Lighttpd

Here is my patch for many servers. It just uses size twice for inodes.

Yet it dosnet remove problem when modification dates differs a little. (You can delete size->mtab, but last-modified will still be bad).

#!patch
--- ./src/etag.c.orig      2006-05-24 22:48:05.000000000 +0200
+++ ./src/etag.c   2006-05-24 22:48:44.000000000 +0200
@@ -9,9 +9,7 @@
 }

 int etag_create(buffer *etag, struct stat *st) {
-       buffer_copy_off_t(etag, st->st_ino);
-       buffer_append_string_len(etag, CONST_STR_LEN("-"));
-       buffer_append_off_t(etag, st->st_size);
+       buffer_copy_off_t(etag, st->st_size);
        buffer_append_string_len(etag, CONST_STR_LEN("-"));
        buffer_append_long(etag, st->st_mtime);

-- Marek Pu?czy?ski

Looking forward to this new Option...

I have many Servers Running Lighty in a cluster and the etags are different..

Sorta Hurts..

--Running Lightly

This would be helpful in my setup also.

I lost this when I move from apache to lighttpd.
(Noone other than me has noticed so far but i'm going to have to explain it to the boss at somepoint)
An I would rather not have to apply a patch to get it working.

Thanks.

-Dan

-- Dan <dmarkham

I have created a patch in ticket 1209 which provides for various options in enabling/disabling different components for etag generation as well as enabling/disabling etag generation globally

http://trac.lighttpd.net/trac/ticket/1209

The options are

etag.use-inode = "enable"|"disable"
etag.use-mtime = "enable"|"disable"
etag.use-size = "enable"|"disable"
static-file.etags = "enable"|"disable"

The patch is relative to 1.4.15. Would appreciate if anyone could test the patch and provide feedback in ticket 1209. Test patch-version4. It contains all the options

Thanks,

-- yusufg