General

Profile

manik

  • Registered on: 2008-11-26
  • Last connection: 2010-02-19

Issues

Activity

2009-02-05

19:14 Lighttpd Support: RE: Following Tutorial: Can't bind to port: 80 Permission denied
i think on the mac, i am not sure, you need to be root to do this not just an admin user.
19:04 Lighttpd Development: RE: Connection: Close with big FastCGI payload
according the HTTP/1.0 and HTTP/1.1 spec, the "Connection" is a hop-by-hop header and not an end-to-end header. hop-b...

2009-02-04

23:12 Lighttpd Development: RE: Connection: Close with big FastCGI payload
i have noticed that lighty doesn't strip hop-by-hop headers. i know little about fastcgi and related functionality, b...

2008-12-17

21:37 Lighttpd Development: RE: Can't send back to client in proxy plugin
add the following line to your code
before you send out the request to the origin server you need to set the corre...

2008-12-10

21:19 Lighttpd Support: RE: many seemingly useless connections reported in status page
i just checked the code, it seems that after the response has been sent to the client, function connection_reset is c...
20:28 Lighttpd Development: RE: joblist_append(srv, con), what is that?
in the case of mod_proxy, you need to call joblist_append whenever you need to have some information propagated to th...
20:13 Lighttpd Development: RE: CPU Imbalance with multiple lighty workers
the point that i am asserting here is the following :
i have multiple processes that are listening on the same por...

2008-12-08

15:00 Lighttpd Support: RE: lighttpd + lighttpd-angel process "gone"
its probably a crash, don't have the core dump ?
>- lighttpd hit "number of open files" limit on OS
>- lighttpd h...
14:57 Lighttpd Development: RE: CPU Imbalance with multiple lighty workers

>That's not true. Suppose one client sends all requests serially. If the server uses the first idle backend, it'll ...

2008-12-06

13:53 Lighttpd Development: RE: CPU Imbalance with multiple lighty workers
>> I assume the web server will prefer an idle backend, so it shouldn't affect response times (in a negative way).
...

Also available in: Atom