50% of Failed requests with load balance

Server NGINX
1

I’m trying to use nginx for load balancing in my server.
Actually I have 2 servers in the same network with Apache listening the
port 8080 (http://IP:8080/ it’s works!). In one of them i install nginx
to
send 50% of requests to one server and 50% to other. The configuration
looks like:

upstream mysite{
server 10.10.14.64:8080;
server 127.0.0.1:8080;
}

server {
server_name 10.10.14.66;

pass all else onto apache waiting at localhost:8080

location / {
proxy_pass http://mysite;
}
}

When i run ‘‘ab -c 100 -n 1000 http://10.10.14.66/prueba.php’’, where
prueba.php only does ‘‘phpinfo()’’. The result show:
…
Complete requests: 1000
Failed requests: 499
(Connect: 0, Receive: 0, Length: 499, Exceptions: 0)
…
But if i run the same benchmark individually, there are 0 Fail requests.
Just commenting one server or the other from upstream nginx
configuration.

If i try with a html or jpg file don’t get fail requests. But the times
don’t improve with the 2 servers. If i does the test directly to port
8080 for each machine, i get 5-6 seg average, but if i does to nginx
with load balance, i get 7-8 seg average. It’s that reasonable?

Can anyone help me?

2

Complete requests: 1000
Failed requests: 499
(Connect: 0, Receive: 0, Length: 499, Exceptions: 0)

It’s just the content-length error and nothing else - ‘ab’ kinda
measures
only the first response and if the following requests differ in size its
considered as error - loadbalancing between 2 servers where <?
phpinfo(); ?>
has different output will lead to ab showing that half requests have
failed.

In general phpinfo() is kinda quite slow for testing - I would suggest
to
test something more generic like <? echo 'hello world'; ?>

If i try with a html or jpg file don’t get fail requests. But the times
don’t improve with the 2 servers. If i does the test directly to port
8080 for each machine, i get 5-6 seg average, but if i does to nginx
with load balance, i get 7-8 seg average. It’s that reasonable?

First - its good to run ‘ab’ from different box than the webserver
itself as
its quite a resource hog and can slow down the webserver itself
(especially
if the box isnt very fast) and then compare the results.

Second - no matter how you look at it a loadbalancer is extra layer
between
client and server so if each server can push 10 requests per second it
doesnt always mean you will see 20 req/sec speed from the balancer.

rr