mirrors/haproxy
1
0
Fork 0
mirror of https://git.haproxy.org/git/haproxy.git/ synced 2025-09-22 22:31:28 +02:00
Code Issues Projects Releases Wiki Activity

MINOR: http: add the "base32" pattern fetch function

Browse Source 
This returns a 32-bit hash of the value returned by the "base"
fetch method above. This is useful to track per-URL activity on
high traffic sites without having to store all URLs. Instead a
shorter hash is stored, saving a lot of memory. The output type
is an unsigned integer.
This commit is contained in:
Willy Tarreau 2012-12-09 13:38:54 +01:00
parent 2406db4b39
commit ab1f7b72fb
2 changed files with 56 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
doc/configuration.txt
View File

@ -9316,6 +9316,12 @@ The list of currently supported pattern fetch functions is the following :
table also allows one to collect statistics about most commonly
requested objects by host/path.
base32 This returns a 32-bit hash of the value returned by the "base"
fetch method above. This is useful to track per-URL activity on
high traffic sites without having to store all URLs. Instead a
shorter hash is stored, saving a lot of memory. The output type
is an unsigned integer.
src This is the source IPv4 address of the client of the session.
It is of type IPv4 and works on both IPv4 and IPv6 tables.
On IPv6 tables, IPv4 address is mapped to its IPv6 equivalent,

50
src/proto_http.c
View File

@ -8489,6 +8489,55 @@ smp_fetch_base(struct proxy *px, struct session *l4, void *l7, unsigned int opt,
return 1;
}
/* This produces a 32-bit hash of the concatenation of the first occurrence of
* the Host header followed by the path component if it begins with a slash ('/').
* This means that '*' will not be added, resulting in exactly the first Host
* entry. If no Host header is found, then the path is used. The resulting value
* is hashed using the url hash followed by a full avalanche hash and provides a
* 32-bit integer value. This fetch is useful for tracking per-URL activity on
* high-traffic sites without having to store whole paths.
*/
static int
smp_fetch_base32(struct proxy *px, struct session *l4, void *l7, unsigned int opt,
const struct arg *args, struct sample *smp)
{
struct http_txn *txn = l7;
struct hdr_ctx ctx;
unsigned int hash = 0;
char *ptr, *beg, *end;
int len;
CHECK_HTTP_MESSAGE_FIRST();
ctx.idx = 0;
if (http_find_header2("Host", 4, txn->req.chn->buf->p + txn->req.sol, &txn->hdr_idx, &ctx)) {
/* OK we have the header value in ctx.line+ctx.val for ctx.vlen bytes */
ptr = ctx.line + ctx.val;
len = ctx.vlen;
while (len--)
hash = *(ptr++) + (hash << 6) + (hash << 16) - hash;
}
/* now retrieve the path */
end = txn->req.chn->buf->p + txn->req.sol + txn->req.sl.rq.u + txn->req.sl.rq.u_l;
beg = http_get_path(txn);
if (!beg)
beg = end;
for (ptr = beg; ptr < end && *ptr != '?'; ptr++);
if (beg < ptr && *beg == '/') {
while (beg < ptr)
hash = *(beg++) + (hash << 6) + (hash << 16) - hash;
}
hash = full_hash(hash);
smp->type = SMP_T_UINT;
smp->data.uint = hash;
smp->flags = SMP_F_VOL_1ST;
return 1;
}
static int
acl_fetch_proto_http(struct proxy *px, struct session *l4, void *l7, unsigned int opt,
const struct arg *args, struct sample *smp)
@ -9110,6 +9159,7 @@ static struct acl_kw_list acl_kws = {{ },{
static struct sample_fetch_kw_list sample_fetch_keywords = {{ },{
{ "hdr", smp_fetch_hdr, ARG2(1,STR,SINT), val_hdr, SMP_T_CSTR, SMP_CAP_L7|SMP_CAP_REQ },
{ "base", smp_fetch_base, 0, NULL, SMP_T_CSTR, SMP_CAP_L7|SMP_CAP_REQ },
{ "base32", smp_fetch_base32, 0, NULL, SMP_T_UINT, SMP_CAP_L7|SMP_CAP_REQ },
{ "path", smp_fetch_path, 0, NULL, SMP_T_CSTR, SMP_CAP_L7|SMP_CAP_REQ },
{ "url", smp_fetch_url, 0, NULL, SMP_T_CSTR, SMP_CAP_L7|SMP_CAP_REQ },
{ "url_ip", smp_fetch_url_ip, 0, NULL, SMP_T_IPV4, SMP_CAP_L7|SMP_CAP_REQ },