<!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="January, 2019" />
<meta property="og:description" content="2019-01-02

Linode alerted that CGSpace (linode18) had a higher outbound traffic rate than normal early this morning
I don&rsquo;t see anything interesting in the web server logs around that time though:

# zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | awk &#39;{print $1}&#39; | sort | uniq -c | sort -n | tail -n 10
     92 40.77.167.4
     99 210.7.29.100
    120 38.126.157.45
    177 35.237.175.180
    177 40.77.167.32
    216 66.249.75.219
    225 18.203.76.93
    261 46.101.86.248
    357 207.46.13.1
    903 54.70.40.11
" />
<meta property="og:type" content="article" />
<meta property="og:url" content="https://alanorth.github.io/cgspace-notes/2019-01/" />
<meta property="article:published_time" content="2019-01-02T09:48:30+02:00" />
<meta property="article:modified_time" content="2019-10-28T13:39:25+02:00" />

<meta name="twitter:card" content="summary"/>
<meta name="twitter:title" content="January, 2019"/>
<meta name="twitter:description" content="2019-01-02

Linode alerted that CGSpace (linode18) had a higher outbound traffic rate than normal early this morning
I don&rsquo;t see anything interesting in the web server logs around that time though:

# zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | awk &#39;{print $1}&#39; | sort | uniq -c | sort -n | tail -n 10
     92 40.77.167.4
     99 210.7.29.100
    120 38.126.157.45
    177 35.237.175.180
    177 40.77.167.32
    216 66.249.75.219
    225 18.203.76.93
    261 46.101.86.248
    357 207.46.13.1
    903 54.70.40.11
"/>
<meta name="generator" content="Hugo 0.73.0" />


    
<script type="application/ld+json">
{
  "@context": "http://schema.org",
  "@type": "BlogPosting",
  "headline": "January, 2019",
  "url": "https://alanorth.github.io/cgspace-notes/2019-01/",
  "wordCount": "5532",
  "datePublished": "2019-01-02T09:48:30+02:00",
  "dateModified": "2019-10-28T13:39:25+02:00",
  "author": {
    "@type": "Person",
    "name": "Alan Orth"
  },
  "keywords": "Notes"
}
</script>



    <link rel="canonical" href="https://alanorth.github.io/cgspace-notes/2019-01/">

    <title>January, 2019 | CGSpace Notes</title>

    
    <!-- combined, minified CSS -->
    
    <link href="https://alanorth.github.io/cgspace-notes/css/style.6da5c906cc7a8fbb93f31cd2316c5dbe3f19ac4aa6bfb066f1243045b8f6061e.css" rel="stylesheet" integrity="sha256-baXJBsx6j7uT8xzSMWxdvj8ZrEqmv7Bm8SQwRbj2Bh4=" crossorigin="anonymous">
    

    <!-- minified Font Awesome for SVG icons -->
    
    <script defer src="https://alanorth.github.io/cgspace-notes/js/fontawesome.min.f3d2a1f5980bab30ddd0d8cadbd496475309fc48e2b1d052c5c09e6facffcb0f.js" integrity="sha256-89Kh9ZgLqzDd0NjK29SWR1MJ/EjisdBSxcCeb6z/yw8=" crossorigin="anonymous"></script>

    <!-- 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" dir="auto"><a href="https://alanorth.github.io/cgspace-notes/" rel="home">CGSpace Notes</a></h1>
        <p class="lead blog-description" dir="auto">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" dir="auto"><a href="https://alanorth.github.io/cgspace-notes/2019-01/">January, 2019</a></h2>
    <p class="blog-post-meta"><time datetime="2019-01-02T09:48:30+02:00">Wed Jan 02, 2019</time> by Alan Orth in 
<span class="fas fa-folder" aria-hidden="true"></span>&nbsp;<a href="/cgspace-notes/categories/notes/" rel="category tag">Notes</a>


</p>
  </header>
  <h2 id="2019-01-02">2019-01-02</h2>
<ul>
<li>Linode alerted that CGSpace (linode18) had a higher outbound traffic rate than normal early this morning</li>
<li>I don&rsquo;t see anything interesting in the web server logs around that time though:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
     92 40.77.167.4
     99 210.7.29.100
    120 38.126.157.45
    177 35.237.175.180
    177 40.77.167.32
    216 66.249.75.219
    225 18.203.76.93
    261 46.101.86.248
    357 207.46.13.1
    903 54.70.40.11
</code></pre><ul>
<li>Analyzing the types of requests made by the top few IPs during that time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | grep 54.70.40.11 | grep -o -E &quot;(bitstream|discover|handle)&quot; | sort | uniq -c
     30 bitstream
    534 discover
    352 handle
# zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | grep 207.46.13.1 | grep -o -E &quot;(bitstream|discover|handle)&quot; | sort | uniq -c
    194 bitstream
    345 handle
# zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;02/Jan/2019:0(1|2|3)&quot; | grep 46.101.86.248 | grep -o -E &quot;(bitstream|discover|handle)&quot; | sort | uniq -c
    261 handle
</code></pre><ul>
<li>It&rsquo;s not clear to me what was causing the outbound traffic spike</li>
<li>Oh nice! The once-per-year cron job for rotating the Solr statistics actually worked now (for the first time ever!):</li>
</ul>
<pre><code>Moving: 81742 into core statistics-2010
Moving: 1837285 into core statistics-2011
Moving: 3764612 into core statistics-2012
Moving: 4557946 into core statistics-2013
Moving: 5483684 into core statistics-2014
Moving: 2941736 into core statistics-2015
Moving: 5926070 into core statistics-2016
Moving: 10562554 into core statistics-2017
Moving: 18497180 into core statistics-2018
</code></pre><ul>
<li>This could by why the outbound traffic rate was high, due to the S3 backup that run at 3:30AM&hellip;</li>
<li>Run all system updates on DSpace Test (linode19) and reboot the server</li>
</ul>
<h2 id="2019-01-03">2019-01-03</h2>
<ul>
<li>Update local Docker image for DSpace PostgreSQL, re-using the existing data volume:</li>
</ul>
<pre><code>$ sudo docker pull postgres:9.6-alpine
$ sudo docker rm dspacedb
$ sudo docker run --name dspacedb -v /home/aorth/.local/lib/containers/volumes/dspacedb_data:/var/lib/postgresql/data -e POSTGRES_PASSWORD=postgres -p 5432:5432 -d postgres:9.6-alpine
</code></pre><ul>
<li>Testing DSpace 5.9 with Tomcat 8.5.37 on my local machine and I see that Atmire&rsquo;s Listings and Reports still doesn&rsquo;t work
<ul>
<li>After logging in via XMLUI and clicking the Listings and Reports link from the sidebar it redirects me to a JSPUI login page</li>
<li>If I log in again there the Listings and Reports work&hellip; hmm.</li>
</ul>
</li>
<li>The JSPUI application—which Listings and Reports depends upon—also does not load, though the error is perhaps unrelated:</li>
</ul>
<pre><code>2019-01-03 14:45:21,727 INFO  org.dspace.browse.BrowseEngine @ anonymous:session_id=9471D72242DAA05BCC87734FE3C66EA6:ip_addr=127.0.0.1:browse_mini:
2019-01-03 14:45:21,971 INFO  org.dspace.app.webui.discovery.DiscoverUtility @ facets for scope, null: 23
2019-01-03 14:45:22,115 WARN  org.dspace.app.webui.servlet.InternalErrorServlet @ :session_id=9471D72242DAA05BCC87734FE3C66EA6:internal_error:-- URL Was: http://localhost:8080/jspui/internal-error
-- Method: GET
-- Parameters were:

org.apache.jasper.JasperException: /home.jsp (line: [214], column: [1]) /discovery/static-tagcloud-facet.jsp (line: [57], column: [8]) No tag [tagcloud] defined in tag library imported with prefix [dspace]
    at org.apache.jasper.compiler.DefaultErrorHandler.jspError(DefaultErrorHandler.java:41)
    at org.apache.jasper.compiler.ErrorDispatcher.dispatch(ErrorDispatcher.java:291)
    at org.apache.jasper.compiler.ErrorDispatcher.jspError(ErrorDispatcher.java:97)
    at org.apache.jasper.compiler.Parser.processIncludeDirective(Parser.java:347)
    at org.apache.jasper.compiler.Parser.parseIncludeDirective(Parser.java:380)
    at org.apache.jasper.compiler.Parser.parseDirective(Parser.java:481)
    at org.apache.jasper.compiler.Parser.parseElements(Parser.java:1445)
    at org.apache.jasper.compiler.Parser.parseBody(Parser.java:1683)
    at org.apache.jasper.compiler.Parser.parseOptionalBody(Parser.java:1016)
    at org.apache.jasper.compiler.Parser.parseCustomTag(Parser.java:1291)
    at org.apache.jasper.compiler.Parser.parseElements(Parser.java:1470)
    at org.apache.jasper.compiler.Parser.parse(Parser.java:144)
    at org.apache.jasper.compiler.ParserController.doParse(ParserController.java:244)
    at org.apache.jasper.compiler.ParserController.parse(ParserController.java:105)
    at org.apache.jasper.compiler.Compiler.generateJava(Compiler.java:202)
    at org.apache.jasper.compiler.Compiler.compile(Compiler.java:373)
    at org.apache.jasper.compiler.Compiler.compile(Compiler.java:350)
    at org.apache.jasper.compiler.Compiler.compile(Compiler.java:334)
    at org.apache.jasper.JspCompilationContext.compile(JspCompilationContext.java:595)
    at org.apache.jasper.servlet.JspServletWrapper.service(JspServletWrapper.java:399)
    at org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:386)
    at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:330)
    at javax.servlet.http.HttpServlet.service(HttpServlet.java:742)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:231)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)
    at org.apache.tomcat.websocket.server.WsFilter.doFilter(WsFilter.java:52)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:193)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)
    at org.apache.catalina.core.ApplicationDispatcher.invoke(ApplicationDispatcher.java:728)
    at org.apache.catalina.core.ApplicationDispatcher.processRequest(ApplicationDispatcher.java:470)
    at org.apache.catalina.core.ApplicationDispatcher.doForward(ApplicationDispatcher.java:395)
    at org.apache.catalina.core.ApplicationDispatcher.forward(ApplicationDispatcher.java:316)
    at org.dspace.app.webui.util.JSPManager.showJSP(JSPManager.java:60)
    at org.apache.jsp.index_jsp._jspService(index_jsp.java:191)
    at org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:70)
    at javax.servlet.http.HttpServlet.service(HttpServlet.java:742)
    at org.apache.jasper.servlet.JspServletWrapper.service(JspServletWrapper.java:476)
    at org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:386)
    at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:330)
    at javax.servlet.http.HttpServlet.service(HttpServlet.java:742)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:231)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)
    at org.apache.tomcat.websocket.server.WsFilter.doFilter(WsFilter.java:52)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:193)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)
    at org.dspace.utils.servlet.DSpaceWebappServletFilter.doFilter(DSpaceWebappServletFilter.java:78)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:193)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)
    at org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:198)
    at org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:96)
    at org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:493)
    at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:140)
    at org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:81)
    at org.apache.catalina.valves.CrawlerSessionManagerValve.invoke(CrawlerSessionManagerValve.java:234)
    at org.apache.catalina.valves.AbstractAccessLogValve.invoke(AbstractAccessLogValve.java:650)
    at org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:87)
    at org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:342)
    at org.apache.coyote.http11.Http11Processor.service(Http11Processor.java:800)
    at org.apache.coyote.AbstractProcessorLight.process(AbstractProcessorLight.java:66)
    at org.apache.coyote.AbstractProtocol$ConnectionHandler.process(AbstractProtocol.java:806)
    at org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1498)
    at org.apache.tomcat.util.net.SocketProcessorBase.run(SocketProcessorBase.java:49)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
    at java.lang.Thread.run(Thread.java:748)
</code></pre><ul>
<li>I notice that I get different JSESSIONID cookies for <code>/</code> (XMLUI) and <code>/jspui</code> (JSPUI) on Tomcat 8.5.37, I wonder if it&rsquo;s the same on Tomcat 7.0.92&hellip; yes I do.</li>
<li>Hmm, on Tomcat 7.0.92 I see that I get a <code>dspace.current.user.id</code> session cookie after logging into XMLUI, and then when I browse to JSPUI I am still logged in&hellip;
<ul>
<li>I didn&rsquo;t see that cookie being set on Tomcat 8.5.37</li>
</ul>
</li>
<li>I sent a message to the dspace-tech mailing list to ask</li>
</ul>
<h2 id="2019-01-04">2019-01-04</h2>
<ul>
<li>Linode sent a message last night that CGSpace (linode18) had high CPU usage, but I don&rsquo;t see anything around that time in the web server logs:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;03/Jan/2019:1(7|8|9)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    189 207.46.13.192
    217 31.6.77.23
    340 66.249.70.29
    349 40.77.167.86
    417 34.218.226.147
    630 207.46.13.173
    710 35.237.175.180
    790 40.77.167.87
   1776 66.249.70.27
   2099 54.70.40.11
</code></pre><ul>
<li>I&rsquo;m thinking about trying to validate our <code>dc.subject</code> terms against <a href="http://aims.fao.org/agrovoc/webservices">AGROVOC webservices</a></li>
<li>There seem to be a few APIs and the documentation is kinda confusing, but I found this REST endpoint that does work well, for example searching for <code>SOIL</code>:</li>
</ul>
<pre><code>$ http http://agrovoc.uniroma2.it/agrovoc/rest/v1/search?query=SOIL&amp;lang=en
HTTP/1.1 200 OK
Access-Control-Allow-Origin: *
Connection: Keep-Alive
Content-Length: 493
Content-Type: application/json; charset=utf-8
Date: Fri, 04 Jan 2019 13:44:27 GMT
Keep-Alive: timeout=5, max=100
Server: Apache
Strict-Transport-Security: max-age=63072000; includeSubdomains
Vary: Accept
X-Content-Type-Options: nosniff
X-Frame-Options: ALLOW-FROM http://aims.fao.org

{
    &quot;@context&quot;: {
        &quot;@language&quot;: &quot;en&quot;,
        &quot;altLabel&quot;: &quot;skos:altLabel&quot;,
        &quot;hiddenLabel&quot;: &quot;skos:hiddenLabel&quot;,
        &quot;isothes&quot;: &quot;http://purl.org/iso25964/skos-thes#&quot;,
        &quot;onki&quot;: &quot;http://schema.onki.fi/onki#&quot;,
        &quot;prefLabel&quot;: &quot;skos:prefLabel&quot;,
        &quot;results&quot;: {
            &quot;@container&quot;: &quot;@list&quot;,
            &quot;@id&quot;: &quot;onki:results&quot;
        },
        &quot;skos&quot;: &quot;http://www.w3.org/2004/02/skos/core#&quot;,
        &quot;type&quot;: &quot;@type&quot;,
        &quot;uri&quot;: &quot;@id&quot;
    },
    &quot;results&quot;: [
        {
            &quot;lang&quot;: &quot;en&quot;,
            &quot;prefLabel&quot;: &quot;soil&quot;,
            &quot;type&quot;: [
                &quot;skos:Concept&quot;
            ],
            &quot;uri&quot;: &quot;http://aims.fao.org/aos/agrovoc/c_7156&quot;,
            &quot;vocab&quot;: &quot;agrovoc&quot;
        }
    ],
    &quot;uri&quot;: &quot;&quot;
}
</code></pre><ul>
<li>The API does not appear to be case sensitive (searches for <code>SOIL</code> and <code>soil</code> return the same thing)</li>
<li>I&rsquo;m a bit confused that there&rsquo;s no obvious return code or status when a term is not found, for example <code>SOILS</code>:</li>
</ul>
<pre><code>HTTP/1.1 200 OK
Access-Control-Allow-Origin: *
Connection: Keep-Alive
Content-Length: 367
Content-Type: application/json; charset=utf-8
Date: Fri, 04 Jan 2019 13:48:31 GMT
Keep-Alive: timeout=5, max=100
Server: Apache
Strict-Transport-Security: max-age=63072000; includeSubdomains
Vary: Accept
X-Content-Type-Options: nosniff
X-Frame-Options: ALLOW-FROM http://aims.fao.org

{
    &quot;@context&quot;: {
        &quot;@language&quot;: &quot;en&quot;,
        &quot;altLabel&quot;: &quot;skos:altLabel&quot;,
        &quot;hiddenLabel&quot;: &quot;skos:hiddenLabel&quot;,
        &quot;isothes&quot;: &quot;http://purl.org/iso25964/skos-thes#&quot;,
        &quot;onki&quot;: &quot;http://schema.onki.fi/onki#&quot;,
        &quot;prefLabel&quot;: &quot;skos:prefLabel&quot;,
        &quot;results&quot;: {
            &quot;@container&quot;: &quot;@list&quot;,
            &quot;@id&quot;: &quot;onki:results&quot;
        },
        &quot;skos&quot;: &quot;http://www.w3.org/2004/02/skos/core#&quot;,
        &quot;type&quot;: &quot;@type&quot;,
        &quot;uri&quot;: &quot;@id&quot;
    },
    &quot;results&quot;: [],
    &quot;uri&quot;: &quot;&quot;
}
</code></pre><ul>
<li>I guess the <code>results</code> object will just be empty&hellip;</li>
<li>Another way would be to try with SPARQL, perhaps using the Python 2.7 <a href="https://pypi.org/project/sparql-client/">sparql-client</a>:</li>
</ul>
<pre><code>$ python2.7 -m virtualenv /tmp/sparql
$ . /tmp/sparql/bin/activate
$ pip install sparql-client ipython
$ ipython
In [10]: import sparql
In [11]: s = sparql.Service(&quot;http://agrovoc.uniroma2.it:3030/agrovoc/sparql&quot;, &quot;utf-8&quot;, &quot;GET&quot;)
In [12]: statement=('PREFIX skos: &lt;http://www.w3.org/2004/02/skos/core#&gt; '
    ...: 'SELECT '
    ...: '?label '
    ...: 'WHERE { '
    ...: '{  ?concept  skos:altLabel ?label . } UNION {  ?concept  skos:prefLabel ?label . } '
    ...: 'FILTER regex(str(?label), &quot;^fish&quot;, &quot;i&quot;) . '
    ...: '} LIMIT 10')
In [13]: result = s.query(statement)
In [14]: for row in result.fetchone():
   ...:     print(row)
   ...:
(&lt;Literal &quot;fish catching&quot;@en&gt;,)
(&lt;Literal &quot;fish harvesting&quot;@en&gt;,)
(&lt;Literal &quot;fish meat&quot;@en&gt;,)
(&lt;Literal &quot;fish roe&quot;@en&gt;,)
(&lt;Literal &quot;fish conversion&quot;@en&gt;,)
(&lt;Literal &quot;fisheries catches (composition)&quot;@en&gt;,)
(&lt;Literal &quot;fishtail palm&quot;@en&gt;,)
(&lt;Literal &quot;fishflies&quot;@en&gt;,)
(&lt;Literal &quot;fishery biology&quot;@en&gt;,)
(&lt;Literal &quot;fish production&quot;@en&gt;,)
</code></pre><ul>
<li>The SPARQL query comes from my notes in <a href="/cgspace-notes/2017-08/">2017-08</a></li>
</ul>
<h2 id="2019-01-06">2019-01-06</h2>
<ul>
<li>I built a clean DSpace 5.8 installation from the upstream <code>dspace-5.8</code> tag and the issue with the XMLUI/JSPUI login is still there with Tomcat 8.5.37
<ul>
<li>If I log into XMLUI and then nagivate to JSPUI I need to log in again</li>
<li>XMLUI does not set the <code>dspace.current.user.id</code> session cookie in Tomcat 8.5.37 for some reason</li>
<li>I sent an update to the dspace-tech mailing list to ask for more help troubleshooting</li>
</ul>
</li>
</ul>
<h2 id="2019-01-07">2019-01-07</h2>
<ul>
<li>I built a clean DSpace 6.3 installation from the upstream <code>dspace-6.3</code> tag and the issue with the XMLUI/JSPUI login is still there with Tomcat 8.5.37
<ul>
<li>If I log into XMLUI and then nagivate to JSPUI I need to log in again</li>
<li>XMLUI does not set the <code>dspace.current.user.id</code> session cookie in Tomcat 8.5.37 for some reason</li>
<li>I sent an update to the dspace-tech mailing list to ask for more help troubleshooting</li>
</ul>
</li>
</ul>
<h2 id="2019-01-08">2019-01-08</h2>
<ul>
<li>Tim Donohue responded to my thread about the cookies on the dspace-tech mailing list
<ul>
<li>He suspects it&rsquo;s a change of behavior in Tomcat 8.5, and indeed I see a mention of new cookie processing in the <a href="https://tomcat.apache.org/migration-85.html#Cookies">Tomcat 8.5 migration guide</a></li>
<li>I tried to switch my XMLUI and JSPUI contexts to use the <code>LegacyCookieProcessor</code>, but it didn&rsquo;t seem to help</li>
<li>I <a href="https://jira.duraspace.org/browse/DS-4140">filed DS-4140 on the DSpace issue tracker</a></li>
</ul>
</li>
</ul>
<h2 id="2019-01-11">2019-01-11</h2>
<ul>
<li>Tezira wrote to say she has stopped receiving the <code>DSpace Submission Approved and Archived</code> emails from CGSpace as of January 2nd
<ul>
<li>I told her that I haven&rsquo;t done anything to disable it lately, but that I would check</li>
<li>Bizu also says she hasn&rsquo;t received them lately</li>
</ul>
</li>
</ul>
<h2 id="2019-01-14">2019-01-14</h2>
<ul>
<li>Day one of CGSpace AReS meeting in Amman</li>
</ul>
<h2 id="2019-01-15">2019-01-15</h2>
<ul>
<li>Day two of CGSpace AReS meeting in Amman
<ul>
<li>Discuss possibly extending the <a href="https://github.com/ilri/dspace-statistics-api">dspace-statistics-api</a> to make community and collection statistics available</li>
<li>Discuss new &ldquo;final&rdquo; CG Core document and some changes that we&rsquo;ll need to do on CGSpace and other repositories</li>
<li>We agreed to try to stick to pure Dublin Core where possible, then use fields that exist in standard DSpace, and use &ldquo;cg&rdquo; namespace for everything else</li>
<li>Major changes are to move <code>dc.contributor.author</code> to <code>dc.creator</code> (which MELSpace and WorldFish are already using in their DSpace repositories)</li>
</ul>
</li>
<li>I am testing the speed of the WorldFish DSpace repository&rsquo;s REST API and it&rsquo;s five to ten times faster than CGSpace as I tested in <a href="/cgspace-notes/2018-10/">2018-10</a>:</li>
</ul>
<pre><code>$ time http --print h 'https://digitalarchive.worldfishcenter.org/rest/items?expand=metadata,bitstreams,parentCommunityList&amp;limit=100&amp;offset=0'

0.16s user 0.03s system 3% cpu 5.185 total
0.17s user 0.02s system 2% cpu 7.123 total
0.18s user 0.02s system 6% cpu 3.047 total
</code></pre><ul>
<li>In other news, Linode sent a mail last night that the CPU load on CGSpace (linode18) was high, here are the top IPs in the logs around those few hours:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;14/Jan/2019:(17|18|19|20)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    157 31.6.77.23
    192 54.70.40.11
    202 66.249.64.157
    207 40.77.167.204
    220 157.55.39.140
    326 197.156.105.116
    385 207.46.13.158
   1211 35.237.175.180
   1830 66.249.64.155
   2482 45.5.186.2
</code></pre><h2 id="2019-01-16">2019-01-16</h2>
<ul>
<li>Day three of CGSpace AReS meeting in Amman
<ul>
<li>We discussed CG Core 2.0 metadata and decided some action points</li>
<li>We discussed branding of AReS tool</li>
</ul>
</li>
<li>Notes from our CG Core 2.0 metadata discussion:
<ul>
<li>Not Dublin Core:
<ul>
<li>dc.subtype</li>
<li>dc.peer-reviewed</li>
</ul>
</li>
<li>Dublin Core, possible action for CGSpace:
<ul>
<li>dc.description:
<ul>
<li>We use dc.description.abstract, dc.description (Notes), dc.description.version (Peer review status), dc.description.sponsorship (Funder)</li>
<li>Maybe move abstract to dc.description</li>
<li>Maybe notes moves to cg.description.notes???</li>
<li>Maybe move dc.description.version to cg.peer-reviewed or cg.peer-review-status???</li>
<li>Move dc.description.sponsorship to cg.contributor.donor???</li>
</ul>
</li>
<li>dc.subject:
<ul>
<li>Wait for guidance, evaluate technical implications (Google indexing, OAI, etc)</li>
</ul>
</li>
<li>Move dc.contributor.author to dc.creator</li>
<li>dc.contributor Project
<ul>
<li>Recommend against creating new fields for all projects</li>
<li>We use collections projects/themes/etc</li>
</ul>
</li>
<li>dc.contributor Project Lead Center
<ul>
<li>MELSpace uses cg.contributor.project-lead-institute (institute is more generic than center)</li>
<li>Maybe we use?</li>
</ul>
</li>
<li>dc.contributor Partner
<ul>
<li>Wait for guidance</li>
<li>MELSpace uses cg.contibutor.center (?)</li>
</ul>
</li>
<li>dc.contributor Donor
<ul>
<li>Use cg.contributor.donor</li>
</ul>
</li>
<li>dc.date
<ul>
<li>Wait for guidance, maybe move dc.date.issued?</li>
<li>dc.date.accessioned and dc.date.available are automatic in DSpace</li>
</ul>
</li>
<li>dc.language
<ul>
<li>Move dc.language.iso to dc.language</li>
</ul>
</li>
<li>dc.identifier
<ul>
<li>Move cg.identifier.url to dc.identifier</li>
</ul>
</li>
<li>dc.identifier  bibliographicCitation
<ul>
<li>dc.identifier.citation should move to dc.bibliographicCitation</li>
</ul>
</li>
<li>dc.description.notes
<ul>
<li>Wait for guidance, maybe move to cg.description.notes ???</li>
</ul>
</li>
<li>dc.relation
<ul>
<li>Maybe move cg.link.reference</li>
<li>Perhaps consolodate cg.link.audio etc there&hellip;?</li>
</ul>
</li>
<li>dc.relation.isPartOf
<ul>
<li>Move dc.relation.ispartofseries to dc.relation.isPartOf</li>
</ul>
</li>
<li>dc.audience
<ul>
<li>Move cg.targetaudience to dc.audience</li>
</ul>
</li>
</ul>
</li>
</ul>
</li>
<li>Something happened to the Solr usage statistics on CGSpace
<ul>
<li>I looked on the server and the Solr cores are there (56GB!), and I don&rsquo;t see any obvious errors in dmesg or anything</li>
<li>I see that the server hasn&rsquo;t been rebooted in 26 days so I rebooted it</li>
</ul>
</li>
<li>After reboot the Solr stats are still messed up in the Atmire Usage Stats module, it only shows 2019-01!</li>
</ul>
<p><img src="/cgspace-notes/2019/01/solr-stats-incorrect.png" alt="Solr stats fucked up"></p>
<ul>
<li>In the Solr admin UI I see the following error:</li>
</ul>
<pre><code>statistics-2018: org.apache.solr.common.SolrException:org.apache.solr.common.SolrException: Error opening new searcher
</code></pre><ul>
<li>Looking in the Solr log I see this:</li>
</ul>
<pre><code>2019-01-16 13:37:55,395 ERROR org.apache.solr.core.CoreContainer @ Error creating core [statistics-2018]: Error opening new searcher
org.apache.solr.common.SolrException: Error opening new searcher
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:873)
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:646)
    at org.apache.solr.core.CoreContainer.create(CoreContainer.java:491)
    at org.apache.solr.core.CoreContainer.create(CoreContainer.java:466)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleCreateAction(CoreAdminHandler.java:575)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleRequestInternal(CoreAdminHandler.java:199)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleRequestBody(CoreAdminHandler.java:188)
    at org.apache.solr.handler.RequestHandlerBase.handleRequest(RequestHandlerBase.java:135)
    at org.apache.solr.servlet.SolrDispatchFilter.handleAdminRequest(SolrDispatchFilter.java:729)
    at org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:258)
    at org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:207)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:241)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:208)
    at org.dspace.solr.filters.LocalHostRestrictionFilter.doFilter(LocalHostRestrictionFilter.java:50)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:241)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:208)
    at org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:221)
    at org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:122)
    at org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:505)
    at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:169)
    at org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:103)
    at org.apache.catalina.valves.CrawlerSessionManagerValve.invoke(CrawlerSessionManagerValve.java:180)
    at org.apache.catalina.valves.AccessLogValve.invoke(AccessLogValve.java:956)
    at org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:116)
    at org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:436)
    at org.apache.coyote.http11.AbstractHttp11Processor.process(AbstractHttp11Processor.java:1078)
    at org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:625)
    at org.apache.tomcat.util.net.JIoEndpoint$SocketProcessor.run(JIoEndpoint.java:316)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
    at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.solr.common.SolrException: Error opening new searcher
    at org.apache.solr.core.SolrCore.openNewSearcher(SolrCore.java:1565)
    at org.apache.solr.core.SolrCore.getSearcher(SolrCore.java:1677)
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:845)
    ... 31 more
Caused by: org.apache.lucene.store.LockObtainFailedException: Lock obtain timed out: NativeFSLock@/home/cgspace.cgiar.org/solr/statistics-2018/data/index/write.lock
    at org.apache.lucene.store.Lock.obtain(Lock.java:89)
    at org.apache.lucene.index.IndexWriter.&lt;init&gt;(IndexWriter.java:753)
    at org.apache.solr.update.SolrIndexWriter.&lt;init&gt;(SolrIndexWriter.java:77)
    at org.apache.solr.update.SolrIndexWriter.create(SolrIndexWriter.java:64)
    at org.apache.solr.update.DefaultSolrCoreState.createMainIndexWriter(DefaultSolrCoreState.java:279)
    at org.apache.solr.update.DefaultSolrCoreState.getIndexWriter(DefaultSolrCoreState.java:111)
    at org.apache.solr.core.SolrCore.openNewSearcher(SolrCore.java:1528)
    ... 33 more
2019-01-16 13:37:55,401 ERROR org.apache.solr.core.SolrCore @ org.apache.solr.common.SolrException: Error CREATEing SolrCore 'statistics-2018': Unable to create core [statistics-2018] Caused by: Lock obtain timed out: NativeFSLock@/home/cgspace.cgiar.org/solr/statistics-2018/data/index/write.lock
    at org.apache.solr.handler.admin.CoreAdminHandler.handleCreateAction(CoreAdminHandler.java:613)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleRequestInternal(CoreAdminHandler.java:199)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleRequestBody(CoreAdminHandler.java:188)
    at org.apache.solr.handler.RequestHandlerBase.handleRequest(RequestHandlerBase.java:135)
    at org.apache.solr.servlet.SolrDispatchFilter.handleAdminRequest(SolrDispatchFilter.java:729)
    at org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:258)
    at org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:207)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:241)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:208)
    at org.dspace.solr.filters.LocalHostRestrictionFilter.doFilter(LocalHostRestrictionFilter.java:50)
    at org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:241)
    at org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:208)
    at org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:221)
    at org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:122)
    at org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:505)
    at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:169)
    at org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:103)
    at org.apache.catalina.valves.CrawlerSessionManagerValve.invoke(CrawlerSessionManagerValve.java:180)
    at org.apache.catalina.valves.AccessLogValve.invoke(AccessLogValve.java:956)
    at org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:116)
    at org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:436)
    at org.apache.coyote.http11.AbstractHttp11Processor.process(AbstractHttp11Processor.java:1078)
    at org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:625)
    at org.apache.tomcat.util.net.JIoEndpoint$SocketProcessor.run(JIoEndpoint.java:316)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
    at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.solr.common.SolrException: Unable to create core [statistics-2018]
    at org.apache.solr.core.CoreContainer.create(CoreContainer.java:507)
    at org.apache.solr.core.CoreContainer.create(CoreContainer.java:466)
    at org.apache.solr.handler.admin.CoreAdminHandler.handleCreateAction(CoreAdminHandler.java:575)
    ... 27 more
Caused by: org.apache.solr.common.SolrException: Error opening new searcher
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:873)
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:646)
    at org.apache.solr.core.CoreContainer.create(CoreContainer.java:491)
    ... 29 more
Caused by: org.apache.solr.common.SolrException: Error opening new searcher
    at org.apache.solr.core.SolrCore.openNewSearcher(SolrCore.java:1565)
    at org.apache.solr.core.SolrCore.getSearcher(SolrCore.java:1677)
    at org.apache.solr.core.SolrCore.&lt;init&gt;(SolrCore.java:845)
    ... 31 more
Caused by: org.apache.lucene.store.LockObtainFailedException: Lock obtain timed out: NativeFSLock@/home/cgspace.cgiar.org/solr/statistics-2018/data/index/write.lock
    at org.apache.lucene.store.Lock.obtain(Lock.java:89)
    at org.apache.lucene.index.IndexWriter.&lt;init&gt;(IndexWriter.java:753)
    at org.apache.solr.update.SolrIndexWriter.&lt;init&gt;(SolrIndexWriter.java:77)
    at org.apache.solr.update.SolrIndexWriter.create(SolrIndexWriter.java:64)
    at org.apache.solr.update.DefaultSolrCoreState.createMainIndexWriter(DefaultSolrCoreState.java:279)
    at org.apache.solr.update.DefaultSolrCoreState.getIndexWriter(DefaultSolrCoreState.java:111)
    at org.apache.solr.core.SolrCore.openNewSearcher(SolrCore.java:1528)
    ... 33 more
</code></pre><ul>
<li>I found some threads on StackOverflow etc discussing this and several suggested increasing the address space for the shell with ulimit</li>
<li>I added <code>ulimit -v unlimited</code> to the <code>/etc/default/tomcat7</code> and restarted Tomcat and now Solr is working again:</li>
</ul>
<p><img src="/cgspace-notes/2019/01/solr-stats-incorrect.png" alt="Solr stats working"></p>
<ul>
<li>Some StackOverflow discussions related to this:
<ul>
<li><a href="https://stackoverflow.com/questions/2895417/solrexception-internal-server-error/3035916#3035916">https://stackoverflow.com/questions/2895417/solrexception-internal-server-error/3035916#3035916</a></li>
<li><a href="https://stackoverflow.com/questions/11683850/how-much-memory-could-vm-use">https://stackoverflow.com/questions/11683850/how-much-memory-could-vm-use</a></li>
<li><a href="https://stackoverflow.com/questions/8892143/error-when-opening-a-lucene-index-map-failed/8893684#8893684">https://stackoverflow.com/questions/8892143/error-when-opening-a-lucene-index-map-failed/8893684#8893684</a></li>
</ul>
</li>
<li>Abenet was asking if the Atmire Usage Stats are correct because they are over 2 million the last few months&hellip;</li>
<li>For 2019-01 alone the Usage Stats are already around 1.2 million</li>
<li>I tried to look in the nginx logs to see how many raw requests there are so far this month and it&rsquo;s about 1.4 million:</li>
</ul>
<pre><code># time zcat --force /var/log/nginx/* | grep -cE &quot;[0-9]{1,2}/Jan/2019&quot;
1442874

real    0m17.161s
user    0m16.205s
sys     0m2.396s
</code></pre><h2 id="2019-01-17">2019-01-17</h2>
<ul>
<li>Send reminder to Atmire about purchasing the <a href="https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=657">MQM module</a></li>
<li>Trying to decide the solid action points for CGSpace on the CG Core 2.0 metadata&hellip;</li>
<li>It&rsquo;s difficult to decide some of these because the current CG Core 2.0 document does not provide guidance or rationale (yet)!</li>
<li>Also, there is not a good Dublin Core reference (or maybe I just don&rsquo;t understand?)</li>
<li>Several authoritative documents on Dublin Core appear to be:
<ul>
<li><a href="http://dublincore.org/documents/dces/">Dublin Core Metadata Element Set, Version 1.1: Reference Description</a></li>
<li><a href="http://www.dublincore.org/documents/dcmi-terms/">DCMI Metadata Terms</a></li>
</ul>
</li>
<li>And what is the relationship between DC and DCTERMS?</li>
<li>DSpace uses DCTERMS in the metadata it embeds in XMLUI item views!</li>
<li>We really need to look at this more carefully and see the impacts that might be made from switching core fields like languages, abstract, authors, etc</li>
<li>We can check WorldFish and MELSpace repositories to see what effects these changes have had on theirs because they have already adopted some of these changes&hellip;</li>
<li>I think I understand the difference between DC and DCTERMS finally: DC is the original set of fifteen elements and DCTERMS is the newer version that was supposed to address much of the drawbacks of the original with regards to digital content</li>
<li>We might be able to use some proper fields for citation, abstract, etc that are part of DCTERMS</li>
<li>To make matters more confusing, there is also &ldquo;qualified Dublin Core&rdquo; that uses the original fifteen elements of legacy DC and qualifies them, like <code>dc.date.accessioned</code>
<ul>
<li>According to Wikipedia <a href="https://en.wikipedia.org/wiki/Dublin_Core">Qualified Dublin Core was superseded by DCTERMS in 2008</a>!</li>
</ul>
</li>
<li>So we should be trying to use DCTERMS where possible, unless it is some internal thing that might mess up DSpace (like dates)</li>
<li>&ldquo;Elements 1.1&rdquo; means legacy DC</li>
<li>Possible action list for CGSpace:
<ul>
<li>dc.description.abstract → dcterms.abstract</li>
<li>dc.description.version → cg.peer-reviewed (or cg.peer-review-status?)</li>
<li>dc.description.sponsorship → cg.contributor.donor</li>
<li>dc.contributor.author → dc.creator</li>
<li>dc.language.iso → dcterms.language</li>
<li>cg.identifier.url → dcterms.identifier</li>
<li>dc.identifier.citation → dcterms.bibliographicCitation</li>
<li>dc.relation.ispartofseries → dcterms.isPartOf</li>
<li>cg.targetaudience → dcterms.audience</li>
</ul>
</li>
</ul>
<h2 id="2019-01-19">2019-01-19</h2>
<ul>
<li>
<p>There&rsquo;s no official set of Dublin Core qualifiers so I can&rsquo;t tell if things like <code>dc.contributor.author</code> that are used by DSpace are official</p>
</li>
<li>
<p>I found a great <a href="https://www.dri.ie/sites/default/files/files/qualified-dublin-core-metadata-guidelines.pdf">presentation from 2015 by the Digital Repository of Ireland</a> that discusses using MARC Relator Terms with Dublin Core elements</p>
</li>
<li>
<p>It seems that <code>dc.contributor.author</code> would be a supported term according to this <a href="https://memory.loc.gov/diglib/loc.terms/relators/dc-contributor.html">Library of Congress list</a> linked from the <a href="http://dublincore.org/usage/documents/relators/">Dublin Core website</a></p>
</li>
<li>
<p>The Library of Congress document specifically says:</p>
<p>These terms conform with the DCMI Abstract Model and may be used in DCMI application profiles. DCMI endorses their use with Dublin Core elements as indicated.</p>
</li>
</ul>
<h2 id="2019-01-20">2019-01-20</h2>
<ul>
<li>That&rsquo;s weird, I logged into DSpace Test (linode19) and it says it has been up for 213 days:</li>
</ul>
<pre><code># w
 04:46:14 up 213 days,  7:25,  4 users,  load average: 1.94, 1.50, 1.35
</code></pre><ul>
<li>I&rsquo;ve definitely rebooted it several times in the past few months&hellip; according to <code>journalctl -b</code> it was a few weeks ago on 2019-01-02</li>
<li>I re-ran the Ansible DSpace tag, ran all system updates, and rebooted the host</li>
<li>After rebooting I notice that the Linode kernel went down from 4.19.8 to 4.18.16&hellip;</li>
<li>Atmire sent a quote on our <a href="https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=657">ticket about purchasing the Metadata Quality Module (MQM) for DSpace 5.8</a></li>
<li>Abenet asked me for an <a href="https://cgspace.cgiar.org/open-search/discover?query=crpsubject:Livestock&amp;sort_by=3&amp;order=DESC">OpenSearch query that could generate and RSS feed for items in the Livestock CRP</a></li>
<li>According to my notes, <code>sort_by=3</code> is accession date (as configured in `dspace.cfg)</li>
<li>The query currently shows 3023 items, but a <a href="https://cgspace.cgiar.org/discover?filtertype_1=crpsubject&amp;filter_relational_operator_1=equals&amp;filter_1=Livestock&amp;submit_apply_filter=&amp;query=">Discovery search for Livestock CRP only returns 858 items</a></li>
<li>That query seems to return items tagged with <code>Livestock and Fish</code> CRP as well&hellip; hmm.</li>
</ul>
<h2 id="2019-01-21">2019-01-21</h2>
<ul>
<li>Investigating running Tomcat 7 on Ubuntu 18.04 with the tarball and a custom systemd package instead of waiting for our DSpace to get compatible with Ubuntu 18.04&rsquo;s Tomcat 8.5</li>
<li>I could either run with a simple <code>tomcat7.service</code> like this:</li>
</ul>
<pre><code>[Unit]
Description=Apache Tomcat 7 Web Application Container
After=network.target
[Service]
Type=forking
ExecStart=/path/to/apache-tomcat-7.0.92/bin/startup.sh
ExecStop=/path/to/apache-tomcat-7.0.92/bin/shutdown.sh
User=aorth
Group=aorth
[Install]
WantedBy=multi-user.target
</code></pre><ul>
<li>Or try to use adapt a real systemd service like Arch Linux&rsquo;s:</li>
</ul>
<pre><code>[Unit]
Description=Tomcat 7 servlet container
After=network.target

[Service]
Type=forking
PIDFile=/var/run/tomcat7.pid
Environment=CATALINA_PID=/var/run/tomcat7.pid
Environment=TOMCAT_JAVA_HOME=/usr/lib/jvm/default-runtime
Environment=CATALINA_HOME=/usr/share/tomcat7
Environment=CATALINA_BASE=/usr/share/tomcat7
Environment=CATALINA_OPTS=
Environment=ERRFILE=SYSLOG
Environment=OUTFILE=SYSLOG

ExecStart=/usr/bin/jsvc \
            -Dcatalina.home=${CATALINA_HOME} \
            -Dcatalina.base=${CATALINA_BASE} \
            -Djava.io.tmpdir=/var/tmp/tomcat7/temp \
            -cp /usr/share/java/commons-daemon.jar:/usr/share/java/eclipse-ecj.jar:${CATALINA_HOME}/bin/bootstrap.jar:${CATALINA_HOME}/bin/tomcat-juli.jar \
            -user tomcat7 \
            -java-home ${TOMCAT_JAVA_HOME} \
            -pidfile /var/run/tomcat7.pid \
            -errfile ${ERRFILE} \
            -outfile ${OUTFILE} \
            $CATALINA_OPTS \
            org.apache.catalina.startup.Bootstrap

ExecStop=/usr/bin/jsvc \
            -pidfile /var/run/tomcat7.pid \
            -stop \
            org.apache.catalina.startup.Bootstrap

[Install]
WantedBy=multi-user.target
</code></pre><ul>
<li>I see that <code>jsvc</code> and <code>libcommons-daemon-java</code> are both available on Ubuntu so that should be easy to port</li>
<li>We probably don&rsquo;t need Eclipse Java Bytecode Compiler (ecj)</li>
<li>I tested Tomcat 7.0.92 on Arch Linux using the <code>tomcat7.service</code> with <code>jsvc</code> and it works&hellip; nice!</li>
<li>I think I might manage this the same way I do the restic releases in the <a href="https://github.com/ilri/rmg-ansible-public">Ansible infrastructure scripts</a>, where I download a specific version and symlink to some generic location without the version number</li>
<li>I verified that there is indeed an issue with sharded Solr statistics cores on DSpace, which will cause inaccurate results in the dspace-statistics-api:</li>
</ul>
<pre><code>$ http 'http://localhost:3000/solr/statistics/select?indent=on&amp;rows=0&amp;q=type:2+id:11576&amp;fq=isBot:false&amp;fq=statistics_type:view' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;33&quot; start=&quot;0&quot;&gt;
$ http 'http://localhost:3000/solr/statistics-2018/select?indent=on&amp;rows=0&amp;q=type:2+id:11576&amp;fq=isBot:false&amp;fq=statistics_type:view' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;241&quot; start=&quot;0&quot;&gt;
</code></pre><ul>
<li>I opened an issue on the GitHub issue tracker (<a href="https://github.com/ilri/dspace-statistics-api/issues/10">#10</a>)</li>
<li>I don&rsquo;t think the <a href="https://solrclient.readthedocs.io/en/latest/">SolrClient library</a> we are currently using supports these type of queries so we might have to just do raw queries with requests</li>
<li>The <a href="https://github.com/django-haystack/pysolr">pysolr</a> library says it supports multicore indexes, but I am not sure it does (or at least not with our setup):</li>
</ul>
<pre><code>import pysolr
solr = pysolr.Solr('http://localhost:3000/solr/statistics')
results = solr.search('type:2', **{'fq': 'isBot:false AND statistics_type:view', 'facet': 'true', 'facet.field': 'id', 'facet.mincount': 1, 'facet.limit': 10, 'facet.offset': 0, 'rows': 0})
print(results.facets['facet_fields'])
{'id': ['77572', 646, '93185', 380, '92932', 375, '102499', 372, '101430', 337, '77632', 331, '102449', 289, '102485', 276, '100849', 270, '47080', 260]}
</code></pre><ul>
<li>If I double check one item from above, for example <code>77572</code>, it appears this is only working on the current statistics core and not the shards:</li>
</ul>
<pre><code>import pysolr
solr = pysolr.Solr('http://localhost:3000/solr/statistics')
results = solr.search('type:2 id:77572', **{'fq': 'isBot:false AND statistics_type:view'})
print(results.hits)
646
solr = pysolr.Solr('http://localhost:3000/solr/statistics-2018/')
results = solr.search('type:2 id:77572', **{'fq': 'isBot:false AND statistics_type:view'})
print(results.hits)
595
</code></pre><ul>
<li>So I guess I need to figure out how to use join queries and maybe even switch to using raw Python requests with JSON</li>
<li>This enumerates the list of Solr cores and returns JSON format:</li>
</ul>
<pre><code>http://localhost:3000/solr/admin/cores?action=STATUS&amp;wt=json
</code></pre><ul>
<li>I think I figured out how to search across shards, I needed to give the whole URL to each other core</li>
<li>Now I get more results when I start adding the other statistics cores:</li>
</ul>
<pre><code>$ http 'http://localhost:3000/solr/statistics/select?&amp;indent=on&amp;rows=0&amp;q=*:*' | grep numFound&lt;result name=&quot;response&quot; numFound=&quot;2061320&quot; start=&quot;0&quot;&gt;
$ http 'http://localhost:3000/solr/statistics/select?&amp;shards=localhost:8081/solr/statistics-2018&amp;indent=on&amp;rows=0&amp;q=*:*' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;16280292&quot; start=&quot;0&quot; maxScore=&quot;1.0&quot;&gt;
$ http 'http://localhost:3000/solr/statistics/select?&amp;shards=localhost:8081/solr/statistics-2018,localhost:8081/solr/statistics-2017&amp;indent=on&amp;rows=0&amp;q=*:*' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;25606142&quot; start=&quot;0&quot; maxScore=&quot;1.0&quot;&gt;
$ http 'http://localhost:3000/solr/statistics/select?&amp;shards=localhost:8081/solr/statistics-2018,localhost:8081/solr/statistics-2017,localhost:8081/solr/statistics-2016&amp;indent=on&amp;rows=0&amp;q=*:*' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;31532212&quot; start=&quot;0&quot; maxScore=&quot;1.0&quot;&gt;
</code></pre><ul>
<li>I should be able to modify the dspace-statistics-api to check the shards via the Solr core status, then add the <code>shards</code> parameter to each query to make the search distributed among the cores</li>
<li>I implemented a proof of concept to query the Solr STATUS for active cores and to add them with a <code>shards</code> query string</li>
<li>A few things I noticed:
<ul>
<li>Solr doesn&rsquo;t mind if you use an empty <code>shards</code> parameter</li>
<li>Solr doesn&rsquo;t mind if you have an extra comma at the end of the <code>shards</code> parameter</li>
<li>If you are searching multiple cores, you need to include the base core in the <code>shards</code> parameter as well</li>
<li>For example, compare the following two queries, first including the base core and the shard in the <code>shards</code> parameter, and then only including the shard:</li>
</ul>
</li>
</ul>
<pre><code>$ http 'http://localhost:8081/solr/statistics/select?indent=on&amp;rows=0&amp;q=type:2+id:11576&amp;fq=isBot:false&amp;fq=statistics_type:view&amp;shards=localhost:8081/solr/statistics,localhost:8081/solr/statistics-2018' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;275&quot; start=&quot;0&quot; maxScore=&quot;12.205825&quot;&gt;
$ http 'http://localhost:8081/solr/statistics/select?indent=on&amp;rows=0&amp;q=type:2+id:11576&amp;fq=isBot:false&amp;fq=statistics_type:view&amp;shards=localhost:8081/solr/statistics-2018' | grep numFound
&lt;result name=&quot;response&quot; numFound=&quot;241&quot; start=&quot;0&quot; maxScore=&quot;12.205825&quot;&gt;
</code></pre><h2 id="2019-01-22">2019-01-22</h2>
<ul>
<li>Release <a href="https://github.com/ilri/dspace-statistics-api/releases/tag/v0.9.0">version 0.9.0 of the dspace-statistics-api</a> to address the issue of querying multiple Solr statistics shards</li>
<li>I deployed it on DSpace Test (linode19) and restarted the indexer and now it shows all the stats from 2018 as well (756 pages of views, intead of 6)</li>
<li>I deployed it on CGSpace (linode18) and restarted the indexer as well</li>
<li>Linode sent an alert that CGSpace (linode18) was using high CPU this afternoon, the top ten IPs during that time were:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;22/Jan/2019:1(4|5|6)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    155 40.77.167.106
    176 2003:d5:fbda:1c00:1106:c7a0:4b17:3af8
    189 107.21.16.70
    217 54.83.93.85
    310 46.174.208.142
    346 83.103.94.48
    360 45.5.186.2
    595 154.113.73.30
    716 196.191.127.37
    915 35.237.175.180
</code></pre><ul>
<li>35.237.175.180 is known to us</li>
<li>I don&rsquo;t think we&rsquo;ve seen 196.191.127.37 before. Its user agent is:</li>
</ul>
<pre><code>Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.87 UBrowser/7.0.185.1002 Safari/537.36
</code></pre><ul>
<li>Interestingly this IP is located in Addis Ababa&hellip;</li>
<li>Another interesting one is 154.113.73.30, which is apparently at IITA Nigeria and uses the user agent:</li>
</ul>
<pre><code>Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36
</code></pre><h2 id="2019-01-23">2019-01-23</h2>
<ul>
<li>Peter noticed that some goo.gl links in our tweets from Feedburner are broken, for example this one from last week:</li>
</ul>
<blockquote class="twitter-tweet"><p lang="en" dir="ltr"><a href="https://twitter.com/hashtag/ILRI?src=hash&amp;ref_src=twsrc%5Etfw">#ILRI</a> research:  Towards unlocking the potential of the hides and skins value chain in Somaliland <a href="https://t.co/EZH7ALW4dp">https://t.co/EZH7ALW4dp</a></p>&mdash; ILRI Communications (@ILRI) <a href="https://twitter.com/ILRI/status/1086330519904673793?ref_src=twsrc%5Etfw">January 18, 2019</a></blockquote>
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

<ul>
<li>The shortened link is <a href="goo.gl/fb/VRj9Gq">goo.gl/fb/VRj9Gq</a> and it shows a &ldquo;Dynamic Link not found&rdquo; error from Firebase:</li>
</ul>
<p><img src="/cgspace-notes/2019/01/firebase-link-not-found.png" alt="Dynamic Link not found"></p>
<ul>
<li>
<p>Apparently Google announced last year that they plan to <a href="https://developers.googleblog.com/2018/03/transitioning-google-url-shortener.html">discontinue the shortner and transition to Firebase Dynamic Links in March, 2019</a>, so maybe this is related&hellip;</p>
</li>
<li>
<p>Very interesting discussion of methods for <a href="https://jdebp.eu/FGA/systemd-house-of-horror/tomcat.html">running Tomcat under systemd</a></p>
</li>
<li>
<p>We can set the ulimit options that used to be in <code>/etc/default/tomcat7</code> with systemd&rsquo;s <code>LimitNOFILE</code> and <code>LimitAS</code> (see the <code>systemd.exec</code> man page)</p>
<ul>
<li>Note that we need to use <code>infinity</code> instead of <code>unlimited</code> for the address space</li>
</ul>
</li>
<li>
<p>Create accounts for Bosun from IITA and Valerio from ICARDA / CGMEL on DSpace Test</p>
</li>
<li>
<p>Maria Garruccio asked me for a list of author affiliations from all of their submitted items so she can clean them up</p>
</li>
<li>
<p>I got a list of their collections from the CGSpace XMLUI and then used an SQL query to dump the unique values to CSV:</p>
</li>
</ul>
<pre><code>dspace=# \copy (select distinct text_value, count(*) from metadatavalue where metadata_field_id = (select metadata_field_id from metadatafieldregistry where element = 'contributor' and qualifier = 'affiliation') AND resource_type_id = 2 AND resource_id IN (select item_id from collection2item where collection_id IN (select resource_id from handle where handle in ('10568/35501', '10568/41728', '10568/49622', '10568/56589', '10568/56592', '10568/65064', '10568/65718', '10568/65719', '10568/67373', '10568/67731', '10568/68235', '10568/68546', '10568/69089', '10568/69160', '10568/69419', '10568/69556', '10568/70131', '10568/70252', '10568/70978'))) group by text_value order by count desc) to /tmp/bioversity-affiliations.csv with csv;
COPY 1109
</code></pre><ul>
<li>Send a mail to the dspace-tech mailing list about the OpenSearch issue we had with the Livestock CRP</li>
<li>Linode sent an alert that CGSpace (linode18) had a high load this morning, here are the top ten IPs during that time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;23/Jan/2019:0(4|5|6)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    222 54.226.25.74
    241 40.77.167.13
    272 46.101.86.248
    297 35.237.175.180
    332 45.5.184.72
    355 34.218.226.147
    404 66.249.64.155
   4637 205.186.128.185
   4637 70.32.83.92
   9265 45.5.186.2
</code></pre><ul>
<li>
<p>I think it&rsquo;s the usual IPs:</p>
<ul>
<li>45.5.186.2 is CIAT</li>
<li>70.32.83.92 is CCAFS</li>
<li>205.186.128.185 is CCAFS or perhaps another Macaroni Bros harvester (new ILRI website?)</li>
</ul>
</li>
<li>
<p>Following up on the thumbnail issue that we had in <a href="/cgspace-notes/2018-12/">2018-12</a></p>
</li>
<li>
<p>It looks like the two items with problematic PDFs both have thumbnails now:</p>
<ul>
<li><a href="https://hdl.handle.net/10568/98390">10568/98390</a></li>
<li><a href="https://hdl.handle.net/10568/98391">10568/98391</a></li>
</ul>
</li>
<li>
<p>Just to make sure these were not uploaded by the user or something, I manually forced the regeneration of these with DSpace&rsquo;s <code>filter-media</code>:</p>
</li>
</ul>
<pre><code>$ schedtool -D -e ionice -c2 -n7 nice -n19 /home/cgspace.cgiar.org/bin/dspace filter-media -v -f -i 10568/98390
$ schedtool -D -e ionice -c2 -n7 nice -n19 /home/cgspace.cgiar.org/bin/dspace filter-media -v -f -i 10568/98391
</code></pre><ul>
<li>Both of these were successful, so there must have been an update to ImageMagick or Ghostscript in Ubuntu since early 2018-12</li>
<li>Looking at the apt history logs I see that on 2018-12-07 a security update for Ghostscript was installed (version 9.26~dfsg+0-0ubuntu0.16.04.3)</li>
<li>I think this Launchpad discussion is relevant: <a href="https://bugs.launchpad.net/ubuntu/+source/ghostscript/+bug/1806517">https://bugs.launchpad.net/ubuntu/+source/ghostscript/+bug/1806517</a></li>
<li>As well as the original Ghostscript bug report: <a href="https://bugs.ghostscript.com/show_bug.cgi?id=699815">https://bugs.ghostscript.com/show_bug.cgi?id=699815</a></li>
</ul>
<h2 id="2019-01-24">2019-01-24</h2>
<ul>
<li>I noticed Ubuntu&rsquo;s Ghostscript 9.26 works on some troublesome PDFs where Arch&rsquo;s Ghostscript 9.26 doesn&rsquo;t, so the fix for the first/last page crash is not the patch I found yesterday</li>
<li>Ubuntu&rsquo;s Ghostscript uses another <a href="http://git.ghostscript.com/?p=ghostpdl.git;h=fae21f1668d2b44b18b84cf0923a1d5f3008a696">patch from Ghostscript git</a> (<a href="https://bugs.ghostscript.com/show_bug.cgi?id=700315">upstream bug report</a>)</li>
<li>I re-compiled Arch&rsquo;s ghostscript with the patch and then I was able to generate a thumbnail from one of the <a href="https://cgspace.cgiar.org/handle/10568/98390">troublesome PDFs</a></li>
<li>Before and after:</li>
</ul>
<pre><code>$ identify Food\ safety\ Kenya\ fruits.pdf\[0\]
zsh: abort (core dumped)  identify Food\ safety\ Kenya\ fruits.pdf\[0\]
$ identify Food\ safety\ Kenya\ fruits.pdf\[0\]
Food safety Kenya fruits.pdf[0]=&gt;Food safety Kenya fruits.pdf PDF 612x792 612x792+0+0 16-bit sRGB 64626B 0.000u 0:00.000
identify: CorruptImageProfile `xmp' @ warning/profile.c/SetImageProfileInternal/1747.
</code></pre><ul>
<li>I reported it to the Arch Linux bug tracker (<a href="https://bugs.archlinux.org/task/61513">61513</a>)</li>
<li>I told Atmire to go ahead with the Metadata Quality Module addition based on our <code>5_x-dev</code> branch (<a href="https://tracker.atmire.com/tickets-cgiar-ilri/view-ticket?id=657">657</a>)</li>
<li>Linode sent alerts last night to say that CGSpace (linode18) was using high CPU last night, here are the top ten IPs from the nginx logs around that time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;23/Jan/2019:(18|19|20)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    305 3.81.136.184
    306 3.83.14.11
    306 52.54.252.47
    325 54.221.57.180
    378 66.249.64.157
    424 54.70.40.11
    497 47.29.247.74
    783 35.237.175.180
   1108 66.249.64.155
   2378 45.5.186.2
</code></pre><ul>
<li>45.5.186.2 is CIAT and 66.249.64.155 is Google&hellip; hmmm.</li>
<li>Linode sent another alert this morning, here are the top ten IPs active during that time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;24/Jan/2019:0(4|5|6)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    360 3.89.134.93
    362 34.230.15.139
    366 100.24.48.177
    369 18.212.208.240
    377 3.81.136.184
    404 54.221.57.180
    506 66.249.64.155
   4642 70.32.83.92
   4643 205.186.128.185
   8593 45.5.186.2
</code></pre><ul>
<li>Just double checking what CIAT is doing, they are mainly hitting the REST API:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;24/Jan/2019:&quot; | grep 45.5.186.2 | grep -Eo &quot;GET /(handle|bitstream|rest|oai)/&quot; | sort | uniq -c | sort -n
</code></pre><ul>
<li>CIAT&rsquo;s community currently has 12,000 items in it so this is normal</li>
<li>The issue with goo.gl links that we saw yesterday appears to be resolved, as links are working again&hellip;</li>
<li>For example: <a href="https://goo.gl/fb/VRj9Gq">https://goo.gl/fb/VRj9Gq</a></li>
<li>The full <a href="http://id.loc.gov/vocabulary/relators.html">list of MARC Relators on the Library of Congress website</a> linked from the <a href="http://dublincore.org/usage/documents/relators/">DMCI relators page</a> is very confusing</li>
<li>Looking at the default DSpace XMLUI crosswalk in <a href="https://github.com/DSpace/DSpace/blob/dspace-5_x/dspace/config/crosswalks/xhtml-head-item.properties">xhtml-head-item.properties</a> I see a very complete mapping of DSpace DC and QDC fields to DCTERMS
<ul>
<li>This is good for standards-compliant web crawlers, but what about for those harvesting via REST or OAI APIs?</li>
</ul>
</li>
<li>I sent a message titled &ldquo;<a href="https://groups.google.com/forum/#!topic/dspace-tech/phV_t51TGuE">DC, QDC, and DCTERMS: reviewing our metadata practices</a>&rdquo; to the dspace-tech mailing list to ask about some of this</li>
</ul>
<h2 id="2019-01-25">2019-01-25</h2>
<ul>
<li>A little bit more work on getting Tomcat to run from a tarball on our <a href="https://github.com/ilri/rmg-ansible-public">Ansible infrastructure playbooks</a>
<ul>
<li>I tested by doing a Tomcat 7.0.91 installation, then switching it to 7.0.92 and it worked&hellip; nice!</li>
<li>I refined the tasks so much that I was confident enough to deploy them on DSpace Test and it went very well</li>
<li>Basically I just stopped tomcat7, created a dspace user, removed tomcat7, chown&rsquo;d everything to the dspace user, then ran the playbook</li>
<li>So now DSpace Test (linode19) is running Tomcat 7.0.92&hellip; w00t</li>
<li>Now we need to monitor it for a few weeks to see if there is anything we missed, and then I can change CGSpace (linode18) as well, and we&rsquo;re ready for Ubuntu 18.04 too!</li>
</ul>
</li>
</ul>
<h2 id="2019-01-27">2019-01-27</h2>
<ul>
<li>Linode sent an email that the server was using a lot of CPU this morning, and these were the top IPs in the web server logs at the time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;27/Jan/2019:0(6|7|8)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    189 40.77.167.108
    191 157.55.39.2
    263 34.218.226.147
    283 45.5.184.2
    332 45.5.184.72
    608 5.9.6.51
    679 66.249.66.223
   1116 66.249.66.219
   4644 205.186.128.185
   4644 70.32.83.92
</code></pre><ul>
<li>I think it&rsquo;s the usual IPs:
<ul>
<li>70.32.83.92 is CCAFS</li>
<li>205.186.128.185 is CCAFS or perhaps another Macaroni Bros harvester (new ILRI website?)</li>
</ul>
</li>
</ul>
<h2 id="2019-01-28">2019-01-28</h2>
<ul>
<li>Udana from WLE asked me about the interaction between their publication website and their items on CGSpace
<ul>
<li>There is an item that is mapped into their collection from IWMI and is missing their <code>cg.identifier.wletheme</code> metadata</li>
<li>I told him that, as far as I remember, when WLE introduced Phase II research themes in 2017 we decided to infer theme ownership from the collection hierarchy and we created a <a href="https://cgspace.cgiar.org/handle/10568/81268">WLE Phase II Research Themes</a> subCommunity</li>
<li>Perhaps they need to ask Macaroni Bros about the mapping</li>
</ul>
</li>
<li>Linode alerted that CGSpace (linode18) was using too much CPU again this morning, here are the active IPs from the web server log at the time:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;28/Jan/2019:0(6|7|8)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
     67 207.46.13.50
    105 41.204.190.40
    117 34.218.226.147
    126 35.237.175.180
    203 213.55.99.121
    332 45.5.184.72
    377 5.9.6.51
    512 45.5.184.2
   4644 205.186.128.185
   4644 70.32.83.92
</code></pre><ul>
<li>There seems to be a pattern with <code>70.32.83.92</code> and <code>205.186.128.185</code> lately!</li>
<li>Every morning at 8AM they are the top users&hellip; I should tell them to stagger their requests&hellip;</li>
<li>I signed up for a <a href="https://visualping.io/">VisualPing</a> of the <a href="https://jdbc.postgresql.org/download.html">PostgreSQL JDBC driver download page</a> to my CGIAR email address
<ul>
<li>Hopefully this will one day alert me that a new driver is released!</li>
</ul>
</li>
<li>Last night Linode sent an alert that CGSpace (linode18) was using high CPU, here are the most active IPs in the hours just before, during, and after the alert:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;28/Jan/2019:(17|18|19|20|21)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    310 45.5.184.2
    425 5.143.231.39
    526 54.70.40.11
   1003 199.47.87.141
   1374 35.237.175.180
   1455 5.9.6.51
   1501 66.249.66.223
   1771 66.249.66.219
   2107 199.47.87.140
   2540 45.5.186.2
</code></pre><ul>
<li>Of course there is CIAT&rsquo;s <code>45.5.186.2</code>, but also <code>45.5.184.2</code> appears to be CIAT&hellip; I wonder why they have two harvesters?</li>
<li><code>199.47.87.140</code> and <code>199.47.87.141</code> is TurnItIn with the following user agent:</li>
</ul>
<pre><code>TurnitinBot (https://turnitin.com/robot/crawlerinfo.html)
</code></pre><h2 id="2019-01-29">2019-01-29</h2>
<ul>
<li>Linode sent an alert about CGSpace (linode18) CPU usage this morning, here are the top IPs in the web server logs just before, during, and after the alert:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;29/Jan/2019:0(3|4|5|6|7)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    334 45.5.184.72
    429 66.249.66.223
    522 35.237.175.180
    555 34.218.226.147
    655 66.249.66.221
    844 5.9.6.51
   2507 66.249.66.219
   4645 70.32.83.92
   4646 205.186.128.185
   9329 45.5.186.2
</code></pre><ul>
<li><code>45.5.186.2</code> is CIAT as usual&hellip;</li>
<li><code>70.32.83.92</code> and <code>205.186.128.185</code> are CCAFS as usual&hellip;</li>
<li><code>66.249.66.219</code> is Google&hellip;</li>
<li>I&rsquo;m thinking it might finally be time to increase the threshold of the Linode CPU alerts
<ul>
<li>I adjusted the alert threshold from 250% to 275%</li>
</ul>
</li>
</ul>
<h2 id="2019-01-30">2019-01-30</h2>
<ul>
<li>Got another alert from Linode about CGSpace (linode18) this morning, here are the top IPs before, during, and after the alert:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;30/Jan/2019:0(5|6|7|8|9)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    273 46.101.86.248
    301 35.237.175.180
    334 45.5.184.72
    387 5.9.6.51
    527 2a01:4f8:13b:1296::2
   1021 34.218.226.147
   1448 66.249.66.219
   4649 205.186.128.185
   4649 70.32.83.92
   5163 45.5.184.2
</code></pre><ul>
<li>I might need to adjust the threshold again, because the load average this morning was 296% and the activity looks pretty normal (as always recently)</li>
</ul>
<h2 id="2019-01-31">2019-01-31</h2>
<ul>
<li>Linode sent alerts about CGSpace (linode18) last night and this morning, here are the top IPs before, during, and after those times:</li>
</ul>
<pre><code># zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;30/Jan/2019:(16|17|18|19|20)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    436 18.196.196.108
    460 157.55.39.168
    460 207.46.13.96
    500 197.156.105.116
    728 54.70.40.11
   1560 5.9.6.51
   1562 35.237.175.180
   1601 85.25.237.71
   1894 66.249.66.219
   2610 45.5.184.2
# zcat --force /var/log/nginx/*.log /var/log/nginx/*.log.1 | grep -E &quot;31/Jan/2019:0(2|3|4|5|6)&quot; | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
    318 207.46.13.242
    334 45.5.184.72
    486 35.237.175.180
    609 34.218.226.147
    620 66.249.66.219
   1054 5.9.6.51
   4391 70.32.83.92
   4428 205.186.128.185
   6758 85.25.237.71
   9239 45.5.186.2
</code></pre><ul>
<li><code>45.5.186.2</code> and <code>45.5.184.2</code> are CIAT as always</li>
<li><code>85.25.237.71</code> is some new server in Germany that I&rsquo;ve never seen before with the user agent:</li>
</ul>
<pre><code>Linguee Bot (http://www.linguee.com/bot; bot@linguee.com)
</code></pre><!-- raw HTML omitted -->

  

  

</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/2020-07/">July, 2020</a></li>

<li><a href="/cgspace-notes/2020-06/">June, 2020</a></li>

<li><a href="/cgspace-notes/2020-05/">May, 2020</a></li>

<li><a href="/cgspace-notes/2020-04/">April, 2020</a></li>

<li><a href="/cgspace-notes/2020-03/">March, 2020</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 dir="auto">
      
      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>