Nginx auto apeend "@@@Accept-Encoding:gzip" in request url

some kind of problems happened at my nginx servers.
1.The error log shows like this:
2009/10/21 18:43:30 [error] 22636#0: *8345 open()
“//mnt/iscsi/che168www/file1/Upload/photo/small/20091021/80_60_514555811.jpg@@@Accept-Encoding:gzip”
failed
(2: No such file or directory), client: 192.168.0.183, server:
file1.che168.com, request: “GET
/Upload/photo/small/20091021/80_60_514555811.jpg@@@Accept-Enc
oding:gzip HTTP/1.0”, host: “file1.che168.com”, referrer:
http://www.che168.com/picture/564/bjf/184568.htm
2009/10/21 18:43:52 [error] 22635#0: *8700 open()
“/mnt/iscsi/che168www/publish/zhuanti/i30china/index.php@@@Accept-Encoding:gzip”
failed (2: No such file or
directory), client: 192.168.0.181, server: www.che168.com, request:
“GET /zhuanti/i30china/index.php@@@Accept-Encoding:gzip HTTP/1.0”, host:
"www.che168.com
"

2.nginx version:
/usr/local/webserver/nginx/sbin/nginx -V
nginx version: nginx/0.8.20
built by gcc 4.1.2 20071124 (Red Hat 4.1.2-42)
configure arguments: --user=www --group=www
–prefix=/usr/local/webserver/nginx

3.and here is part of the configuration file:
default_type application/octet-stream;

charset gbk;

server_names_hash_bucket_size 128;
client_header_buffer_size 32k;
large_client_header_buffers 4 32k;
client_max_body_size 8m;

sendfile on;
tcp_nopush on;

keepalive_timeout 60;
server_tokens off;

access_log off;

proxy_intercept_errors on;
proxy_redirect off;
tcp_nodelay on;

ssi on;
#ssi_silent_errors on;
ssi_types text/shtml text/plain application/x-javascript text/css
application/xml application/javascript;

fastcgi_connect_timeout 300;
fastcgi_send_timeout 300;
fastcgi_read_timeout 300;
fastcgi_buffer_size 64k;
fastcgi_buffers 4 64k;
fastcgi_busy_buffers_size 128k;
fastcgi_temp_file_write_size 128k;

gzip on;
gzip_min_length 0k;
gzip_buffers 8 32k;
gzip_http_version 1.0;
gzip_comp_level 9;
gzip_types text/plain application/x-javascript text/css
application/xml application/javascript;
gzip_vary on;
gzip_proxied any;
gzip_disable “MSIE [1-6].”;

4.we use squid(squid/3.0.STABLE15) in front of nginx servers.

5.the same problem exists in version 0.7.62

Is this a bug or just a misconfiguration error?

On Wed, Oct 21, 2009 at 12:55:06PM +0200, Yu Sun wrote:

2009/10/21 18:43:52 [error] 22635#0: *8700 open()
built by gcc 4.1.2 20071124 (Red Hat 4.1.2-42)
large_client_header_buffers 4 32k;
proxy_intercept_errors on;
fastcgi_read_timeout 300;
gzip_types text/plain application/x-javascript text/css

Is this a bug or just a misconfiguration error?

In error_log message the request field is a client’s original request
line,
so this “GET /zhuanti/i30china/index.php@@@Accept-Encoding:gzip
HTTP/1.0”
is what nginx was asked for. It was not added by nginx itself.

Igor S. wrote:

On Wed, Oct 21, 2009 at 12:55:06PM +0200, Yu Sun wrote:

2009/10/21 18:43:52 [error] 22635#0: *8700 open()
built by gcc 4.1.2 20071124 (Red Hat 4.1.2-42)
large_client_header_buffers 4 32k;
proxy_intercept_errors on;
fastcgi_read_timeout 300;
gzip_types text/plain application/x-javascript text/css

Is this a bug or just a misconfiguration error?

In error_log message the request field is a client’s original request
line,
so this “GET /zhuanti/i30china/index.php@@@Accept-Encoding:gzip
HTTP/1.0”
is what nginx was asked for. It was not added by nginx itself.

Yes, this is not caused by nginx.
I added a rewrite rule and fix it without finding why:
rewrite ^(.*)@@@Accept-Encoding:gzip$ $1;