We have 5 physical servers. All are identical. 16 gb ram, 4×146 10k sas in raid 10. So, we need high availability. Current plan is following: 2 boxes with Heartbeat for load balancing. At this moment nginx is configured as balancer. then problem is web & mysql. I plan to set up 2 boxes running mysql and nginx. Between theses boxes set up mysql replication in master <-> master. Then add latest box which has [...]

I’m building an application that requires high-availability. We currently run: 4 webservers, 2 varnish servers (both with Pound for SSL proxy), 3 Cassandra nodes. The varnish servers are used as entry points for the system. We want to configure everything that Varnish A always works, until failure, then Varnish B should receive the requests and distribute them to the webservers. As Varnish A and Varnish B are not in the same DC and some other [...]

We have a GlusterFS cluster we use for our processing function. We want to get Windows integrated into it, but are having some trouble figuring out how to avoid the single-point-of-failure that is a Samba server serving a GlusterFS volume. Our file-flow works like this: Files are read by a Linux processing node. The files are processed. Results (can be small, can be quite large) are written back to the GlusterFS volume as they’re done. [...]

I’ve been reading up on OpenStack and how we can re-create an EC2/S3-style cloud for our internal development and I’m having a hard time finding information on how the OpenStack cloud controller provides redundancy of the cloud management services. I know I can setup multiple Swift and Nova nodes, but not a single document/article/howto/wiki contains information on: a) what happens if the cloud controller node dies; and b) how to setup redundant cloud controllers. It [...]

The advice below applies to Apache, but it seems generic enough to apply regardless of system. My architecture is 3 web servers behind a load balancer. My IIS servers currently have keep-alive enabled with the default connection timeout of 2 minutes. Would disabling my site’s keep-alive increase performance and throughput? First, disable keep-alive. This is the nastiest thing against performance. It was designed at a time sites were running NCSA httpd forked off inetd at [...]

I have a problem when the mail server can’t access the shared storage and then the emails start piling up in the mail queue. Now I would like postfix to try and send the emails to the secondary mail server if it can’t do the delivery locally. Is it possible? I don’t know if posfix even supports something like this, because if the domain is in my_destinations then the delivery is local and can’t be [...]

For a project I have the task of planning a high availability setup for a web shop and CMS system. However, of course the project is on a tight budget. So a high end solution might not be in the budget. There will be two machines running web server (CMS, shop), one machine running the database, and one machine for running a fax server needed for delivering orders to partners. All systems run Linux. All [...]

For our SaaS app, I want to have a system in place for a datacenter wide failure. In the main datacentre, we have two servers set up with MSSQL Database Mirroring (synchronised). This provides us with a good enough high availability solution for server failures. If a server dies, it fails over automatically (with the help of a 3rd witness server) within seconds. I was thinking of using MSSQL Replication or Log Shipping from the [...]

I found out that proxmox doc says a lot abou Fencing. I haven’t found anything about it in Ganeti doc. Also I screeched Ganeti Google Group Archive and I found similar question but answer was that Split Brain is not a problem in Ganeti and questioner should read archive. So I wonder if Ganeti solved in some creative way or just ignored the problem? I’m especially interested in setups with only two nodes. Asked by [...]

I would like to setup a two nodes PROXMOX cluster with DRBD storage. The nodes will be connected by gigabit ethernet. What would be the best storage setup in terms of reliability, performance and costs electiveness? My favorite would be RAID 10, but my colleague argues that RAID 0 would be sufficient since DRBD works as RAID 1. I also heard RAID 1 is good option. Asked by Maciek Sawicki “What will be the best [...]