commit 0b40c15e266f55c907b0ccba7e58046c91a475fc
parent 48138d0986aa809d5a0255d6f4722bbc6183caaa
Author: Oliver Lowe <o@olowe.co>
Date: Thu, 15 Jun 2023 12:12:38 +1000
Add HTTP request counter
Diffstat:
| A | bin/hits | | | 18 | ++++++++++++++++++ |
| A | man/hits.1 | | | 45 | +++++++++++++++++++++++++++++++++++++++++++++ |
2 files changed, 63 insertions(+), 0 deletions(-)
diff --git a/bin/hits b/bin/hits
@@ -0,0 +1,18 @@
+#!/usr/bin/awk -f
+
+# skip requests with bad method
+<UNKNOWN> {
+ next
+}
+
+{
+ host = $1
+ path = host $8
+ hits[path]++
+}
+
+END {
+ for (path in hits) {
+ printf "%d %s\n", hits[path], path
+ }
+}
diff --git a/man/hits.1 b/man/hits.1
@@ -0,0 +1,45 @@
+.Dd
+.Dt HITS 1
+.Sh NAME
+.Nm hits
+.Nd count web traffic
+.Sh SYNOPSIS
+.Nm
+.Op Ar
+.Sh DESCRIPTION
+.Nm
+counts HTTP requests from web server request log files
+and reports the number of requests per path.
+If no files are specified,
+.Nm
+reads from the standard input.
+.Pp
+The log must be in the default format produced by
+OpenBSD's
+.Xr httpd 8
+and
+.Xr nginx 8 .
+Here is an example line:
+.Dl default 192.0.2.1 - - [19/Dec/2022:23:12:34 +1100] "GET /some/file.txt HTTP/1.1" 200 256
+.Sh EXIT STATUS
+.Ex
+.Sh EXAMPLES
+Print the most-requested paths
+from the latest access log:
+.Bd -literal -offset indent
+hits /var/www/logs/access.log | sort | tail
+.Ed
+.Pp
+Print the least-requested blog entries from 2021:
+.Bd -literal -offset indent
+grep 'blog/2021/' /var/www/logs/access.log | hits | sort | sed 10q
+.Ed
+.Pp
+Print the 20 most-requested paths
+from all access logs:
+.Bd -literal -offset indent
+cd /var/www/logs
+gunzip -c access*.gz | cat access.log | hits | sort -r | sed 20q
+.Ed
+.Sh SEE ALSO
+.Xr awk 1
