Re: Can stream module proxy grpc traffics transparently?

Hi, Thanks for the quick reply. Yes, I know that ngx_http_grpc_module already supports grpc proxy. the fact is that we have a legacy product running with a pretty old nginx version and hope that it can be used to proxy grpc as well, so, can stream module proxy the grpc request/response c

Can stream module proxy grpc traffics transparently?

If yes and what are the drawbacks comparing with the grpc support by http module? thanks, Allen Posted at Nginx Forum: https://forum.nginx.org/read.php?2,293868,293868#msg-293868 ___ nginx mailing list -- nginx@nginx.org To unsubscribe send an email

How to configure nginx to support multipart/form-data upload

Hi, Can somebody show the minimal configuration to suppport large file upload with multipart/form-data Content-Type? When I upload 8GB file with the following global configs, I will always get 600s timeout with the error: upstream timed out (110: Connection timed out) while reading response header

Re: How nginx stream module reuse tcp connections?

Hi, Thanks for the reply! So, if there is no error and the downstream/upstream didn't actively close the connection, the nginx won't timeout and close the tcp connection with the downstream or with the upstream at all? is that correct? and I suppose the SO_KEEPALIVE option is turn on by default

How nginx stream module reuse tcp connections?

Hi, As we know there are some keepalive options in the nginx http modules to reuse tcp connections, But are there corresponding options in the nginx stream module to achieve the same? How nginx persist tcp connection with downstream? How nginx persist tcp connection with upstream? What is the "s

Re: Keepalived Connections Reset after reloading the configuration (HUP Signal)

Looking into the big loop code, it may happen that the worker process may close the keepalive connection before consuming any pending read events? for ( ;; ) { if (ngx_exiting) { if (ngx_event_no_timers_left() == NGX_OK) { ngx_log_error(NGX_LOG_NOTICE, cycl

Re: Keepalived Connections Reset after reloading the configuration (HUP Signal)

Hi Maxim Dounin, Is it possible that the nginx is closing the keepalive connection while there is input data queued? As we know If a stream socket is closed when there is input data queued, the TCP connection is reset rather than being cleanly closed. Br, Allen Posted at Nginx Forum: https://f

how nginx handle websocket proxying

Can someone elaborate this a little bit? "NGINX supports WebSocket by allowing a tunnel to be set up between both client and back-end servers." what is the "tunnel" here? Does it mean the client will talk with the back-end server directly after the http Upgrade handshakes? Posted at Nginx Forum:

how to enable non root user to execute nginx reload

A non root process needs to signal reload to nginx master (as root) without sudo I've tried using setcap and setpriv with CAP_KILL, both not work. # getcap nginx/sbin/nginx nginx/sbin/nginx = cap_kill+ip #su user01 -s /bin/sh -c 'nginx/sbin/nginx -s reload' nginx: [alert] kill(68, 1) failed (1:

Re: Worker process core dumped

Hi, I found most times using "r" after ngx_http_free_request() won't have any problem. the core dump would happen once for a while in the high load. I am wondering if the "ngx.pfree" does not return the memory back to the os when it's called? Posted at Nginx Forum: https://forum.nginx.org/read.p

Re: Worker process core dumped

to be more self assurance, can somebody confirm that the "r" is no longer accessable after this? ngx_http_free_request(r, 0); thank you in advance! Posted at Nginx Forum: https://forum.nginx.org/read.php?2,289296,289299#msg-289299 ___

Re: Worker process core dumped

I was wrong, the request object was created on the fly with the pool object. here the pool was detroyed before the r was referenced which caused the core dump. Posted at Nginx Forum: https://forum.nginx.org/read.php?2,289296,289298#msg-289298 ___ nginx

Worker process core dumped

version: 1.17.8 debug log: 2020/09/03 14:09:21 [error] 320#320: *873195 upstream timed out (110: Connection timed out) wh

Re: How nignx handled the header to the client in this situation

Hi Francis, Thanks for the reply! w.r.t. the "http://nginx.org/r/proxy_buffering";, the doc does not mention if the buffering works for header, body or both, I'm wondering if nginx can postpone the sending of upstream header in any ways? otherwise the client will get wrong status code in this case

Re: How nignx handled the header to the client in this situation

Will nginx buffer the header before receiving of the whole body? If not, what if error happens in the middle of body receiving? nginx has no chance to resend the error status then. Posted at Nginx Forum: https://forum.nginx.org/read.php?2,287990,288039#msg-288039

How nignx handled the header to the client in this situation

I understand the nginx would proxy the header first and then the body, in the case the connection with the upstream is broken during the transfer of body, what status code the client would get? since the nginx would proxy the 200 OK from upstream first to the client, but will nginx send another 5xx

Re: Multiple master processes per reloading

Patrick Wrote: --- > On 2019-07-07 22:39, allenhe wrote: > > Per my understanding, the reloading would only replace the old > workers with > > new ones, while during testing (constantly reloading), I found the > output

Multiple master processes per reloading

Hi, Per my understanding, the reloading would only replace the old workers with new ones, while during testing (constantly reloading), I found the output of "ps -ef" shows multiple masters and shutting down workers which would fade away very quickly, so I guess the master process may undergo the s

what does "deny 0.0.0.1;" do

Hi, I found this is valid, and want to know what scenario it's used for. deny 0.0.0.1; Thanks, Allen Posted at Nginx Forum: https://forum.nginx.org/read.php?2,283857,283857#msg-283857 ___ nginx mailing list nginx@nginx.org http://mailman.nginx.org/

Re: Worker other than the master is listening to the socket

Nginx version: 1.13.6.1 1) In our use case, the Nginx is reloaded constantly. you will see lots worker process hanging at "nginx: worker process is shutting down" after couple days: 58 root 0:00 nginx: master process ./openresty/nginx/sbin/nginx -p /opt/applicatio 1029 nobody 0:2

Re: nginx has problem sending request body to upstream

I see. so in this case the request was completely sent in single write without blocking that there is no need to schedule a write timer anymore, otherwise it is necessary. Thanks for the explanations! b.t.w, have you ever seen the work process is listening on the socket? Posted at Nginx Forum:

Re: nginx has problem sending request body to upstream

I understand the connection establish timer, write timer and read timer should be setup and removed in order, but where is the write timer? are there lines in the logs saying I'm going to send the bytes, the sending is on-going, and the bytes has been sent out? Posted at Nginx Forum: https://foru

Re: nginx has problem sending request body to upstream

but it looks to me the timer was set for the write not the read, also the subsequent message isn't telling the nginx was interrupted while sending the request? Posted at Nginx Forum: https://forum.nginx.org/read.php?2,283735,283739#msg-283739 ___ nginx

nginx has problem sending request body to upstream

Hi, The Nginx hangs at proxying request body to the upstream server but with no error indicating what's happening until the client close the font-end connection. can somebody here help give me any clue? following is the debug log snippet: 2019/04/12 14:49:38 [debug] 92#92: *405 epoll add connecti

Nginx didn't try the next available backend server

Hi, My Nginx is configured with: proxy_next_upstream error timeout http_429 http_503; But I find it won't try the next available upstream server with the following error returned: 2019/04/05 20:11:41 [error] 85#85: *4903418 recv() failed (104: Connection reset by peer) while reading response hea

Worker other than the master is listening to the socket

Hi, I understand it is the master process listen to the binding socket since that's what I see from the netstat output in most time: tcp0 0 0.0.0.0:28002 0.0.0.0:* LISTEN 12990/nginx: master while sometimes I found the worker process also doing the same t