Solved

Load balancing Squid and Web Servers with Squid

Posted on 2010-09-23
9
693 Views
Last Modified: 2012-06-27
Greetings,

We have 2 squid servers (each one with a public IP) and 2 web servers.


We have 2 DNS records published so users will load balance across the 2 squid servers, but how do I load balance the squid servers across the 2 web servers?

for example, users go to www.test.com.

they arrive at either cache1 or cache2.

How do I tell cache1 to load balance between web1 and web2? same with cache2?

We are doing this through multiple DNS entries which our local DNS server respond with both web server IP addresses; but what happens if one web server goes down?

How can I tell squid to load balance across multiple web servers for the same domain and if one of the web servers go down to easily take it of the configuration? We have implemented round robin DNS but the problem is that if a server goes down DNS won't know about it and even if we remove it from DNS the cache server will have that IP address in dns cache


What's the better way to do this?
0
Comment
Question by:ddiazp
  • 5
  • 4
9 Comments
 
LVL 5

Accepted Solution

by:
mooodiecr earned 500 total points
ID: 33748873
Squid has a similar round -robin technique you can try.  We use this very successfully for an extremely busy website.  In the squid configuration file add:

http_port 1.2.3.4.:80 defaultsite=www.thefraggle.com vhost
cache_peer 1.2.3.5 parent 80 0 no-query originserver round-robin
cache_peer 1.2.3.6 parent 80 0 no-query originserver round-robin

This can be placed at the very top of the squid config file.
0
 
LVL 5

Expert Comment

by:mooodiecr
ID: 33748896
0
 
LVL 10

Author Comment

by:ddiazp
ID: 33749031
We already have something similar:


http_port 80 accel defaultsite=www.test.com vhost
cache_peer www1.test.com parent 80 0 no-query no-digest originserver name=cache_www default

where www1.test.com is our other cache server. Is this wrong?



Or do we need to remove:

cache_peer www1.test.com parent 80 0 no-query no-digest originserver name=cache_www default

and add:

cache_peer web1.test.com parent 80 0 no-query no-digest originserver name=cache_www default
cache_peer web2.test.com parent 80 0 no-query no-digest originserver name=cache_www default
0
Free learning courses: Active Directory Deep Dive

Get a firm grasp on your IT environment when you learn Active Directory best practices with Veeam! Watch all, or choose any amount, of this three-part webinar series to improve your skills. From the basics to virtualization and backup, we got you covered.

 
LVL 5

Expert Comment

by:mooodiecr
ID: 33749290
That is correct:
"
Or do we need to remove:

cache_peer www1.test.com parent 80 0 no-query no-digest originserver name=cache_www default

and add:

cache_peer web1.test.com parent 80 0 no-query no-digest originserver name=cache_www default
cache_peer web2.test.com parent 80 0 no-query no-digest originserver name=cache_www default
"

the port and accel entry is the site that is to be used and then there needs to be two separate entries for the load balancing sites.  Just to note the IP address or DNS name can resolve locally if needed.  i.e. web2.test.com can read 127.0.0.1
0
 
LVL 10

Author Comment

by:ddiazp
ID: 33749365
Awesome, one last question.

What will Squid do in the event that one of the webX.test.com is down or cannot contact it?
0
 
LVL 5

Expert Comment

by:mooodiecr
ID: 33754927
Squid tracks the reachability status of its neighbour caches. When using ICP, or any other protocol you choose, Squid marks a neighbour as down when the neighbour fails to reply to 20 consecutive ICP queries. As soon as a reply is received again, the neighbour is marked up.

In addition to ICP, Squid also monitors TCP connections. When a TCP connection to a neighbour fails, Squid marks the neighbour down and begins a process to periodically (every 80 seconds) retry a diagnostic TCP connection. When the diagnostic connection succeeds, the neighbour is marked up.

So, in order for Squid to consider a neighbour cache as up, these two conditions must be true:

   1. The last TCP connection must have been successful.
   2. There must be fewer than 20 consecutive unacknowledged ICP queries.
0
 
LVL 10

Author Comment

by:ddiazp
ID: 33755180
You say cache neighbors. Are these the actual webservers? We'll implement this and give it a shot.

Thanks
0
 
LVL 5

Expert Comment

by:mooodiecr
ID: 33756140
In your scenario it would be whatever site is being served/cached so yes.
0
 
LVL 10

Author Closing Comment

by:ddiazp
ID: 34403583
This worked, forgot about closing this ages ago!
0

Featured Post

Ransomware: The New Cyber Threat & How to Stop It

This infographic explains ransomware, type of malware that blocks access to your files or your systems and holds them hostage until a ransom is paid. It also examines the different types of ransomware and explains what you can do to thwart this sinister online threat.  

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

I will assume you are running a non-server version of some sort of Windows throughout this article. There are many flavors of Windows since Windows Server 2000 - 2008, XP Home & Pro, Vista Home & Pro, and Windows 7 Starter, Home, Pro, Ultimate, etc.…
It is possible to boost certain documents at query time in Solr. Query time boosting can be a powerful resource for finding the most relevant and "best" content. Of course the more information you index, the more fields you will be able to use for y…
In a recent question (https://www.experts-exchange.com/questions/29004105/Run-AutoHotkey-script-directly-from-Notepad.html) here at Experts Exchange, a member asked how to run an AutoHotkey script (.AHK) directly from Notepad++ (aka NPP). This video…

856 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question