Describe the ipng_stats_logtail UDP feature in debian/control alongside the per-VIP / per-device counter description, so the package metadata reflects what the module actually ships. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
50 lines
2.0 KiB
Plaintext
50 lines
2.0 KiB
Plaintext
# SPDX-License-Identifier: Apache-2.0
|
|
Source: nginx-ipng-stats-plugin
|
|
Section: httpd
|
|
Priority: optional
|
|
Maintainer: Pim van Pelt <pim@ipng.ch>
|
|
Build-Depends:
|
|
debhelper-compat (= 13),
|
|
nginx-dev,
|
|
libpcre2-dev,
|
|
zlib1g-dev,
|
|
libssl-dev
|
|
Standards-Version: 4.6.2
|
|
Homepage: https://git.ipng.ch/ipng/nginx-ipng-stats-plugin
|
|
Rules-Requires-Root: no
|
|
|
|
Package: libnginx-mod-http-ipng-stats
|
|
Architecture: any
|
|
Depends:
|
|
${misc:Depends},
|
|
${shlibs:Depends},
|
|
nginx (>= 1.26~)
|
|
Description: nginx dynamic module for per-VIP, per-device traffic counters
|
|
This package ships ngx_http_ipng_stats_module as a dynamic module
|
|
loadable into stock upstream nginx on Debian. The module records
|
|
per-VIP HTTP request counters, status code distribution, bytes in
|
|
and out, and request-duration histograms, and attributes each
|
|
request to the interface it arrived on. Counters are exposed as
|
|
Prometheus text and JSON from a single scrape endpoint.
|
|
.
|
|
Attribution is done by reading the ingress ifindex per connection
|
|
from the kernel's IP_PKTINFO / IPV6_PKTINFO cmsg; listening sockets
|
|
stay plain wildcards so outgoing packets follow the normal routing
|
|
table (which matters for DSR / maglev setups). The module adds
|
|
device= and ipng_source_tag= parameters to the nginx listen
|
|
directive, mapping interface names to source tags.
|
|
.
|
|
The module also provides an ipng_stats_logtail http-level directive
|
|
that ships every request as a UDP datagram to a collector, using a
|
|
named nginx log_format (so $ipng_source_tag and $server_addr are
|
|
available alongside the standard access-log variables). Each worker
|
|
holds a private in-memory buffer flushed on size or timer, over a
|
|
per-worker SOCK_DGRAM socket resolved once at configuration time.
|
|
Delivery is fire-and-forget — zero disk I/O, no backpressure — and
|
|
an if=$variable parameter suppresses selected requests (e.g. health
|
|
checks) before the format is rendered.
|
|
.
|
|
Typical use cases include GRE tunnel fleets, VLAN trunks, or any
|
|
deployment where traffic arrives on distinct interfaces and
|
|
per-interface observability is needed.
|