6

How to queue in lighttpd java and php requests

view full story
linux-howto

http://stackoverflow.com – We have a scenario as mentioned 1.Main Load Balancer 2.Data1 Load Balancer 3.Data2 Load Balancer From 1. request goes to 2. and data is fetched and that fetched data is sent to 3. where with some JAVA process reports are generated. All the 3 laod balancer are having lighty (lighttpd). Current Policy to generate mass data reports -- > manually booting machines under 2 and 3. Target to achieve -- > To make queue on 2 and 3 via lightty so that reports are generated without any error and without any JAVA process going into defunct state. (HowTos)

fooubuntu's picture
Submitted by fooubuntu on 12/17/2012 – Made popular on 12/17/2012

Stories similar to How to queue in lighttpd java and php requests

We have a aws load balancer. The load balancer uses the TCP connections 443 to encrypt the data.

Since the load balancer encrypts the IP of the client, in our apache log file we can see only the IP of the load balancer rather then the client. Even though we have used %{X-Forwarded-For}i in our log format.

Is there any way to handle this issue?

I have heard people say "oh that server is off the load balancer so you can run that expensive script on it".

What implications does a server off the load balancer have? Fundamental to answering this, I understand load balancing, but I don't know what the dynamics of a system with 5 servers (4 on the load balancer) has.

Thanks

We are using aws load-balancer. There is one machine which is behind the load balancer. So for this reason we have configured apache2.conf such that %h are replaced by %{X-Forwarded-For}i in our file.

But something wired is happening. We can't see any ip of the client in the logs section.

But when we change back from %{X-Forwarded-For}i to %h all client ip's are visible in the log section.

When ever I try to load test a site that is load balanced on EC2 I get the following message:

ab -n 20 -c 20 http://www.somesite.com/

Benchmarking www.somesite.com (be patient)...Send request failed! apr_socket_recv: Connection reset by peer (54)

Any sites that are hosted on EC2 without the load balancer can be load tested successfully, so I figure it's not related to the security group I have

The stickiness of an elastic load balancer can be set to one of three options

Disable Stickiness Enable Load Balancer Generated Cookie Stickiness Enable Application Generated Cookie Stickiness

What are the differences between 2) and 3)?

If I set up a load balancer on aws that terminates ssl at the load balancer, then the requests being sent to the application server is in plain text. That has definite security implications. How do folks mitigate this implication in practice?

If alternatively, I set up https traffic between load balancer and the application servers also, what are the performance implications?

I run apache2 on Ubuntu as a caching load-balancing reverse proxy in front of a group of application servers.

We're in the process of getting Passenger Enterprise purchased and setup, but in the meantime we'd like to get some pseudo-"rolling restarts" setup, by issuing commands on each server to sleep X seconds and then restart passenger when we deploy.

I have a service where servers upload 20kb xml files to my server.

There is no session, it is a single POST request and that's it.