On Sun, Jun 18, 2017 at 09:59:43PM -0400, David Woodstuck wrote:
Hi there,
> I have a host Nginx server running in port: 9000, This Nginx will proxy
> http://www.myserver.com:10085/. Some pages from
> http://www.myserver.com:10085/ have a lot of iframes whose srcs are
>
On Wed, Jun 21, 2017 at 03:48:22PM -0400, Olaf van der Spek wrote:
Hi there,
> > proxy_pass http://backend;
>
> Is this an error in the docs? I think the http:// here isn't right.
proxy_pass can speak http or https. The config above is the way you tell
it which one to speak to this
On Wed, Jun 21, 2017 at 03:26:10PM +0800, George / wrote:
Hi there,
For the question in the Subject: line, see
http://nginx.org/en/docs/http/request_processing.html#mixed_name_ip_based_servers
> i am running nginx version: nginx/1.12.0.i got following server block
> config as below, all
Hello,
https://nginx.org/r/ssl_session_ticket_key mention session ticket key rotation.
Which process read these files? master or worker?
Must it be readable for root only or nginx-user?
Must I signal nginx processes the rotation? If yes, how? via SIGHUP?
thanks for clarification,
Andreas
That'll be it then. I'd added cache control headers.
Thanks for the swift reply. Mass edit coming up!
Steve
On 22/06/17 08:34, Reinis Rozitis wrote:
The solution was to also add that header to the location{} block that
I use to manage the relevant static resources.
This seems rather
The solution was to also add that header to the location{} block that I
use to manage the relevant static resources.
This seems rather strange. Is it supposed to work this way?
If your location blocks contain any add_header directives then those comming
from server {} block are ignored (and
Hi folks,
As a precaution against CORS, I
add_header Access-Control-Allow-Origin *;
outside any location{} block in my server{} definition.
I have recently had a problem where I deliver .css via a CDN, and that
CDN references font files also on the CDN, and this was triggering CORS,
so
> proxy_pass http://backend;
Is this an error in the docs? I think the http:// here isn't right.
http://nginx.org/en/docs/http/ngx_http_upstream_module.html
Posted at Nginx Forum:
https://forum.nginx.org/read.php?2,275028,275038#msg-275038
___
nginx
I'm sending 403 responses now, so I screwed up by mistaking the fields
in the logs. I'm going back to lurking mode again with my tail
shamefully between my legs.
This code in the image location section will block the google app:
if ($http_user_agent ~*
Hi Maxim,
Thank so much for your response. You are correct, and the configuration of
proxy_cache_use_stale updating was present.
Stale while revalidate is a great option, and we use in a websites with more
than 50 millions users/month.
This graph is our time to waiting to delivery pages
So, the user will never know what happened on the backend (just get 415).
Get it, thank you very much!
-- Original --
From: "Maxim Dounin";;
Send time: Thursday, Jun 22, 2017 1:24 AM
To: "nginx-devel";
Subject: Re:
Hi,
On Tuesday, Jun 20, 2017 11:01 PM +0300, Maxim Dounin wrote:
>On Tue, Jun 20, 2017 at 10:39:44AM +0800, 胡聪 (hucc) wrote:
>
>> >416 will be returned when the request has no Range HEADER and the target
>> >file is
>> >empty. Apparently, it does not conform to the HTTP protocol. Empty file
Hello!
On Thu, Jun 22, 2017 at 01:05:37AM +0800, 胡聪 (hucc) wrote:
> Hi,
>
> On Wednesday, Jun 21, 2017 1:38 AM +0300, Maxim Dounin wrote:
>
> >When the response is read from a static file, not proxied, and/or
> >proxied with "proxy_method GET;", an empty response means exacly
> >that: an empty
Hi,
On Wednesday, Jun 21, 2017 1:38 AM +0300, Maxim Dounin wrote:
>When the response is read from a static file, not proxied, and/or
>proxied with "proxy_method GET;", an empty response means exacly
>that: an empty response, and returning 415 is perfectly correct.
>
>The ngx_buf_size() test will
Hello!
On Wed, Jun 21, 2017 at 12:09:17PM -0400, Olaf van der Spek wrote:
> What does it take to enable KeepAlive for FastCGI upstream servers?
> I've set upstream { keepalive 99; } and location { fastcgi_keep_conn on; ]
> but nginx is still closing the connection after each request.
With
Hello!
On Wed, Jun 21, 2017 at 04:51:55PM +0100, Steven Hartland wrote:
> We're seeing an 502 bad gateway responses to client on an nginx load
> balanced upstream due to "no live upstreams".
>
> The upstream in question has 2 servers defined with default settings
> running over https
What does it take to enable KeepAlive for FastCGI upstream servers?
I've set upstream { keepalive 99; } and location { fastcgi_keep_conn on; ]
but nginx is still closing the connection after each request.
Posted at Nginx Forum:
https://forum.nginx.org/read.php?2,275028,275028#msg-275028
We're seeing an 502 bad gateway responses to client on an nginx load
balanced upstream due to "no live upstreams".
The upstream in question has 2 servers defined with default settings
running over https (proxy_pass https://myupstream).
When this happens we see "no live upstreams while
Thank you Jim
Best Regards
Rebaca
Posted at Nginx Forum:
https://forum.nginx.org/read.php?2,275020,275026#msg-275026
___
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx
Thanks a lot for a very quick response
Regards,
Rebaca
Posted at Nginx Forum:
https://forum.nginx.org/read.php?2,275020,275025#msg-275025
___
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx
Hi,
Refer to http://nginx.org/en/docs/http/ngx_http_upstream_module.html#server ,
if all primary server are unavailable, backup server will handle request.
I have two question?
1. What's the meaning of unavailable?
2. When will primary server come back, after fail_timeout?
Thanks,
Linbo
Hello,
On 06/21/2017 09:31 AM, rebaca wrote:
Hi
I am trying to experiment with purging cache content in Nginx server. The
Nginx I have is a free version (Not Nginx Plus version).
As per one of the document from Nginx website, in order to enable purging on
Nginx, I need to use the directive -
Hi,
On Wed, Jun 21, 2017 at 09:31:03AM -0400, rebaca wrote:
> Hi
>
> I am trying to experiment with purging cache content in Nginx server. The
> Nginx I have is a free version (Not Nginx Plus version).
>
> As per one of the document from Nginx website, in order to enable purging on
> Nginx, I
В логи заглядывал еще до создания данной темы.
лог настроен следующим образом
> error_log /var/log/nginx/error.log notice;
При этом до недавнего времени в логе проскакивали только сообщения
> [alert] 5092#5092: send() failed (90: Message too long)
А теперь еще появились
> [crit] 5093#5093:
Hi
I am trying to experiment with purging cache content in Nginx server. The
Nginx I have is a free version (Not Nginx Plus version).
As per one of the document from Nginx website, in order to enable purging on
Nginx, I need to use the directive - proxy_cache_purge.
But when I try to add this
Hi Reinis,
Thanks a lot for quick and insightful response.
From,
Rebaca
Posted at Nginx Forum:
https://forum.nginx.org/read.php?2,275008,275019#msg-275019
___
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx
Добрый день, товарищи.
Не сведущ в теме, потому вопрос может звучать глупо.
Не сталкивался ли кто-то с примерами использования NGINX в виде клиента для
Google Cloud Pub/Sub? Как lua-resty-redis для pub/sub в Redis, к примеру.
Любая помощь приветствуется. Заранее спасибо.
--
Sincerely,
Dmitry
Currently NGINX supports static file caching wherein if the file is
present
in the location (derived from the config), then it will serve the client
directly. Else it just intimates the client that the file is not present.
There is no capability to forward the request to origin server, get the
>
> Вам же порекомендовали не использовать proxy_temp_path на другом разделе
Так так и сделал, почитал доку и приписал параметр use_temp_path=off к
своему конфигу.
Вот цитата из доки.
> Поэтому лучше, если кэш будет находиться на той же файловой системе, что и
> каталог с временными файлами.
21 июня 2017 г., 11:09 пользователь Иван Мишин
написал:
> Максим, кеш дира /tmp/ram/ была забита на 100% 28Гб из 28Гб. Сбросил
> часть кеша, получилось 25Гб занято из 28Гб. Затем исправил то о чем ты
> говорил и начальный конфиг (приведенный в первом письме) стал
Максим, кеш дира /tmp/ram/ была забита на 100% 28Гб из 28Гб. Сбросил часть
кеша, получилось 25Гб занято из 28Гб. Затем исправил то о чем ты говорил и
начальный конфиг (приведенный в первом письме) стал выглядеть вот так:
> proxy_cache_path /tmp/ram/ levels=1:2 use_temp_path=off
>
Hi all,
i am running nginx version: nginx/1.12.0.i got following server block config
as below, all request match regular expression work well,but request to server
s01.example.com return 404.what's wrong? i googled for a while,most of the
article said,it first try to match literal string
Actually I think I was mistaken and the field is the user agent. I will
change the variable and see what happens. I did some experiments to
show the pattern match works.
On Tue, 20 Jun 2017 20:56:46 -0700
li...@lazygranch.com wrote:
> I want to block by referrer. I provided a more "normal"
BACKGROUND:
---
Currently NGINX supports static file caching wherein if the file is present
in the location (derived from the config), then it will serve the client
directly. Else it just intimates the client that the file is not present.
There is no capability to forward the
34 matches
Mail list logo