Update RI spec to better-specify statistics
ref: http://zzz.i2p/topics/3515
This commit is contained in:
@ -1,7 +1,7 @@
|
||||
{% extends "global/layout.html" %}
|
||||
{% block title %}{% trans %}The Network Database{% endtrans %}{% endblock %}
|
||||
{% block lastupdated %}{% trans %}August 2019{% endtrans %}{% endblock %}
|
||||
{% block accuratefor %}0.9.42{% endblock %}
|
||||
{% block lastupdated %}2023-01{% endblock %}
|
||||
{% block accuratefor %}0.9.57{% endblock %}
|
||||
{% block content %}
|
||||
<h2>{% trans %}Overview{% endtrans %}</h2>
|
||||
|
||||
@ -116,16 +116,60 @@ Current statistics are limited to:
|
||||
<li>{% trans %}1 hour average number of participating tunnels{% endtrans %}
|
||||
</ul>
|
||||
|
||||
<p>{% trans -%}
|
||||
Floodfill routers publish additional data on the number of entries in their network database.
|
||||
{%- endtrans %}</p>
|
||||
<p>
|
||||
These are optional, but if included, help analysis of network-wide performance.
|
||||
As of API 0.9.58, these statistics are simplified and standardized, as follows:
|
||||
</p>
|
||||
<ul>
|
||||
<li>Option keys are stat_(statname).(statperiod)
|
||||
<li>Option values are ';' -separated
|
||||
<li>Stats for event counts or normalized percentages use the 4th value;
|
||||
the first three values are unused but must be present
|
||||
<li>Stats for average values use the 1st value, and no ';' separator is required
|
||||
<li>For equal weighting of all routers in stats analysis,
|
||||
and for additional anonymity,
|
||||
routers should include these stats only after an uptime of one hour or more,
|
||||
and only one time every 16 times that the RI is published.
|
||||
</ul>
|
||||
|
||||
<p>
|
||||
Example:
|
||||
<pre>
|
||||
stat_tunnel.buildExploratoryExpire.60m = 0;0;0;53.14
|
||||
stat_tunnel.buildExploratoryReject.60m = 0;0;0;15.51
|
||||
stat_tunnel.buildExploratorySuccess.60m = 0;0;0;31.35
|
||||
stat_tunnel.participatingTunnels.60m = 289.20
|
||||
</pre>
|
||||
</p>
|
||||
|
||||
|
||||
|
||||
<p>{% trans -%}
|
||||
The data published can be seen in the router's user interface,
|
||||
but is not used or trusted within the router.
|
||||
As the network has matured, we have gradually removed most of the published
|
||||
statistics to improve anonymity, and we plan to remove more in future releases.
|
||||
Floodfill routers may publish additional data on the number of entries in their network database.
|
||||
These are optional, but if included, help analysis of network-wide performance.
|
||||
{%- endtrans %}</p>
|
||||
<p>
|
||||
The following two options should be included by floodfill routers in every published RI:
|
||||
</p>
|
||||
<ul>
|
||||
<li><b>netdb.knownLeaseSets</b>
|
||||
<li><b>netdb.knownRouters</b>
|
||||
</ul>
|
||||
<p>
|
||||
Example:
|
||||
<pre>
|
||||
netdb.knownLeaseSets = 158
|
||||
netdb.knownRouters = 11374
|
||||
</pre>
|
||||
</p>
|
||||
|
||||
|
||||
|
||||
|
||||
<p>
|
||||
The data published can be seen in the router's user interface,
|
||||
but is not used or trusted by any other router.
|
||||
</p>
|
||||
|
||||
<h3>{% trans %}Family Options{% endtrans %}</h3>
|
||||
|
||||
|
Reference in New Issue
Block a user