NGINX and 100-Continue

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,

Yes, I read this. I responded with new question.
I tried to use tengine, but my problem was not solved.
I can't find solution. 

You don't have this problem in your case?
I'm using boto (client for S3), at the end when file
is uploaded at 100%, boto hangs after command PUT /[file] 100, example:

IP_REMOVED - - [29/May/2014:11:27:53 +0000] "PUT /foo.bar HTTP/1.1" 100 0 "-" "Boto/2.27.0 Python/2.7.6 Linux/3.13.0-24-generic"

Transfer ending when boto get timeout.
When I switch to Apache there are no problem with that. 

Some sources from Intenrert report that this might be involved with mod fastcgi (fcgi can't handle this, only fastcgi can handle http 100-Continue). I can't find how this might be correlated with fastcgi in nginx.

Michael



Don't use nginx.  The current version buffers all the uploads to the local disk, which causes all sorts of problems with radosgw (timeouts, clock skew errors, etc).  Use tengine instead (or apache).  I sent the mailing list some info on tengine a couple weeks ago.

On 5/29/2014 6:11 AM, Michael Lukzak wrote:
NGINX and 100-Continue Hi,

I have a question about Nginx and 100-Continue.
If I use client like boto or Cyberduck all works fine,
but when I want to upload file on 100% upload, progress bar
hangs and after about 30s Cyberduck reports that 
HTTP 100-Continue timeouted.

I use nginx v1.4.1
Only when I use Apache 2 with fastcgi all works fine.

Anyone has a solution?

My nginx config is, works with socket from RADOS.

server {
       listen 80;
       server_name cli.rados *.cli.rados;
        server_tokens off;

       client_max_body_size 1g;

       location / {
               fastcgi_pass_header Authorization;
               fastcgi_pass_request_headers on;
                       if ($request_method = PUT ) {
                               rewrite ^ /PUT$request_uri;
                       }
               include fastcgi_params;
               client_max_body_size 0;
               fastcgi_busy_buffers_size 512k;
               fastcgi_buffer_size 512k;
               fastcgi_buffers 16 512k;
               fastcgi_read_timeout 2s;
               fastcgi_send_timeout 1s;
               fastcgi_connect_timeout 1s;
               fastcgi_next_upstream error timeout http_500 http_503;

               fastcgi_pass unix:/var/run/radosgw.sock;
       }

       location /PUT/ {
               internal;
               fastcgi_pass_header Authorization;
               fastcgi_pass_request_headers on;
               include fastcgi_params;
               client_max_body_size 0;
               fastcgi_param CONTENT_LENGTH $content_length;
               fastcgi_busy_buffers_size 512k;
               fastcgi_buffer_size 512k;
               fastcgi_buffers 16 512k;

               fastcgi_pass unix:/var/run/radosgw.sock;
       }
# end of http
}

Best Regards,
Michael 

_______________________________________________
ceph-users mailing list
ceph-users at lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20140529/3004f71b/attachment.htm>


[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux