New answers tagged

2

No, this is not possible. Haproxy does not understand the FTP protocol, and therefore can't do things like re-write the FTP commands and responses that include the true IP address of the FTP server. If that's a requirement, then you'll have to find a different proxy solution.


0

This was actually something I had missed – linkerd supports HTTP/1.1 out of the box and nginx was using 1.0 to forward to the linkerd service. Adding proxy_http_version 1.1 Fixed the issue for me.


0

This should be pretty easy. You just need create server blocks for each subdomain, get an SSL certificate for the subdomain (unless the main cert is a wildcard cert), and copy the applicable location blocks in with a couple of tweaks. Something like this should work, though I haven't tested it so it will almost certainly need some tweaking. server { ...


0

From what I can tell mod_proxy_html only really understands HTML 4.01 and XHTML 1.0 based on https://httpd.apache.org/docs/2.4/mod/mod_proxy_html.html#proxyhtmldoctype. Searching whether a <table> is allowed within an <a> turns up http://stackoverflow.com/questions/6013943/table-tag-inside-anchor-tag amongst other answers which suggests that it'...


0

This should pretty much "just work" with nginx using upstreams and, optionally, nginx reloads. I'm not sure what you mean by "specific to a connection" in this context, but making some assumptions about your setup... Assuming the upload point application will be called "uploadpoint" and that initially it is going to point1 and that point2 is what you want ...


0

Ended up with this table: **Nginx**: ngx_http_sub_module --> http://p.defau.lt/?A6UukYFKZrYLalJ5KOuTKg **Nginx**: Nginx + lua. LUA body_filter --> http://p.defau.lt/?0ZxLWENoEkFIRD17QlEtbQ **Apache**: Apache Module mod_substitute --> http://p.defau.lt/?IS8u2mnAQTSHs4qEl_MSFQ **Apache**: As of version 2.3 LUA can be used as a scripting language ...


0

Fire up another VPS somewhere, install forward proxy software, and then configure your application to use that server as a web proxy. Keep in mind that open proxies are bad, so make sure to put in place firewall rules such that only your server can make requests to the proxy.


0

The directive ProxyPass "/server3" "http://192.168.1.3:80" does not need the default port number 80 for HTTP and I would add a trailing slash: ProxyPass "/server3/" "http://192.168.1.3/. As of the the 2.4 release the manual recommends the functionally identical syntax with the ProxyPass directive enclosed in a <Location> directive: <VirtualHost *...


0

There is not such thing as "Last Modified doesn't work". This header simply indicates the date resource was last modified. The clients may issue "if not modified since" requests afterwards, but this is not important to your situation. It is natural that subsequent requests will return exactly the same headers. Objects in Varnish are cached including the ...


2

So I have found a solution. I remove all newlines in the certificate and send them as a single HTTP Header from the proxy to the back-end, similar as explained here: https://forum.nginx.org/read.php?2,236546,236546 In my back-end I reconstruct the certificate by adding a newline every 64 characters. The updated code for the reverse proxy is the following ...


1

I see 2 solutions here: Enable LUA scripting and reformat the string in a no-end-of-line format. Use the Set Misc plugin and use base64_encode on that variable.


0

If you don't want to go the "hardware route", have a look at this. In this answer I've described how one could / would implement such a setup using software.


0

I had some problem, I resolved in this way: in my wp-config.php, I added these lines: if (strpos($_SERVER['HTTP_X_FORWARDED_PROTO'], 'https') !== false) $_SERVER['HTTPS']='on'; BEFORE everything in the code. Hope it helps!


1

Nginx will always create new proxy connection for incoming connection unless it is cached localy. As MJPG is an infinite loop, it doesn't seem as an option here. What I would however try is setup a ffmpeg/vlc on the proxy server to consume mjpg stream from rpi. Nginx then can connect to local ffmpeg/vlc to get stream. You will then have only one connection ...


0

Yes, the way you have things configured will essentially load balance all incoming connections between the backend servers. What you want is basically one backend definition for each real web server. I think the gist of what you want is multiple backend definitions, one for each real web server. Then in your frontend section, you'd want to use various ...


2

today, nginx Supports ssl pass thru: https://serversforhackers.com/tcp-load-balancing-with-nginx-ssl-pass-thru


1

The following configuration works fine on my server. proxy_buffers 16 16k; proxy_buffer_size 16k;


0

I found a solution. Here it is: location = /do/uploads.php { proxy_pass http://domain/switch/do; proxy_set_header Host $host; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Proto $scheme; proxy_set_header X-Real-IP $remote_addr; }


0

I was able to solve the first issue as well. The reason for this odd behavior was caused because the reverse proxy dropped the query part of the request (everything behind the ?). This again was caused by misconfigured proxy_pass lines. The solution to this was to remove /$uri from both proxy_pass lines in the configuration file (details see http://nginx....


0

The typical config is that your users/visitors browse to http://www.example.com.com/sonar and then they get the content reverse proxied from http://sonar.example.com:9000/. <VirtualHost *:80> ServerName www.example.com ServerAlias example.com ProxyPass /sonar http://sonar.example.com:9000/ ProxyPassReverse /sonar http://sonar.example....


1

it won't work with https:// extentions no matter in what way I try to set it Your <VirtualHost *:80> is only matching port 80 traffic, so it cannot work with "https:// extentions." Whatever problem there might be with your proxy setup, your first issue is with your virtual host setup. EDIT per OP comment to answer: what I want to do is redirect ...


1

Not sure about your environment, just as a hint: Is the machine in question publicly reachable? If so, implement authentication / authorization, or use a VPN. If this is valid, how can I fix the error "Too many open files"? Probably, because it's listening to that many ports, your nginx, runs against limits, specifically the ones dealing with file ...


0

I had this problem and it wasn't due to DNS since I was not using hostnames, it was the MaxClient and related settings that got it fixed in httpd.conf: <IfModule prefork.c> StartServers 20 MinSpareServers 15 MaxSpareServers 100 ServerLimit 2000 MaxClients 1000 MaxRequestsPerChild 10000 </IfModule> <IfModule worker.c> ...


0

I was able to solve the second of the issues (error 1006) by changing some of the configuration options. As the other problem still persists I cannot confirm that the NoVNC console works from within the Horizon UI but when I call the NoVNC URL directly I get a connection and can interact with the instance. This is the working solution till now: ...


0

you can achieve that using the unix socket commands. Take a look at point 9.2 of the management.txt doc file. There are third party tools that use this 'api' like haproxyctl, but I do not have any experience with them.


1

The problem is your second nginx instance that's trying to serve files locally: location / { try_files $uri $uri/ =404; } Just remove that whole thing if you want to serve using http://server2.com/app/. If you want to use http://server2.com/ instead, update the config to reflect that: listen [::]:80 default_server ipv6only=on; ...


0

It was a mod rewrite thing RewriteRule ^ - [E=HTTP_AUTHORIZATION:%{HTTP:Authorization}] The above rule needs to be applied at the proxy.


1

DerfK is right about dynamic content at the full page level, you absolutely don't want cookie leakage to occur. However you can do more caching at the layers upstream - with WordPress specifically, it has a decent pluggable object cache. This means, that while the page itself won't be cached, you can store much of the database load within a ephemeral store ...


3

In general caching works poorly with dynamic content, especially with content that contains things like "recently viewed" that will change every singe page load. If you really want to use caching then the first step is to disable that plug in so that there's a theoretical chance that someone might be served the same page twice. The second step is to only ...



Top 50 recent answers are included