<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I seem to be getting rate limited by RADB with update-asn-db /
update-prefix-db. It will get partway through and then die. Doubly
unfortunate is the fact that IXP Manager doesn't keep track of
where it was in the cache run, so the next time it starts over
again at the top. As a result, networks farther down probably
never get updated.</p>
<p>Are other people running into this?<br>
</p>
<p><br>
</p>
<p><a moz-do-not-send="true"
href="https://www.radb.net/support/informational/query.html"
class="moz-txt-link-freetext">https://www.radb.net/support/informational/query.html</a>
says:<br>
<br>
<blockquote type="cite">Please note: The RADb WHOIS service is
rate-limited. For large queries, we recommend using the RADb API<br>
<br>
Programmatic access is considered abuse of the service.</blockquote>
<br>
Isn't that exactly what IXP Manager / bgpq3 / bgpq4 are doing???</p>
<p><br>
Later, it says:<br>
<br>
<blockquote type="cite"><span style="white-space: pre-wrap; display: block; width: 98vw;">If you plan on making a large number of queries please invoke a persistent TCP/IP session. This is done by telnetting directly to whois.radb.net and issuing the !! command. This will spare our server having to establish and teardown connections for every query.</span></blockquote>
<br>
bgpq4, at least, does that. But IXP Manager is still separately
invoking bgpq4 for each AS when running update-asn-db /
update-prefix-db.</p>
<pre class="moz-signature" cols="72">--
Richard</pre>
</body>
</html>