performed well.

It prevents intrusions. It does what it says it will do quite accurately.

It does intrusion protection.

It does not have any reporting on attempted intrusions.

It is perfectly stable; no downtime.

We have scaled it. It is sufficient to meet our needs.

I have not used technical support.

We were not previously using another solution.

The initial setup was a bit complicated, as there are a lot of pieces that you have to look at.

The pricing and licensing are good.

Nobody else was on the shortlist. We chose it because we were using it.

Most important criteria when selecting a vendor: We look at the characteristics of the vendor, then see if they suit us.

Fujitsu's Oracle/Intel platform has been specifically designed with Oracle in mind using Oracle VM, Oracle Linux, for our customers wanting to use Oracle product, applications, databases. We've designed it in a way that we get the best possible performance from the applications and databases on our engineered system.

What it's allowed us to do, initially, it allowed us to develop an Intel platform specifically for Oracle. What's most important for us, where it comes across is the licensing. It's very difficult - sometimes you can build a platform that is optimal, but when you apply Oracle licenses across that platform, it isn't the most economical. All of our Intel platform for this has been optimized towards which Oracle solutions are going to be running on it, to get both the best performance but also that will be economical for our customers.

Because it's specifically built for Oracle, with Oracle applications and solutions in mind, we have standard pricing, a standard way of working, a standard cost for each organization. That allows us to save time, on both bid and, once new requirements come along for each organization, we know exactly what it takes to add to that solution, to add to that platform. The saving for us is, we can feed back quickly to grow, respond to new requirements.

With Oracle Linux Ksplice specifically, we have organizations looking for minimum downtime. We're able to apply hot-patching at any time; once we've proven they're tested, ready to go, we don't need to take downtime to apply them.

We have a shared services platform with multiple organizations set on it. So planning downtime across all those organizations becomes more and more difficult. The more organizations we get onto the platform, the less "white space" is available. Ksplice allows us to do hot-patching without the downtime. That, for us, is quite key.

Also, the virtualization, Oracle VM, allows us to get the best performance for our Oracle applications and database solutions. We know it's proven to be more performant with Oracle applications, so we get the best performance out of it on our platform.

What we found in moving from Oracle Linux 6 to Oracle Linux 7 was the whole interfacing with the application and the fact that operating had all changed, all the commands had changed. You need to be aware that there is some kind of training, some kind of handover required for your technical guys, understanding different ways of interacting with it. Bear that in mind.

What we experienced is, the stability is key. What we can't take into account with customers is how they're going to want to use the platform, once we've installed it, once we've got different solutions running on that platform.

We have a use case of a shared platform where we have one large organization set on our Intel platform. The virtualization then allows us to grow out for when we get more and more organizations on.

We've just added another huge organization, DHL, they are now set on that shared platform along with another organization. That hasn't impacted it in the least. We are able to scale out and scale with that organization. That organization itself, that specific program, could grow and grow. So it allows us that flexibility to grow that whichever way. If that organization's business case grows and becomes bigger and bigger, the platform can scale out to that.

It also allows us to add in more organizations on the same platform with one overview of managing. For us, as an organization we can manage it from a single point with multiple organizations using it, with no impact on each other.

We don't have any problems with Oracle technical support. Our guys can normally resolve most of the issues themselves, but where we do require further help, we have direct contact with Oracle, and the turnaround is what we'd expect.

There is a gap for the type of Intel platform we're now providing, from an Oracle perspective. For a lot of the platforms we have our own cloud at Fujitsu, our K5, which is not geared towards Oracle specifically, because of the licensing implications. So we knew there was a requirement for a quick, economical, engineered system, so that the customers can either sit in their own datacenters or we'll place it in our datacenters and manage the service that way.

With Oracle VM and Oracle Linux, it then allows us to scale up, scale out as and when the customers want, their requirements grow, their enterprises grow. Or the requirements change over time; it could be an easy path for them to move from on-premise to cloud, or they may want to bring the cloud, themselves, on-premise.

It's the perfect step for them, if they're not quite ready to move to the cloud - they might never want to go to the cloud, but they want to control security, data, data integrity. All the features they're after as an organization - they may want to go one way, they may not want to go the other way. This fits that platform at that point.

For us to work with any vendor, it's the support and ongoing roadmap with that vendor. We need to understand where it's going, where it's going to end up in the next one to two years, as well as then three to four years. We also need to be able to work closely with them to almost guide that roadmap from our experience, and be able to have input into it as well. That is key with any partner and vendor.

The key for us with our engineered systems is specifically how quick and easy it is to "plug in and play," with a solution. We got the platform in place within a couple of weeks and then another week or so to get everything up and running with the virtualizations, and then the Oracle Linux with all the solutions and applications on top of that.

End-to-end it will take us three to five weeks, depending on the install.

We use our in house expertise at Fujitsu.

As per above, pay attention to how Oracle license their products and make sure you are clear as to the implications of choosing products which can have a significant impact on license cost and supportability. 

We were driven to some part by how the cost of licensing of Oracle databases and needed to ensure the most cost effective way to do this, so really OVM was the only option for us .

I am the Oracle practice CTO. I work for Fujitsu. We cover all the aspects of IT, for enterprise, for infrastructure, through to applications and managed services. I work for the Business Applications Services, we cover anything around enterprise solutions, enterprise architecture, anything that will aid them in their business process. In my role at Fujitsu I oversee all of the Oracle architects, so any solution owners from infrastructure to applications, and all the bits in between. All architects and solution owners report to me.

In the context of, if you're wanting to use the Oracle workloads, absolutely, this is the way you need to go. For non-Oracle workloads, again, no problems with that at all. From Fujitsu's point of view, and where it sits on our Intel platform, this is a no-brainer. We specifically built it with Oracle in mind. Therefore, using Oracle VM and Oracle Linux was the way forward.

If that's the way you're going, if you're looking to use Oracle applications, Oracle Databases, I would definitely recommend using the OVM and Oracle Linux.

It performs perfectly for what we require it to do. There are, obviously, certain issues that have been highlighted in the next version. That's not the product itself, that's just the usability of it. We would rate the Oracle OVM, the Oracle Linux, eight to nine out of 10.

Healthcare IT.

Stable Oracle deployments.

Easy Oracle integration: Oracle preinstall packages available on YUM.

Kernel updates.

Fujitsu uses OVM and Oracle Linux extensively with Fujitsu hardware to provide our customers with a certified, scalable, pre-tested platform to run Oracle Applications and technology.

The use of a pre-defined and certified architecture that combines a market leading Fujitsu server architecture along with Oracle's virtualisation technology and Linux operating system, has allowed Fujitsu to reduce the cost and time to create customer bids.

Optimising the use of Oracle licenses on behalf of our customers to get most efficient cost of ownership.

Fujitsu works closely with Oracle's product development team to provide feedback on areas for improvement on behalf of our customers.

In the early days of using the product, Fujitsu had some issues with both the stability and functionality of the product. Oracle was very responsive to this feedback and, as a result, has incorporated many of the ideas that were shared by Fujitsu. The result is a market-leading solution.

No issues with scalability.

Excellent. Oracle also encouraged a direct working relationship with the product development team, which has served to help in getting support.

Work with Fujitsu to learn from our experience. We are able to provide a pre-certified architecture that incorporates the best of both Oracle and Fujitsu technology to provide a highly scalable platform for Oracle workloads.

Ksplice is a really cool feature. The availability is maximized because it can be patched without downtime. Oracle Linux provides free Ksplice.

Productivity has improved as it is easier to deploy and use. In particular, various open source packages can be more easily installed and managed, and systematically maintained.

Ksplice is a really cool feature. The availability is maximized because it can be patched without downtime. Oracle Linux provides free Ksplice.

I hope you have a built-in package to visualize your performance and analysis tools.

I can install and use the open-source tools, but I hope to use the proven packages.

We have not encountered any stability issues. It was very stable. Safety is at the highest level and there has never been a problem.

We have not encountered any scalability issues; scalability was also very satisfactory.

Customer Service:

Customer service is wonderful.

Technical Support:

Technical support is very skilled and stable. However, in Korea, 24-hour call service is only available in English.

I switched some of my UNIX systems to x86. As a result of the U2L project, I chose the OS as Linux.

The installation was very simple. Installation was quick and easy with a few clicks.

We used to use UNIX. And I'm using some Red Hat and CentOS. I switched some of CentOS to Oracle Linux. We do it directly. We have many engineers with various levels of experience.

• Service continuity through zero downtime and low-cost subscription

Oracle Linux is provided by Oracle subscription and is equivalent or better than other Linux technical support. However, support costs are about half that level.

Before choosing this product, we also considered Red Hat and SUSE. However, we chose Oracle Linux to use the Oracle kernel optimized for Oracle applications.

If you convert UNIX to Linux... and if stability and service downtime are to be minimized, Oracle Linux is the solution.

Docker containers allow you to deploy a lot of workloads at scale. If you can think of the old batch jobs, what they do is allow us to deploy that into the cloud so we can elastically expand or contract and only pay for what we use. I think that's part of it.

You have high availability, so you can run your containers in multiple availability zones. They're very cost-effective, much more cost-effective than writing your own scripts. And they're layered so they're very lightweight; they don't consume quite as many resources as how we would traditionally deploy this.

I would really like to see more frequent releases and I would like to see a very lightweight kernel for embedded systems. I'm really anticipating the new Oracle Database XE as it relates to Oracle Linux because now I can run that in production, and that was just announced as well.

It's young, so I think it's fair that they have some work to do. A little bit more variability, the ability to expand, take advantage of bare metal. I think that that's really going to be a key as they grow.

The Oracle container engine is actually built on the Enterprise Linux kernel. So I think it's extremely stable and secure. I think it's one of the most stable and secure Linux variants in the world. When we actually build our Docker containers we utilize Oracle Linux as a basis for those as well.

I think what's great about it is if you're a small customer you can install Oracle Linux, just pull it down off of the site, update it. And if you're a large customer you can go with the unbreakable kernel. You can run that on on-premise, and when you go to the cloud which, of course, is scaled out to literally millions of nodes, it's the basis for all of Oracle's cloud.

What's great is having the Oracle Linux, also having Oracle products. You've really goy one vendor to call. Some people like to say one neck to choke but I say one hand to high-five.

You might have a development team that kind of runs off and goes rogue and installs several different operating systems. Or they've assembled a Docker container and deployed it in the Cloud under the guise of microservices. The first time you have a hiccup with that, or the credit card doesn't process and you don't know where to find the code or the developer, I think at that point you really need to put some controls in place.

When it comes to important criteria when selecting a vendor, I think experience is there, but they really have to care. They have to own the problems; I think owning your client's problems is the number one thing.