mirror of
https://github.com/alanorth/cgspace-notes.git
synced 2024-11-27 00:48:19 +01:00
1212 lines
52 KiB
HTML
1212 lines
52 KiB
HTML
<!DOCTYPE html>
|
||
<html lang="en">
|
||
|
||
<head>
|
||
<meta charset="utf-8">
|
||
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
|
||
|
||
<meta property="og:title" content="November, 2017" />
|
||
<meta property="og:description" content="2017-11-01
|
||
|
||
|
||
The CORE developers responded to say they are looking into their bot not respecting our robots.txt
|
||
|
||
|
||
2017-11-02
|
||
|
||
|
||
Today there have been no hits by CORE and no alerts from Linode (coincidence?)
|
||
|
||
# grep -c "CORE" /var/log/nginx/access.log
|
||
0
|
||
|
||
|
||
Generate list of authors on CGSpace for Peter to go through and correct:
|
||
|
||
dspace=# \copy (select distinct text_value, count(*) as count from metadatavalue where metadata_field_id = (select metadata_field_id from metadatafieldregistry where element = 'contributor' and qualifier = 'author') AND resource_type_id = 2 group by text_value order by count desc) to /tmp/authors.csv with csv;
|
||
COPY 54701
|
||
|
||
" />
|
||
<meta property="og:type" content="article" />
|
||
<meta property="og:url" content="https://alanorth.github.io/cgspace-notes/2017-11/" />
|
||
<meta property="article:published_time" content="2017-11-02T09:37:54+02:00" />
|
||
<meta property="article:modified_time" content="2018-04-10T08:27:55+03:00" />
|
||
|
||
<meta name="twitter:card" content="summary"/>
|
||
<meta name="twitter:title" content="November, 2017"/>
|
||
<meta name="twitter:description" content="2017-11-01
|
||
|
||
|
||
The CORE developers responded to say they are looking into their bot not respecting our robots.txt
|
||
|
||
|
||
2017-11-02
|
||
|
||
|
||
Today there have been no hits by CORE and no alerts from Linode (coincidence?)
|
||
|
||
# grep -c "CORE" /var/log/nginx/access.log
|
||
0
|
||
|
||
|
||
Generate list of authors on CGSpace for Peter to go through and correct:
|
||
|
||
dspace=# \copy (select distinct text_value, count(*) as count from metadatavalue where metadata_field_id = (select metadata_field_id from metadatafieldregistry where element = 'contributor' and qualifier = 'author') AND resource_type_id = 2 group by text_value order by count desc) to /tmp/authors.csv with csv;
|
||
COPY 54701
|
||
|
||
"/>
|
||
<meta name="generator" content="Hugo 0.58.1" />
|
||
|
||
|
||
|
||
<script type="application/ld+json">
|
||
{
|
||
"@context": "http://schema.org",
|
||
"@type": "BlogPosting",
|
||
"headline": "November, 2017",
|
||
"url": "https:\/\/alanorth.github.io\/cgspace-notes\/2017-11\/",
|
||
"wordCount": "5428",
|
||
"datePublished": "2017-11-02T09:37:54\x2b02:00",
|
||
"dateModified": "2018-04-10T08:27:55\x2b03:00",
|
||
"author": {
|
||
"@type": "Person",
|
||
"name": "Alan Orth"
|
||
},
|
||
"keywords": "Notes"
|
||
}
|
||
</script>
|
||
|
||
|
||
|
||
<link rel="canonical" href="https://alanorth.github.io/cgspace-notes/2017-11/">
|
||
|
||
<title>November, 2017 | CGSpace Notes</title>
|
||
|
||
<!-- combined, minified CSS -->
|
||
<link href="https://alanorth.github.io/cgspace-notes/css/style.css" rel="stylesheet" integrity="sha384-G5B34w7DFTumWTswxYzTX7NWfbvQEg1HbFFEg6ItN03uTAAoS2qkPS/fu3LhuuSA" crossorigin="anonymous">
|
||
|
||
<!-- RSS 2.0 feed -->
|
||
|
||
|
||
|
||
|
||
|
||
|
||
</head>
|
||
|
||
<body>
|
||
|
||
|
||
<div class="blog-masthead">
|
||
<div class="container">
|
||
<nav class="nav blog-nav">
|
||
<a class="nav-link " href="https://alanorth.github.io/cgspace-notes/">Home</a>
|
||
</nav>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<header class="blog-header">
|
||
<div class="container">
|
||
<h1 class="blog-title"><a href="https://alanorth.github.io/cgspace-notes/" rel="home">CGSpace Notes</a></h1>
|
||
<p class="lead blog-description">Documenting day-to-day work on the <a href="https://cgspace.cgiar.org">CGSpace</a> repository.</p>
|
||
</div>
|
||
</header>
|
||
|
||
|
||
|
||
|
||
<div class="container">
|
||
<div class="row">
|
||
<div class="col-sm-8 blog-main">
|
||
|
||
|
||
|
||
|
||
<article class="blog-post">
|
||
<header>
|
||
<h2 class="blog-post-title"><a href="https://alanorth.github.io/cgspace-notes/2017-11/">November, 2017</a></h2>
|
||
<p class="blog-post-meta"><time datetime="2017-11-02T09:37:54+02:00">Thu Nov 02, 2017</time> by Alan Orth in
|
||
|
||
<i class="fa fa-tag" aria-hidden="true"></i> <a href="/cgspace-notes/tags/notes" rel="tag">Notes</a>
|
||
|
||
</p>
|
||
</header>
|
||
<h2 id="2017-11-01">2017-11-01</h2>
|
||
|
||
<ul>
|
||
<li>The CORE developers responded to say they are looking into their bot not respecting our robots.txt</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-02">2017-11-02</h2>
|
||
|
||
<ul>
|
||
<li><p>Today there have been no hits by CORE and no alerts from Linode (coincidence?)</p>
|
||
|
||
<pre><code># grep -c "CORE" /var/log/nginx/access.log
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>Generate list of authors on CGSpace for Peter to go through and correct:</p>
|
||
|
||
<pre><code>dspace=# \copy (select distinct text_value, count(*) as count from metadatavalue where metadata_field_id = (select metadata_field_id from metadatafieldregistry where element = 'contributor' and qualifier = 'author') AND resource_type_id = 2 group by text_value order by count desc) to /tmp/authors.csv with csv;
|
||
COPY 54701
|
||
</code></pre></li>
|
||
</ul>
|
||
|
||
<ul>
|
||
<li>Abenet asked if it would be possible to generate a report of items in Listing and Reports that had “International Fund for Agricultural Development” as the <em>only</em> investor</li>
|
||
<li>I opened a ticket with Atmire to ask if this was possible: <a href="https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=540">https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=540</a></li>
|
||
<li>Work on making the thumbnails in the item view clickable</li>
|
||
|
||
<li><p>Basically, once you read the METS XML for an item it becomes easy to trace the structure to find the bitstream link</p>
|
||
|
||
<pre><code>//mets:fileSec/mets:fileGrp[@USE='CONTENT']/mets:file/mets:FLocat[@LOCTYPE='URL']/@xlink:href
|
||
</code></pre></li>
|
||
|
||
<li><p>METS XML is available for all items with this pattern: /metadata/handle/10568/95947/mets.xml</p></li>
|
||
|
||
<li><p>I whipped up a quick hack to print a clickable link with this URL on the thumbnail but it needs to check a few corner cases, like when there is a thumbnail but no content bitstream!</p></li>
|
||
|
||
<li><p>Help proof fifty-three CIAT records for Sisay: <a href="https://dspacetest.cgiar.org/handle/10568/95895">https://dspacetest.cgiar.org/handle/10568/95895</a></p></li>
|
||
|
||
<li><p>A handful of issues with <code>cg.place</code> using format like “Lima, PE” instead of “Lima, Peru”</p></li>
|
||
|
||
<li><p>Also, some dates like with completely invalid format like “2010- 06” and “2011-3-28”</p></li>
|
||
|
||
<li><p>I also collapsed some consecutive whitespace on a handful of fields</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-03">2017-11-03</h2>
|
||
|
||
<ul>
|
||
<li>Atmire got back to us to say that they estimate it will take two days of labor to implement the change to Listings and Reports</li>
|
||
<li>I said I’d ask Abenet if she wants that feature</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-04">2017-11-04</h2>
|
||
|
||
<ul>
|
||
<li>I finished looking through Sisay’s CIAT records for the “Alianzas de Aprendizaje” data</li>
|
||
<li>I corrected about half of the authors to standardize them</li>
|
||
<li>Linode emailed this morning to say that the CPU usage was high again, this time at 6:14AM</li>
|
||
<li>It’s the first time in a few days that this has happened</li>
|
||
|
||
<li><p>I had a look to see what was going on, but it isn’t the CORE bot:</p>
|
||
|
||
<pre><code># awk '{print $1}' /var/log/nginx/access.log | sort -n | uniq -c | sort -h | tail
|
||
306 68.180.229.31
|
||
323 61.148.244.116
|
||
414 66.249.66.91
|
||
507 40.77.167.16
|
||
618 157.55.39.161
|
||
652 207.46.13.103
|
||
666 157.55.39.254
|
||
1173 104.196.152.243
|
||
1737 66.249.66.90
|
||
23101 138.201.52.218
|
||
</code></pre></li>
|
||
|
||
<li><p>138.201.52.218 is from some Hetzner server, and I see it making 40,000 requests yesterday too, but none before that:</p>
|
||
|
||
<pre><code># zgrep -c 138.201.52.218 /var/log/nginx/access.log*
|
||
/var/log/nginx/access.log:24403
|
||
/var/log/nginx/access.log.1:45958
|
||
/var/log/nginx/access.log.2.gz:0
|
||
/var/log/nginx/access.log.3.gz:0
|
||
/var/log/nginx/access.log.4.gz:0
|
||
/var/log/nginx/access.log.5.gz:0
|
||
/var/log/nginx/access.log.6.gz:0
|
||
</code></pre></li>
|
||
|
||
<li><p>It’s clearly a bot as it’s making tens of thousands of requests, but it’s using a “normal” user agent:</p>
|
||
|
||
<pre><code>Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2227.0 Safari/537.36
|
||
</code></pre></li>
|
||
|
||
<li><p>For now I don’t know what this user is!</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-05">2017-11-05</h2>
|
||
|
||
<ul>
|
||
<li>Peter asked if I could fix the appearance of “International Livestock Research Institute” in the author lookup during item submission</li>
|
||
<li>It looks to be just an issue with the user interface expecting authors to have both a first and last name:</li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/author-lookup.png" alt="Author lookup" />
|
||
<img src="/cgspace-notes/2017/11/add-author.png" alt="Add author" /></p>
|
||
|
||
<ul>
|
||
<li><p>But in the database the authors are correct (none with weird <code>, /</code> characters):</p>
|
||
|
||
<pre><code>dspace=# select distinct text_value, authority, confidence from metadatavalue value where resource_type_id=2 and metadata_field_id=3 and text_value like 'International Livestock Research Institute%';
|
||
text_value | authority | confidence
|
||
--------------------------------------------+--------------------------------------+------------
|
||
International Livestock Research Institute | 8f3865dc-d056-4aec-90b7-77f49ab4735c | 0
|
||
International Livestock Research Institute | f4db1627-47cd-4699-b394-bab7eba6dadc | 0
|
||
International Livestock Research Institute | | -1
|
||
International Livestock Research Institute | 8f3865dc-d056-4aec-90b7-77f49ab4735c | 600
|
||
International Livestock Research Institute | f4db1627-47cd-4699-b394-bab7eba6dadc | -1
|
||
International Livestock Research Institute | | 600
|
||
International Livestock Research Institute | 8f3865dc-d056-4aec-90b7-77f49ab4735c | -1
|
||
International Livestock Research Institute | 8f3865dc-d056-4aec-90b7-77f49ab4735c | 500
|
||
(8 rows)
|
||
</code></pre></li>
|
||
|
||
<li><p>So I’m not sure if this is just a graphical glitch or if editors have to edit this metadata field prior to approval</p></li>
|
||
|
||
<li><p>Looking at monitoring Tomcat’s JVM heap with Prometheus, it looks like we need to use JMX + <a href="https://github.com/prometheus/jmx_exporter">jmx_exporter</a></p></li>
|
||
|
||
<li><p>This guide shows how to <a href="https://geekflare.com/enable-jmx-tomcat-to-monitor-administer/">enable JMX in Tomcat</a> by modifying <code>CATALINA_OPTS</code></p></li>
|
||
|
||
<li><p>I was able to successfully connect to my local Tomcat with jconsole!</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-07">2017-11-07</h2>
|
||
|
||
<ul>
|
||
<li>CGSpace when down and up a few times this morning, first around 3AM, then around 7</li>
|
||
<li>Tsega had to restart Tomcat 7 to fix it temporarily</li>
|
||
|
||
<li><p>I will start by looking at bot usage (access.log.1 includes usage until 6AM today):</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log.1 | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
619 65.49.68.184
|
||
840 65.49.68.199
|
||
924 66.249.66.91
|
||
1131 68.180.229.254
|
||
1583 66.249.66.90
|
||
1953 207.46.13.103
|
||
1999 207.46.13.80
|
||
2021 157.55.39.161
|
||
2034 207.46.13.36
|
||
4681 104.196.152.243
|
||
</code></pre></li>
|
||
|
||
<li><p>104.196.152.243 seems to be a top scraper for a few weeks now:</p>
|
||
|
||
<pre><code># zgrep -c 104.196.152.243 /var/log/nginx/access.log*
|
||
/var/log/nginx/access.log:336
|
||
/var/log/nginx/access.log.1:4681
|
||
/var/log/nginx/access.log.2.gz:3531
|
||
/var/log/nginx/access.log.3.gz:3532
|
||
/var/log/nginx/access.log.4.gz:5786
|
||
/var/log/nginx/access.log.5.gz:8542
|
||
/var/log/nginx/access.log.6.gz:6988
|
||
/var/log/nginx/access.log.7.gz:7517
|
||
/var/log/nginx/access.log.8.gz:7211
|
||
/var/log/nginx/access.log.9.gz:2763
|
||
</code></pre></li>
|
||
|
||
<li><p>This user is responsible for hundreds and sometimes thousands of Tomcat sessions:</p>
|
||
|
||
<pre><code>$ grep 104.196.152.243 dspace.log.2017-11-07 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
954
|
||
$ grep 104.196.152.243 dspace.log.2017-11-03 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
6199
|
||
$ grep 104.196.152.243 dspace.log.2017-11-01 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
7051
|
||
</code></pre></li>
|
||
|
||
<li><p>The worst thing is that this user never specifies a user agent string so we can’t lump it in with the other bots using the Tomcat Session Crawler Manager Valve</p></li>
|
||
|
||
<li><p>They don’t request dynamic URLs like “/discover” but they seem to be fetching handles from XMLUI instead of REST (and some with <code>//handle</code>, note the regex below):</p>
|
||
|
||
<pre><code># grep -c 104.196.152.243 /var/log/nginx/access.log.1
|
||
4681
|
||
# grep 104.196.152.243 /var/log/nginx/access.log.1 | grep -c -P 'GET //?handle'
|
||
4618
|
||
</code></pre></li>
|
||
|
||
<li><p>I just realized that <code>ciat.cgiar.org</code> points to 104.196.152.243, so I should contact Leroy from CIAT to see if we can change their scraping behavior</p></li>
|
||
|
||
<li><p>The next IP (207.46.13.36) seem to be Microsoft’s bingbot, but all its requests specify the “bingbot” user agent and there are no requests for dynamic URLs that are forbidden, like “/discover”:</p>
|
||
|
||
<pre><code>$ grep -c 207.46.13.36 /var/log/nginx/access.log.1
|
||
2034
|
||
# grep 207.46.13.36 /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>The next IP (157.55.39.161) also seems to be bingbot, and none of its requests are for URLs forbidden by robots.txt either:</p>
|
||
|
||
<pre><code># grep 157.55.39.161 /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>The next few seem to be bingbot as well, and they declare a proper user agent and do not request dynamic URLs like “/discover”:</p>
|
||
|
||
<pre><code># grep -c -E '207.46.13.[0-9]{2,3}' /var/log/nginx/access.log.1
|
||
5997
|
||
# grep -E '207.46.13.[0-9]{2,3}' /var/log/nginx/access.log.1 | grep -c "bingbot"
|
||
5988
|
||
# grep -E '207.46.13.[0-9]{2,3}' /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>The next few seem to be Googlebot, and they declare a proper user agent and do not request dynamic URLs like “/discover”:</p>
|
||
|
||
<pre><code># grep -c -E '66.249.66.[0-9]{2,3}' /var/log/nginx/access.log.1
|
||
3048
|
||
# grep -E '66.249.66.[0-9]{2,3}' /var/log/nginx/access.log.1 | grep -c Google
|
||
3048
|
||
# grep -E '66.249.66.[0-9]{2,3}' /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>The next seems to be Yahoo, which declares a proper user agent and does not request dynamic URLs like “/discover”:</p>
|
||
|
||
<pre><code># grep -c 68.180.229.254 /var/log/nginx/access.log.1
|
||
1131
|
||
# grep 68.180.229.254 /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>The last of the top ten IPs seems to be some bot with a weird user agent, but they are not behaving too well:</p>
|
||
|
||
<pre><code># grep -c -E '65.49.68.[0-9]{3}' /var/log/nginx/access.log.1
|
||
2950
|
||
# grep -E '65.49.68.[0-9]{3}' /var/log/nginx/access.log.1 | grep -c "GET /discover"
|
||
330
|
||
</code></pre></li>
|
||
|
||
<li><p>Their user agents vary, ie:</p>
|
||
|
||
<ul>
|
||
<li><code>Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36</code></li>
|
||
<li><code>Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.97 Safari/537.11</code></li>
|
||
<li><code>Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.1; WOW64; Trident/7.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; .NET4.0E)</code></li>
|
||
</ul></li>
|
||
|
||
<li><p>I’ll just keep an eye on that one for now, as it only made a few hundred requests to dynamic discovery URLs</p></li>
|
||
|
||
<li><p>While it’s not in the top ten, Baidu is one bot that seems to not give a fuck:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep "7/Nov/2017" | grep -c Baiduspider
|
||
8912
|
||
# cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep "7/Nov/2017" | grep Baiduspider | grep -c -E "GET /(browse|discover|search-filter)"
|
||
2521
|
||
</code></pre></li>
|
||
|
||
<li><p>According to their documentation their bot <a href="http://www.baidu.com/search/robots_english.html">respects <code>robots.txt</code></a>, but I don’t see this being the case</p></li>
|
||
|
||
<li><p>I think I will end up blocking Baidu as well…</p></li>
|
||
|
||
<li><p>Next is for me to look and see what was happening specifically at 3AM and 7AM when the server crashed</p></li>
|
||
|
||
<li><p>I should look in nginx access.log, rest.log, oai.log, and DSpace’s dspace.log.2017-11-07</p></li>
|
||
|
||
<li><p>Here are the top IPs making requests to XMLUI from 2 to 8 AM:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep -E '07/Nov/2017:0[2-8]' | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
279 66.249.66.91
|
||
373 65.49.68.199
|
||
446 68.180.229.254
|
||
470 104.196.152.243
|
||
470 197.210.168.174
|
||
598 207.46.13.103
|
||
603 157.55.39.161
|
||
637 207.46.13.80
|
||
703 207.46.13.36
|
||
724 66.249.66.90
|
||
</code></pre></li>
|
||
|
||
<li><p>Of those, most are Google, Bing, Yahoo, etc, except 63.143.42.244 and 63.143.42.242 which are Uptime Robot</p></li>
|
||
|
||
<li><p>Here are the top IPs making requests to REST from 2 to 8 AM:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/rest.log /var/log/nginx/rest.log.1 | grep -E '07/Nov/2017:0[2-8]' | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
8 207.241.229.237
|
||
10 66.249.66.90
|
||
16 104.196.152.243
|
||
25 41.60.238.61
|
||
26 157.55.39.161
|
||
27 207.46.13.103
|
||
27 207.46.13.80
|
||
31 207.46.13.36
|
||
1498 50.116.102.77
|
||
</code></pre></li>
|
||
|
||
<li><p>The OAI requests during that same time period are nothing to worry about:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/oai.log /var/log/nginx/oai.log.1 | grep -E '07/Nov/2017:0[2-8]' | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
1 66.249.66.92
|
||
4 66.249.66.90
|
||
6 68.180.229.254
|
||
</code></pre></li>
|
||
|
||
<li><p>The top IPs from dspace.log during the 2–8 AM period:</p>
|
||
|
||
<pre><code>$ grep -E '2017-11-07 0[2-8]' dspace.log.2017-11-07 | grep -o -E 'ip_addr=[0-9.]+' | sort -n | uniq -c | sort -h | tail
|
||
143 ip_addr=213.55.99.121
|
||
181 ip_addr=66.249.66.91
|
||
223 ip_addr=157.55.39.161
|
||
248 ip_addr=207.46.13.80
|
||
251 ip_addr=207.46.13.103
|
||
291 ip_addr=207.46.13.36
|
||
297 ip_addr=197.210.168.174
|
||
312 ip_addr=65.49.68.199
|
||
462 ip_addr=104.196.152.243
|
||
488 ip_addr=66.249.66.90
|
||
</code></pre></li>
|
||
|
||
<li><p>These aren’t actually very interesting, as the top few are Google, CIAT, Bingbot, and a few other unknown scrapers</p></li>
|
||
|
||
<li><p>The number of requests isn’t even that high to be honest</p></li>
|
||
|
||
<li><p>As I was looking at these logs I noticed another heavy user (124.17.34.59) that was not active during this time period, but made many requests today alone:</p>
|
||
|
||
<pre><code># zgrep -c 124.17.34.59 /var/log/nginx/access.log*
|
||
/var/log/nginx/access.log:22581
|
||
/var/log/nginx/access.log.1:0
|
||
/var/log/nginx/access.log.2.gz:14
|
||
/var/log/nginx/access.log.3.gz:0
|
||
/var/log/nginx/access.log.4.gz:0
|
||
/var/log/nginx/access.log.5.gz:3
|
||
/var/log/nginx/access.log.6.gz:0
|
||
/var/log/nginx/access.log.7.gz:0
|
||
/var/log/nginx/access.log.8.gz:0
|
||
/var/log/nginx/access.log.9.gz:1
|
||
</code></pre></li>
|
||
|
||
<li><p>The whois data shows the IP is from China, but the user agent doesn’t really give any clues:</p>
|
||
|
||
<pre><code># grep 124.17.34.59 /var/log/nginx/access.log | awk -F'" ' '{print $3}' | sort | uniq -c | sort -h
|
||
210 "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36"
|
||
22610 "Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.2; Win64; x64; Trident/7.0; LCTE)"
|
||
</code></pre></li>
|
||
|
||
<li><p>A Google search for “LCTE bot” doesn’t return anything interesting, but this <a href="https://stackoverflow.com/questions/42500881/what-is-lcte-in-user-agent">Stack Overflow discussion</a> references the lack of information</p></li>
|
||
|
||
<li><p>So basically after a few hours of looking at the log files I am not closer to understanding what is going on!</p></li>
|
||
|
||
<li><p>I do know that we want to block Baidu, though, as it does not respect <code>robots.txt</code></p></li>
|
||
|
||
<li><p>And as we speak Linode alerted that the outbound traffic rate is very high for the past two hours (about 12–14 hours)</p></li>
|
||
|
||
<li><p>At least for now it seems to be that new Chinese IP (124.17.34.59):</p>
|
||
|
||
<pre><code># grep -E "07/Nov/2017:1[234]:" /var/log/nginx/access.log | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
198 207.46.13.103
|
||
203 207.46.13.80
|
||
205 207.46.13.36
|
||
218 157.55.39.161
|
||
249 45.5.184.221
|
||
258 45.5.187.130
|
||
386 66.249.66.90
|
||
410 197.210.168.174
|
||
1896 104.196.152.243
|
||
11005 124.17.34.59
|
||
</code></pre></li>
|
||
|
||
<li><p>Seems 124.17.34.59 are really downloading all our PDFs, compared to the next top active IPs during this time!</p>
|
||
|
||
<pre><code># grep -E "07/Nov/2017:1[234]:" /var/log/nginx/access.log | grep 124.17.34.59 | grep -c pdf
|
||
5948
|
||
# grep -E "07/Nov/2017:1[234]:" /var/log/nginx/access.log | grep 104.196.152.243 | grep -c pdf
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>About CIAT, I think I need to encourage them to specify a user agent string for their requests, because they are not reuising their Tomcat session and they are creating thousands of sessions per day</p></li>
|
||
|
||
<li><p>All CIAT requests vs unique ones:</p>
|
||
|
||
<pre><code>$ grep -Io -E 'session_id=[A-Z0-9]{32}:ip_addr=104.196.152.243' dspace.log.2017-11-07 | wc -l
|
||
3506
|
||
$ grep -Io -E 'session_id=[A-Z0-9]{32}:ip_addr=104.196.152.243' dspace.log.2017-11-07 | sort | uniq | wc -l
|
||
3506
|
||
</code></pre></li>
|
||
|
||
<li><p>I emailed CIAT about the session issue, user agent issue, and told them they should not scrape the HTML contents of communities, instead using the REST API</p></li>
|
||
|
||
<li><p>About Baidu, I found a link to their <a href="http://ziyuan.baidu.com/robots/">robots.txt tester tool</a></p></li>
|
||
|
||
<li><p>It seems like our robots.txt file is valid, and they claim to recognize that URLs like <code>/discover</code> should be forbidden (不允许, aka “not allowed”):</p></li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/baidu-robotstxt.png" alt="Baidu robots.txt tester" /></p>
|
||
|
||
<ul>
|
||
<li><p>But they literally just made this request today:</p>
|
||
|
||
<pre><code>180.76.15.136 - - [07/Nov/2017:06:25:11 +0000] "GET /discover?filtertype_0=crpsubject&filter_relational_operator_0=equals&filter_0=WATER%2C+LAND+AND+ECOSYSTEMS&filtertype=subject&filter_relational_operator=equals&filter=WATER+RESOURCES HTTP/1.1" 200 82265 "-" "Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)"
|
||
</code></pre></li>
|
||
|
||
<li><p>Along with another thousand or so requests to URLs that are forbidden in robots.txt today alone:</p>
|
||
|
||
<pre><code># grep -c Baiduspider /var/log/nginx/access.log
|
||
3806
|
||
# grep Baiduspider /var/log/nginx/access.log | grep -c -E "GET /(browse|discover|search-filter)"
|
||
1085
|
||
</code></pre></li>
|
||
|
||
<li><p>I will think about blocking their IPs but they have 164 of them!</p>
|
||
|
||
<pre><code># grep "Baiduspider/2.0" /var/log/nginx/access.log | awk '{print $1}' | sort -n | uniq | wc -l
|
||
164
|
||
</code></pre></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-08">2017-11-08</h2>
|
||
|
||
<ul>
|
||
<li>Linode sent several alerts last night about CPU usage and outbound traffic rate at 6:13PM</li>
|
||
<li>Linode sent another alert about CPU usage in the morning at 6:12AM</li>
|
||
|
||
<li><p>Jesus, the new Chinese IP (124.17.34.59) has downloaded 24,000 PDFs in the last 24 hours:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep -E "0[78]/Nov/2017:" | grep 124.17.34.59 | grep -v pdf.jpg | grep -c pdf
|
||
24981
|
||
</code></pre></li>
|
||
|
||
<li><p>This is about 20,000 Tomcat sessions:</p>
|
||
|
||
<pre><code>$ cat dspace.log.2017-11-07 dspace.log.2017-11-08 | grep -Io -E 'session_id=[A-Z0-9]{32}:ip_addr=124.17.34.59' | sort | uniq | wc -l
|
||
20733
|
||
</code></pre></li>
|
||
|
||
<li><p>I’m getting really sick of this</p></li>
|
||
|
||
<li><p>Sisay re-uploaded the CIAT records that I had already corrected earlier this week, erasing all my corrections</p></li>
|
||
|
||
<li><p>I had to re-correct all the publishers, places, names, dates, etc and apply the changes on DSpace Test</p></li>
|
||
|
||
<li><p>Run system updates on DSpace Test and reboot the server</p></li>
|
||
|
||
<li><p>Magdalena had written to say that two of their Phase II project tags were missing on CGSpace, so I added them (<a href="https://github.com/ilri/DSpace/pull/346">#346</a>)</p></li>
|
||
|
||
<li><p>I figured out a way to use nginx’s map function to assign a “bot” user agent to misbehaving clients who don’t define a user agent</p></li>
|
||
|
||
<li><p>Most bots are automatically lumped into one generic session by <a href="https://tomcat.apache.org/tomcat-7.0-doc/config/valve.html#Crawler_Session_Manager_Valve">Tomcat’s Crawler Session Manager Valve</a> but this only works if their user agent matches a pre-defined regular expression like <code>.*[bB]ot.*</code></p></li>
|
||
|
||
<li><p>Some clients send thousands of requests without a user agent which ends up creating thousands of Tomcat sessions, wasting precious memory, CPU, and database resources in the process</p></li>
|
||
|
||
<li><p>Basically, we modify the nginx config to add a mapping with a modified user agent <code>$ua</code>:</p>
|
||
|
||
<pre><code>map $remote_addr $ua {
|
||
# 2017-11-08 Random Chinese host grabbing 20,000 PDFs
|
||
124.17.34.59 'ChineseBot';
|
||
default $http_user_agent;
|
||
}
|
||
</code></pre></li>
|
||
|
||
<li><p>If the client’s address matches then the user agent is set, otherwise the default <code>$http_user_agent</code> variable is used</p></li>
|
||
|
||
<li><p>Then, in the server’s <code>/</code> block we pass this header to Tomcat:</p>
|
||
|
||
<pre><code>proxy_pass http://tomcat_http;
|
||
proxy_set_header User-Agent $ua;
|
||
</code></pre></li>
|
||
|
||
<li><p>Note to self: the <code>$ua</code> variable won’t show up in nginx access logs because the default <code>combined</code> log format doesn’t show it, so don’t run around pulling your hair out wondering with the modified user agents aren’t showing in the logs!</p></li>
|
||
|
||
<li><p>If a client matching one of these IPs connects without a session, it will be assigned one by the Crawler Session Manager Valve</p></li>
|
||
|
||
<li><p>You can verify by cross referencing nginx’s <code>access.log</code> and DSpace’s <code>dspace.log.2017-11-08</code>, for example</p></li>
|
||
|
||
<li><p>I will deploy this on CGSpace later this week</p></li>
|
||
|
||
<li><p>I am interested to check how this affects the number of sessions used by the CIAT and Chinese bots (see above on <a href="#2017-11-07">2017-11-07</a> for example)</p></li>
|
||
|
||
<li><p>I merged the clickable thumbnails code to <code>5_x-prod</code> (<a href="https://github.com/ilri/DSpace/pull/347">#347</a>) and will deploy it later along with the new bot mapping stuff (and re-run the Asible <code>nginx</code> and <code>tomcat</code> tags)</p></li>
|
||
|
||
<li><p>I was thinking about Baidu again and decided to see how many requests they have versus Google to URL paths that are explicitly forbidden in <code>robots.txt</code>:</p>
|
||
|
||
<pre><code># zgrep Baiduspider /var/log/nginx/access.log* | grep -c -E "GET /(browse|discover|search-filter)"
|
||
22229
|
||
# zgrep Googlebot /var/log/nginx/access.log* | grep -c -E "GET /(browse|discover|search-filter)"
|
||
0
|
||
</code></pre></li>
|
||
|
||
<li><p>It seems that they rarely even bother checking <code>robots.txt</code>, but Google does multiple times per day!</p>
|
||
|
||
<pre><code># zgrep Baiduspider /var/log/nginx/access.log* | grep -c robots.txt
|
||
14
|
||
# zgrep Googlebot /var/log/nginx/access.log* | grep -c robots.txt
|
||
1134
|
||
</code></pre></li>
|
||
|
||
<li><p>I have been looking for a reason to ban Baidu and this is definitely a good one</p></li>
|
||
|
||
<li><p>Disallowing <code>Baiduspider</code> in <code>robots.txt</code> probably won’t work because this bot doesn’t seem to respect the robot exclusion standard anyways!</p></li>
|
||
|
||
<li><p>I will whip up something in nginx later</p></li>
|
||
|
||
<li><p>Run system updates on CGSpace and reboot the server</p></li>
|
||
|
||
<li><p>Re-deploy latest <code>5_x-prod</code> branch on CGSpace and DSpace Test (includes the clickable thumbnails, CCAFS phase II project tags, and updated news text)</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-09">2017-11-09</h2>
|
||
|
||
<ul>
|
||
<li><p>Awesome, it seems my bot mapping stuff in nginx actually reduced the number of Tomcat sessions used by the CIAT scraper today, total requests and unique sessions:</p>
|
||
|
||
<pre><code># zcat -f -- /var/log/nginx/access.log.1 /var/log/nginx/access.log.2.gz | grep '09/Nov/2017' | grep -c 104.196.152.243
|
||
8956
|
||
$ grep 104.196.152.243 dspace.log.2017-11-09 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
223
|
||
</code></pre></li>
|
||
|
||
<li><p>Versus the same stats for yesterday and the day before:</p>
|
||
|
||
<pre><code># zcat -f -- /var/log/nginx/access.log.1 /var/log/nginx/access.log.2.gz | grep '08/Nov/2017' | grep -c 104.196.152.243
|
||
10216
|
||
$ grep 104.196.152.243 dspace.log.2017-11-08 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
2592
|
||
# zcat -f -- /var/log/nginx/access.log.2.gz /var/log/nginx/access.log.3.gz | grep '07/Nov/2017' | grep -c 104.196.152.243
|
||
8120
|
||
$ grep 104.196.152.243 dspace.log.2017-11-07 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
3506
|
||
</code></pre></li>
|
||
|
||
<li><p>The number of sessions is over <em>ten times less</em>!</p></li>
|
||
|
||
<li><p>This gets me thinking, I wonder if I can use something like nginx’s rate limiter to automatically change the user agent of clients who make too many requests</p></li>
|
||
|
||
<li><p>Perhaps using a combination of geo and map, like illustrated here: <a href="https://www.nginx.com/blog/rate-limiting-nginx/">https://www.nginx.com/blog/rate-limiting-nginx/</a></p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-11">2017-11-11</h2>
|
||
|
||
<ul>
|
||
<li>I was looking at the Google index and noticed there are 4,090 search results for dspace.ilri.org but only seven for mahider.ilri.org</li>
|
||
<li>Search with something like: inurl:dspace.ilri.org inurl:https</li>
|
||
<li>I want to get rid of those legacy domains eventually!</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-12">2017-11-12</h2>
|
||
|
||
<ul>
|
||
<li>Update the <a href="https://github.com/ilri/rmg-ansible-public">Ansible infrastructure templates</a> to be a little more modular and flexible</li>
|
||
|
||
<li><p>Looking at the top client IPs on CGSpace so far this morning, even though it’s only been eight hours:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep "12/Nov/2017" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
243 5.83.120.111
|
||
335 40.77.167.103
|
||
424 66.249.66.91
|
||
529 207.46.13.36
|
||
554 40.77.167.129
|
||
604 207.46.13.53
|
||
754 104.196.152.243
|
||
883 66.249.66.90
|
||
1150 95.108.181.88
|
||
1381 5.9.6.51
|
||
</code></pre></li>
|
||
|
||
<li><p>5.9.6.51 seems to be a Russian bot:</p>
|
||
|
||
<pre><code># grep 5.9.6.51 /var/log/nginx/access.log | tail -n 1
|
||
5.9.6.51 - - [12/Nov/2017:08:13:13 +0000] "GET /handle/10568/16515/recent-submissions HTTP/1.1" 200 5097 "-" "Mozilla/5.0 (compatible; MegaIndex.ru/2.0; +http://megaindex.com/crawler)"
|
||
</code></pre></li>
|
||
|
||
<li><p>What’s amazing is that it seems to reuse its Java session across all requests:</p>
|
||
|
||
<pre><code>$ grep -c -E 'session_id=[A-Z0-9]{32}:ip_addr=5.9.6.51' dspace.log.2017-11-12
|
||
1558
|
||
$ grep 5.9.6.51 dspace.log.2017-11-12 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
1
|
||
</code></pre></li>
|
||
|
||
<li><p>Bravo to MegaIndex.ru!</p></li>
|
||
|
||
<li><p>The same cannot be said for 95.108.181.88, which appears to be YandexBot, even though Tomcat’s Crawler Session Manager valve regex should match ‘YandexBot’:</p>
|
||
|
||
<pre><code># grep 95.108.181.88 /var/log/nginx/access.log | tail -n 1
|
||
95.108.181.88 - - [12/Nov/2017:08:33:17 +0000] "GET /bitstream/handle/10568/57004/GenebankColombia_23Feb2015.pdf HTTP/1.1" 200 972019 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
|
||
$ grep -c -E 'session_id=[A-Z0-9]{32}:ip_addr=95.108.181.88' dspace.log.2017-11-12
|
||
991
|
||
</code></pre></li>
|
||
|
||
<li><p>Move some items and collections on CGSpace for Peter Ballantyne, running <a href="https://gist.github.com/alanorth/e60b530ed4989df0c731afbb0c640515"><code>move_collections.sh</code></a> with the following configuration:</p>
|
||
|
||
<pre><code>10947/6 10947/1 10568/83389
|
||
10947/34 10947/1 10568/83389
|
||
10947/2512 10947/1 10568/83389
|
||
</code></pre></li>
|
||
|
||
<li><p>I explored nginx rate limits as a way to aggressively throttle Baidu bot which doesn’t seem to respect disallowed URLs in robots.txt</p></li>
|
||
|
||
<li><p>There’s an interesting <a href="https://www.nginx.com/blog/rate-limiting-nginx/">blog post from Nginx’s team about rate limiting</a> as well as a <a href="https://gist.github.com/arosenhagen/8aaf5d7f94171778c0e9">clever use of mapping with rate limits</a></p></li>
|
||
|
||
<li><p>The solution <a href="https://github.com/ilri/rmg-ansible-public/commit/f0646991772660c505bea9c5ac586490e7c86156">I came up with</a> uses tricks from both of those</p></li>
|
||
|
||
<li><p>I deployed the limit on CGSpace and DSpace Test and it seems to work well:</p>
|
||
|
||
<pre><code>$ http --print h https://cgspace.cgiar.org/handle/10568/1 User-Agent:'Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)'
|
||
HTTP/1.1 200 OK
|
||
Connection: keep-alive
|
||
Content-Encoding: gzip
|
||
Content-Language: en-US
|
||
Content-Type: text/html;charset=utf-8
|
||
Date: Sun, 12 Nov 2017 16:30:19 GMT
|
||
Server: nginx
|
||
Strict-Transport-Security: max-age=15768000
|
||
Transfer-Encoding: chunked
|
||
Vary: Accept-Encoding
|
||
X-Cocoon-Version: 2.2.0
|
||
X-Content-Type-Options: nosniff
|
||
X-Frame-Options: SAMEORIGIN
|
||
X-XSS-Protection: 1; mode=block
|
||
$ http --print h https://cgspace.cgiar.org/handle/10568/1 User-Agent:'Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)'
|
||
HTTP/1.1 503 Service Temporarily Unavailable
|
||
Connection: keep-alive
|
||
Content-Length: 206
|
||
Content-Type: text/html
|
||
Date: Sun, 12 Nov 2017 16:30:21 GMT
|
||
Server: nginx
|
||
</code></pre></li>
|
||
|
||
<li><p>The first request works, second is denied with an HTTP 503!</p></li>
|
||
|
||
<li><p>I need to remember to check the Munin graphs for PostgreSQL and JVM next week to see how this affects them</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-13">2017-11-13</h2>
|
||
|
||
<ul>
|
||
<li><p>At the end of the day I checked the logs and it really looks like the Baidu rate limiting is working, HTTP 200 vs 503:</p>
|
||
|
||
<pre><code># zcat -f -- /var/log/nginx/access.log.1 /var/log/nginx/access.log.2.gz | grep "13/Nov/2017" | grep "Baiduspider" | grep -c " 200 "
|
||
1132
|
||
# zcat -f -- /var/log/nginx/access.log.1 /var/log/nginx/access.log.2.gz | grep "13/Nov/2017" | grep "Baiduspider" | grep -c " 503 "
|
||
10105
|
||
</code></pre></li>
|
||
|
||
<li><p>Helping Sisay proof 47 records for IITA: <a href="https://dspacetest.cgiar.org/handle/10568/97029">https://dspacetest.cgiar.org/handle/10568/97029</a></p></li>
|
||
|
||
<li><p>From looking at the data in OpenRefine I found:</p>
|
||
|
||
<ul>
|
||
<li>Errors in <code>cg.authorship.types</code></li>
|
||
<li>Errors in <code>cg.coverage.country</code> (smart quote in “COTE D’IVOIRE”, “HAWAII” is not a country)</li>
|
||
<li>Whitespace issues in some <code>cg.contributor.affiliation</code></li>
|
||
<li>Whitespace issues in some <code>cg.identifier.doi</code> fields and most values are using HTTP instead of HTTPS</li>
|
||
<li>Whitespace issues in some <code>dc.contributor.author</code> fields</li>
|
||
<li>Issue with invalid <code>dc.date.issued</code> value “2011-3”</li>
|
||
<li>Description fields are poorly copy–pasted</li>
|
||
<li>Whitespace issues in <code>dc.description.sponsorship</code></li>
|
||
<li>Lots of inconsistency in <code>dc.format.extent</code> (mixed dash style, periods at the end of values)</li>
|
||
<li>Whitespace errors in <code>dc.identifier.citation</code></li>
|
||
<li>Whitespace errors in <code>dc.subject</code></li>
|
||
<li>Whitespace errors in <code>dc.title</code></li>
|
||
</ul></li>
|
||
|
||
<li><p>After uploading and looking at the data in DSpace Test I saw more errors with CRPs, subjects (one item had four copies of all of its subjects, another had a “.” in it), affiliations, sponsors, etc.</p></li>
|
||
|
||
<li><p>Atmire responded to the <a href="https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=510">ticket about ORCID stuff</a> a few days ago, today I told them that I need to talk to Peter and the partners to see what we would like to do</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-14">2017-11-14</h2>
|
||
|
||
<ul>
|
||
<li>Deploy some nginx configuration updates to CGSpace</li>
|
||
<li>They had been waiting on a branch for a few months and I think I just forgot about them</li>
|
||
<li>I have been running them on DSpace Test for a few days and haven’t seen any issues there</li>
|
||
<li>Started testing DSpace 6.2 and a few things have changed</li>
|
||
|
||
<li><p>Now PostgreSQL needs <code>pgcrypto</code>:</p>
|
||
|
||
<pre><code>$ psql dspace6
|
||
dspace6=# CREATE EXTENSION pgcrypto;
|
||
</code></pre></li>
|
||
|
||
<li><p>Also, local settings are no longer in <code>build.properties</code>, they are now in <code>local.cfg</code></p></li>
|
||
|
||
<li><p>I’m not sure if we can use separate profiles like we did before with <code>mvn -Denv=blah</code> to use blah.properties</p></li>
|
||
|
||
<li><p>It seems we need to use “system properties” to override settings, ie: <code>-Ddspace.dir=/Users/aorth/dspace6</code></p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-15">2017-11-15</h2>
|
||
|
||
<ul>
|
||
<li>Send Adam Hunt an invite to the DSpace Developers network on Yammer</li>
|
||
<li>He is the new head of communications at WLE, since Michael left</li>
|
||
<li>Merge changes to item view’s wording of link metadata (<a href="https://github.com/ilri/DSpace/pull/348">#348</a>)</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-17">2017-11-17</h2>
|
||
|
||
<ul>
|
||
<li>Uptime Robot said that CGSpace went down today and I see lots of <code>Timeout waiting for idle object</code> errors in the DSpace logs</li>
|
||
<li>I looked in PostgreSQL using <code>SELECT * FROM pg_stat_activity;</code> and saw that there were 73 active connections</li>
|
||
<li>After a few minutes the connecitons went down to 44 and CGSpace was kinda back up, it seems like Tsega restarted Tomcat</li>
|
||
|
||
<li><p>Looking at the REST and XMLUI log files, I don’t see anything too crazy:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/rest.log /var/log/nginx/rest.log.1 | grep "17/Nov/2017" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
13 66.249.66.223
|
||
14 207.46.13.36
|
||
17 207.46.13.137
|
||
22 207.46.13.23
|
||
23 66.249.66.221
|
||
92 66.249.66.219
|
||
187 104.196.152.243
|
||
1400 70.32.83.92
|
||
1503 50.116.102.77
|
||
6037 45.5.184.196
|
||
# cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep "17/Nov/2017" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
325 139.162.247.24
|
||
354 66.249.66.223
|
||
422 207.46.13.36
|
||
434 207.46.13.23
|
||
501 207.46.13.137
|
||
647 66.249.66.221
|
||
662 34.192.116.178
|
||
762 213.55.99.121
|
||
1867 104.196.152.243
|
||
2020 66.249.66.219
|
||
</code></pre></li>
|
||
|
||
<li><p>I need to look into using JMX to analyze active sessions I think, rather than looking at log files</p></li>
|
||
|
||
<li><p>After adding appropriate <a href="https://geekflare.com/enable-jmx-tomcat-to-monitor-administer/">JMX listener options to Tomcat’s JAVA_OPTS</a> and restarting Tomcat, I can connect remotely using an SSH dynamic port forward (SOCKS) on port 7777 for example, and then start jconsole locally like:</p>
|
||
|
||
<pre><code>$ jconsole -J-DsocksProxyHost=localhost -J-DsocksProxyPort=7777 service:jmx:rmi:///jndi/rmi://localhost:9000/jmxrmi -J-DsocksNonProxyHosts=
|
||
</code></pre></li>
|
||
|
||
<li><p>Looking at the MBeans you can drill down in Catalina→Manager→webapp→localhost→Attributes and see active sessions, etc</p></li>
|
||
|
||
<li><p>I want to enable JMX listener on CGSpace but I need to do some more testing on DSpace Test and see if it causes any performance impact, for example</p></li>
|
||
|
||
<li><p>If I hit the server with some requests as a normal user I see the session counter increase, but if I specify a bot user agent then the sessions seem to be reused (meaning the Crawler Session Manager is working)</p></li>
|
||
|
||
<li><p>Here is the Jconsole screen after looping <code>http --print Hh https://dspacetest.cgiar.org/handle/10568/1</code> for a few minutes:</p></li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/jconsole-sessions.png" alt="Jconsole sessions for XMLUI" /></p>
|
||
|
||
<ul>
|
||
<li>Switch DSpace Test to using the G1GC for JVM so I can see what the JVM graph looks like eventually, and start evaluating it for production</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-19">2017-11-19</h2>
|
||
|
||
<ul>
|
||
<li>Linode sent an alert that CGSpace was using a lot of CPU around 4–6 AM</li>
|
||
|
||
<li><p>Looking in the nginx access logs I see the most active XMLUI users between 4 and 6 AM:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep -E "19/Nov/2017:0[456]" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
111 66.249.66.155
|
||
171 5.9.6.51
|
||
188 54.162.241.40
|
||
229 207.46.13.23
|
||
233 207.46.13.137
|
||
247 40.77.167.6
|
||
251 207.46.13.36
|
||
275 68.180.229.254
|
||
325 104.196.152.243
|
||
1610 66.249.66.153
|
||
</code></pre></li>
|
||
|
||
<li><p>66.249.66.153 appears to be Googlebot:</p>
|
||
|
||
<pre><code>66.249.66.153 - - [19/Nov/2017:06:26:01 +0000] "GET /handle/10568/2203 HTTP/1.1" 200 6309 "-" "Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)"
|
||
</code></pre></li>
|
||
|
||
<li><p>We know Googlebot is persistent but behaves well, so I guess it was just a coincidence that it came at a time when we had other traffic and server activity</p></li>
|
||
|
||
<li><p>In related news, I see an Atmire update process going for many hours and responsible for hundreds of thousands of log entries (two thirds of all log entries)</p>
|
||
|
||
<pre><code>$ wc -l dspace.log.2017-11-19
|
||
388472 dspace.log.2017-11-19
|
||
$ grep -c com.atmire.utils.UpdateSolrStatsMetadata dspace.log.2017-11-19
|
||
267494
|
||
</code></pre></li>
|
||
|
||
<li><p>WTF is this process doing every day, and for so many hours?</p></li>
|
||
|
||
<li><p>In unrelated news, when I was looking at the DSpace logs I saw a bunch of errors like this:</p>
|
||
|
||
<pre><code>2017-11-19 03:00:32,806 INFO org.apache.pdfbox.pdfparser.PDFParser @ Document is encrypted
|
||
2017-11-19 03:00:32,807 ERROR org.apache.pdfbox.filter.FlateFilter @ FlateFilter: stop reading corrupt stream due to a DataFormatException
|
||
</code></pre></li>
|
||
|
||
<li><p>It’s been a few days since I enabled the G1GC on DSpace Test and the JVM graph definitely changed:</p></li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/tomcat-jvm-g1gc.png" alt="Tomcat G1GC" /></p>
|
||
|
||
<h2 id="2017-11-20">2017-11-20</h2>
|
||
|
||
<ul>
|
||
<li>I found <a href="https://www.cakesolutions.net/teamblogs/low-pause-gc-on-the-jvm">an article about JVM tuning</a> that gives some pointers how to enable logging and tools to analyze logs for you</li>
|
||
<li>Also notes on <a href="https://blog.gceasy.io/2016/11/15/rotating-gc-log-files/">rotating GC logs</a></li>
|
||
<li>I decided to switch DSpace Test back to the CMS garbage collector because it is designed for low pauses and high throughput (like G1GC!) and because we haven’t even tried to monitor or tune it</li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-21">2017-11-21</h2>
|
||
|
||
<ul>
|
||
<li><p>Magdalena was having problems logging in via LDAP and it seems to be a problem with the CGIAR LDAP server:</p>
|
||
|
||
<pre><code>2017-11-21 11:11:09,621 WARN org.dspace.authenticate.LDAPAuthentication @ anonymous:session_id=2FEC0E5286C17B6694567FFD77C3171C:ip_addr=77.241.141.58:ldap_authentication:type=failed_auth javax.naming.CommunicationException\colon; simple bind failed\colon; svcgroot2.cgiarad.org\colon;3269 [Root exception is javax.net.ssl.SSLHandshakeException\colon; sun.security.validator.ValidatorException\colon; PKIX path validation failed\colon; java.security.cert.CertPathValidatorException\colon; validity check failed]
|
||
</code></pre></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-22">2017-11-22</h2>
|
||
|
||
<ul>
|
||
<li>Linode sent an alert that the CPU usage on the CGSpace server was very high around 4 to 6 AM</li>
|
||
|
||
<li><p>The logs don’t show anything particularly abnormal between those hours:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep -E "22/Nov/2017:0[456]" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
136 31.6.77.23
|
||
174 68.180.229.254
|
||
217 66.249.66.91
|
||
256 157.55.39.79
|
||
268 54.144.57.183
|
||
281 207.46.13.137
|
||
282 207.46.13.36
|
||
290 207.46.13.23
|
||
696 66.249.66.90
|
||
707 104.196.152.243
|
||
</code></pre></li>
|
||
|
||
<li><p>I haven’t seen 54.144.57.183 before, it is apparently the CCBot from commoncrawl.org</p></li>
|
||
|
||
<li><p>In other news, it looks like the JVM garbage collection pattern is back to its standard jigsaw pattern after switching back to CMS a few days ago:</p></li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/tomcat-jvm-cms.png" alt="Tomcat JVM with CMS GC" /></p>
|
||
|
||
<h2 id="2017-11-23">2017-11-23</h2>
|
||
|
||
<ul>
|
||
<li>Linode alerted again that CPU usage was high on CGSpace from 4:13 to 6:13 AM</li>
|
||
|
||
<li><p>I see a lot of Googlebot (66.249.66.90) in the XMLUI access logs</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 | grep -E "23/Nov/2017:0[456]" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
88 66.249.66.91
|
||
140 68.180.229.254
|
||
155 54.196.2.131
|
||
182 54.224.164.166
|
||
301 157.55.39.79
|
||
315 207.46.13.36
|
||
331 207.46.13.23
|
||
358 207.46.13.137
|
||
565 104.196.152.243
|
||
1570 66.249.66.90
|
||
</code></pre></li>
|
||
|
||
<li><p>… and the usual REST scrapers from CIAT (45.5.184.196) and CCAFS (70.32.83.92):</p>
|
||
|
||
<pre><code># cat /var/log/nginx/rest.log /var/log/nginx/rest.log.1 | grep -E "23/Nov/2017:0[456]" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
5 190.120.6.219
|
||
6 104.198.9.108
|
||
14 104.196.152.243
|
||
21 112.134.150.6
|
||
22 157.55.39.79
|
||
22 207.46.13.137
|
||
23 207.46.13.36
|
||
26 207.46.13.23
|
||
942 45.5.184.196
|
||
3995 70.32.83.92
|
||
</code></pre></li>
|
||
|
||
<li><p>These IPs crawling the REST API don’t specify user agents and I’d assume they are creating many Tomcat sessions</p></li>
|
||
|
||
<li><p>I would catch them in nginx to assign a “bot” user agent to them so that the Tomcat Crawler Session Manager valve could deal with them, but they seem to create any really — at least not in the dspace.log:</p>
|
||
|
||
<pre><code>$ grep 70.32.83.92 dspace.log.2017-11-23 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
2
|
||
</code></pre></li>
|
||
|
||
<li><p>I’m wondering if REST works differently, or just doesn’t log these sessions?</p></li>
|
||
|
||
<li><p>I wonder if they are measurable via JMX MBeans?</p></li>
|
||
|
||
<li><p>I did some tests locally and I don’t see the sessionCounter incrementing after making requests to REST, but it does with XMLUI and OAI</p></li>
|
||
|
||
<li><p>I came across some interesting PostgreSQL tuning advice for SSDs: <a href="https://amplitude.engineering/how-a-single-postgresql-config-change-improved-slow-query-performance-by-50x-85593b8991b0">https://amplitude.engineering/how-a-single-postgresql-config-change-improved-slow-query-performance-by-50x-85593b8991b0</a></p></li>
|
||
|
||
<li><p>Apparently setting <code>random_page_cost</code> to 1 is “common” advice for systems running PostgreSQL on SSD (the default is 4)</p></li>
|
||
|
||
<li><p>So I deployed this on DSpace Test and will check the Munin PostgreSQL graphs in a few days to see if anything changes</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-24">2017-11-24</h2>
|
||
|
||
<ul>
|
||
<li>It’s too early to tell for sure, but after I made the <code>random_page_cost</code> change on DSpace Test’s PostgreSQL yesterday the number of connections dropped drastically:</li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/postgres-connections-week.png" alt="PostgreSQL connections after tweak (week)" /></p>
|
||
|
||
<ul>
|
||
<li>There have been other temporary drops before, but if I look at the past month and actually the whole year, the trend is that connections are four or five times higher on average:</li>
|
||
</ul>
|
||
|
||
<p><img src="/cgspace-notes/2017/11/postgres-connections-month.png" alt="PostgreSQL connections after tweak (month)" /></p>
|
||
|
||
<ul>
|
||
<li>I just realized that we’re not logging access requests to other vhosts on CGSpace, so it’s possible I have no idea that we’re getting slammed at 4AM on another domain that we’re just silently redirecting to cgspace.cgiar.org</li>
|
||
<li>I’ve enabled logging on the CGIAR Library on CGSpace so I can check to see if there are many requests there</li>
|
||
<li>In just a few seconds I already see a dozen requests from Googlebot (of course they get HTTP 301 redirects to cgspace.cgiar.org)</li>
|
||
|
||
<li><p>I also noticed that CGNET appears to be monitoring the old domain every few minutes:</p>
|
||
|
||
<pre><code>192.156.137.184 - - [24/Nov/2017:20:33:58 +0000] "HEAD / HTTP/1.1" 301 0 "-" "curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.27.1 zlib/1.2.3 libidn/1.18 libssh2/1.4.2"
|
||
</code></pre></li>
|
||
|
||
<li><p>I should probably tell CGIAR people to have CGNET stop that</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-26">2017-11-26</h2>
|
||
|
||
<ul>
|
||
<li>Linode alerted that CGSpace server was using too much CPU from 5:18 to 7:18 AM</li>
|
||
|
||
<li><p>Yet another mystery because the load for all domains looks fine at that time:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/access.log /var/log/nginx/access.log.1 /var/log/nginx/library-access.log /var/log/nginx/library-access.log.1 | grep -E "26/Nov/2017:0[567]" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
190 66.249.66.83
|
||
195 104.196.152.243
|
||
220 40.77.167.82
|
||
246 207.46.13.137
|
||
247 68.180.229.254
|
||
257 157.55.39.214
|
||
289 66.249.66.91
|
||
298 157.55.39.206
|
||
379 66.249.66.70
|
||
1855 66.249.66.90
|
||
</code></pre></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-29">2017-11-29</h2>
|
||
|
||
<ul>
|
||
<li>Linode alerted that CGSpace was using 279% CPU from 6 to 8 AM this morning</li>
|
||
<li>About an hour later Uptime Robot said that the server was down</li>
|
||
|
||
<li><p>Here are all the top XMLUI and REST users from today:</p>
|
||
|
||
<pre><code># cat /var/log/nginx/rest.log /var/log/nginx/rest.log.1 /var/log/nginx/access.log /var/log/nginx/access.log.1 /var/log/nginx/library-access.log /var/log/nginx/library-access.log.1 | grep -E "29/Nov/2017" | awk '{print $1}' | sort -n | uniq -c | sort -h | tail
|
||
540 66.249.66.83
|
||
659 40.77.167.36
|
||
663 157.55.39.214
|
||
681 157.55.39.206
|
||
733 157.55.39.158
|
||
850 66.249.66.70
|
||
1311 66.249.66.90
|
||
1340 104.196.152.243
|
||
4008 70.32.83.92
|
||
6053 45.5.184.196
|
||
</code></pre></li>
|
||
|
||
<li><p>PostgreSQL activity shows 69 connections</p></li>
|
||
|
||
<li><p>I don’t have time to troubleshoot more as I’m in Nairobi working on the HPC so I just restarted Tomcat for now</p></li>
|
||
|
||
<li><p>A few hours later Uptime Robot says the server is down again</p></li>
|
||
|
||
<li><p>I don’t see much activity in the logs but there are 87 PostgreSQL connections</p></li>
|
||
|
||
<li><p>But shit, there were 10,000 unique Tomcat sessions today:</p>
|
||
|
||
<pre><code>$ cat dspace.log.2017-11-29 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
10037
|
||
</code></pre></li>
|
||
|
||
<li><p>Although maybe that’s not much, as the previous two days had more:</p>
|
||
|
||
<pre><code>$ cat dspace.log.2017-11-27 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
12377
|
||
$ cat dspace.log.2017-11-28 | grep -o -E 'session_id=[A-Z0-9]{32}' | sort -n | uniq | wc -l
|
||
16984
|
||
</code></pre></li>
|
||
|
||
<li><p>I think we just need start increasing the number of allowed PostgreSQL connections instead of fighting this, as it’s the most common source of crashes we have</p></li>
|
||
|
||
<li><p>I will bump DSpace’s <code>db.maxconnections</code> from 60 to 90, and PostgreSQL’s <code>max_connections</code> from 183 to 273 (which is using my loose formula of 90 * webapps + 3)</p></li>
|
||
|
||
<li><p>I really need to figure out how to get DSpace to use a PostgreSQL connection pool</p></li>
|
||
</ul>
|
||
|
||
<h2 id="2017-11-30">2017-11-30</h2>
|
||
|
||
<ul>
|
||
<li>Linode alerted about high CPU usage on CGSpace again around 6 to 8 AM</li>
|
||
<li>Then Uptime Robot said CGSpace was down a few minutes later, but it resolved itself I think (or Tsega restarted Tomcat, I don’t know)</li>
|
||
</ul>
|
||
|
||
|
||
|
||
|
||
|
||
</article>
|
||
|
||
|
||
|
||
</div> <!-- /.blog-main -->
|
||
|
||
<aside class="col-sm-3 ml-auto blog-sidebar">
|
||
|
||
|
||
|
||
<section class="sidebar-module">
|
||
<h4>Recent Posts</h4>
|
||
<ol class="list-unstyled">
|
||
|
||
|
||
<li><a href="/cgspace-notes/posts/">Posts</a></li>
|
||
|
||
<li><a href="/cgspace-notes/2019-09/">September, 2019</a></li>
|
||
|
||
<li><a href="/cgspace-notes/2019-08/">August, 2019</a></li>
|
||
|
||
<li><a href="/cgspace-notes/2019-07/">July, 2019</a></li>
|
||
|
||
<li><a href="/cgspace-notes/2019-06/">June, 2019</a></li>
|
||
|
||
</ol>
|
||
</section>
|
||
|
||
|
||
|
||
|
||
<section class="sidebar-module">
|
||
<h4>Links</h4>
|
||
<ol class="list-unstyled">
|
||
|
||
<li><a href="https://cgspace.cgiar.org">CGSpace</a></li>
|
||
|
||
<li><a href="https://dspacetest.cgiar.org">DSpace Test</a></li>
|
||
|
||
<li><a href="https://github.com/ilri/DSpace">CGSpace @ GitHub</a></li>
|
||
|
||
</ol>
|
||
</section>
|
||
|
||
</aside>
|
||
|
||
|
||
</div> <!-- /.row -->
|
||
</div> <!-- /.container -->
|
||
|
||
|
||
|
||
<footer class="blog-footer">
|
||
<p>
|
||
|
||
Blog template created by <a href="https://twitter.com/mdo">@mdo</a>, ported to Hugo by <a href='https://twitter.com/mralanorth'>@mralanorth</a>.
|
||
|
||
</p>
|
||
<p>
|
||
<a href="#">Back to top</a>
|
||
</p>
|
||
</footer>
|
||
|
||
|
||
</body>
|
||
|
||
</html>
|