Gluster in production.. stories from the front line
Hi all,
I am Interested to know if anyone here is using gluster in a production environment, and what you are using it for specifically?
I have personally used it in the past to solve a problem using a distributed replicated volume and it was extremely well received at that site (after it was implemented!), but working in other environments I am finding it difficult to convince others of its benefits.
I personally think it is an excellent fit in many scenarios and am currently looking at potentially replacing NFS in some instances with native glusterfs.. so was wondering if anyone has headed down this path and what their experiences have been?
Thanks in advance.
Responses
Hi,
I am going to share with my experience with RHS storage (glusterfs), I have been used glusterfs storage for one year, it's not comparable to any traditional solution such as NFS.
I implemented glusterfs on four projects that require high intensity read and write files, redundancy, high availability.
We used to deploy distribute-replica method, until this moment we didn't face any issues regarding performance or high availability.
Now we move to RHS storage, it easy to configure from console and create bricks without any problems, the greatest thing that using nagios server to monitor everything related to bricks, volume domain, services, and servers's health.
Hi,
First of all, we tried to use NFS as a share storage to serve the clients, but we considered the high availability aspects, performance, security, of course might these in some how achieved and implemented.
Due to our policy to have a backup of the data, which was easy to handle by glusterfs, and use acl was straightforward to allow or prevent which client to access.
Add to that the if we need scale and extend our existence storage without facing any issues special if we do that online without shutdown the services.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
