On 2019-05-07 23:26, Brendan Doyle wrote:
> I'm trying to get a basic tcp load balancer working on OL :
> ...
> # nmap -p 5000 10.129.87.162
This is probably not working as you expect because the default scan for
nmap is a SYN scan -- since the TCP handshake is not complete, why would
nginx connec
Hi,
I'm trying to get a basic tcp load balancer working on OL :
cat /etc/oracle-release
Oracle Linux Server release 7.5
My config is very basic:
stream {
upstream backend_stream {
server 10.129.87.160:5000;
server 10.129.87.120:5000;
}
s
Hi team,
Hope you all are doing well.
I need some assistance regarding nginx configuration. I am using nginx service
as proxy_pass to other URL's
When I am performing load testing on nginx server with high no. of requests.
For few tests it is running fine, But after some time it shows below up
Hello!
On Mon, May 06, 2019 at 09:30:33PM +, Yuhao Zhang wrote:
> Hi, Thank you for the explanation. I definitely need to learn
> more about the protocol spec. Would you help me understand why
> the "proxy_buffer_size" affects the result even when
> "proxy_buffering" is off?
This is becau
Hi,
Thanks for the response. We fixed the issue related to casting and still
didnt see any change in behavior where intermittently garbage values are
being appended.
However, when we tried to get the ip address using alternate means, as
follows the issue is not happening and Valgrind is also not r
On Tue, May 07, 2019 at 03:21:49PM +1000, Duke Dougal wrote:
Hi there,
> Well you hit the mark thank you well done.
Good that you found the fix -- and that your nginx config was correct
all along.
> The problem was that I needed to wrap the entire curl url in quotes.
Interesting -- had you tes
Patrick, thank you. Mapping works perfect.
Today one of my developers ask me that he'll append ?domain=st when he
wants to see the staging site and ?domain=www when he wants to see the
production while testing end the end of the url. here is examples:
request goes to staging if the domain=st para
