Slashdot Mirror
Building a Linux Virtual Server
prostoalex writes "NewsForge (an OSTG site) has an article that discusses the details of building a cluster for a virtual server - an environment where one can add and remove machines as needed, to account for high-traffic Web site of intensive database application: 'The main advantage of using LVS is that unlike Microsoft network load-balancing clusters, the LVS allows you to add a node running any operating system that supports TCP/IP to the cluster.'"
First of all, the Linux Virtual Server project is a misnomer, because a Virtual Server these days means a virtualized operating environment similar to what is provided by VMWare, Xen, Linux Vserver, etc.
Second, IP Load Balancing is not new and is much better done by a hardware device such as Cisco CSM, Foundry ServerIron, Nortel Alteon, etc. These things boot in seconds, use ASICs, can process orders of magnitudes more packets per second than a Linux box can, have no moving parts (except for the fans), are aesily configured, can be set up in a fail-over configuration, speak routing protocols, etc, etc.
The Linux Virtual Server is a good project to tinker around with if you would like to understand load-balancing concepts better and do not have the $$$ for a hardware device, but I would not even consider designing a mission-critical solution using LVS instead of a hardware load-balancer.
I remember back in my ISP days we purchased a small provider with a few T1's all running on Linux-based routers. They had a tendency to lock up and were very awkward to manage for many reasons (e.g. upgrading the kernel was a risky procedure that required way more downtime than necessary) and we replaced them with real routers pretty quickly. I kinda suspect that Linux Virtual Server may be a similar type of experience - PC's just don't make good network devices for some reason.
Of course I'll probably get flamed for this comment...
Great for http. Bad for nfs and smb/cifs.
Looking at these images (LVS and WLBS) it seems with LVS there is a single point of failure with the load balancer box. With Windows load balancing there is no "balancer box" and if any one server fails the cluster continues without missing a beat.
How does LVS work if the balancer box fails?
Unlike most of the posters sharing their 'wisdom' on the merits (or lack thereof) of using PC's for networking ... I can say that I recently setup a load balancer using LVS in front of my Web cluster (of 6 dual Opteron's). It rocks!
:)
You can easily avoid it becoming a single point of failure by using 2 directors with realtime syncronization and failover. Keepalived works well in this regard, giving subsecond failover performance, and even uses VRRP, Cisco's own hot standby protocol.
As for the level hardware that you need, a low end or older PC, such as a P3-700 for instance, can easily saturate a 100 mbit connection.
I used 2 x P4 2.4G systems. So for less than a fraction of the cost of even 1 hardware based boxes, I get hot standby and should be good for way more throughput than I can afford to provide.
The best feature that LVS brings to the table that the hardware based implementations miss out on, is flexibility. This software is being enhanced all the time and nothing is impossible with it. The same cannot be said for the hardware solutions.
Unfortuneately the article does not touch on any of the nice add-ons, like keepalived, ldirectord, etc., that make LVS a manageable solution. I know that they are trying to keep the examples simple, but showing servers being configured using solely ipvsadm is not too realistic for the average person and make it look undeservedly complicated, which it is not.
Give LVS a look an see if it might not give you the features you need without paying the big iron prices.
DC
Here's another approach: put Linux on a CompactFlash card instead of a hard drive. Keep the filesystem mounted read-only for normal operation. Test upgrades on a different computer and CF card. Upgrade by swapping out CF cards. If you can build a PC that doesn't need fans, then you've removed all moving parts from the equation. For smaller installations, single-board computers such as the Soekris are very reliable. For larger installations, you can usually find a 1U system with the level of performance you need.
It was always just so much simpler to get a more expensive Cisco box in the long run due to its design, documentation, an performance.
It always depends on what you're doing, of course, and also what you're familiar with, but for my routing/firewalling/VPN/load balancing/ etc needs I've simply found Linux to be more flexible.