cgspace-notes/docs/2015-11/index.html

333 lines
11 KiB
HTML
Raw Normal View History

2018-02-11 17:28:23 +01:00
<!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, 2015" />
<meta property="og:description" content="2015-11-22
CGSpace went down
Looks like DSpace exhausted its PostgreSQL connection pool
2019-05-05 15:45:12 +02:00
Last week I had increased the limit from 30 to 60, which seemed to help, but now there are many more idle connections:
2018-02-11 17:28:23 +01:00
$ psql -c &#39;SELECT * from pg_stat_activity;&#39; | grep idle | grep -c cgspace
78
2019-05-05 15:45:12 +02:00
2018-02-11 17:28:23 +01:00
" />
<meta property="og:type" content="article" />
2019-02-02 13:12:57 +01:00
<meta property="og:url" content="https://alanorth.github.io/cgspace-notes/2015-11/" />
<meta property="article:published_time" content="2015-11-23T17:00:57&#43;03:00"/>
2018-03-09 21:16:20 +01:00
<meta property="article:modified_time" content="2018-03-09T22:10:33&#43;02:00"/>
2018-09-30 07:23:48 +02:00
2018-02-11 17:28:23 +01:00
<meta name="twitter:card" content="summary"/>
<meta name="twitter:title" content="November, 2015"/>
<meta name="twitter:description" content="2015-11-22
CGSpace went down
Looks like DSpace exhausted its PostgreSQL connection pool
2019-05-05 15:45:12 +02:00
Last week I had increased the limit from 30 to 60, which seemed to help, but now there are many more idle connections:
2018-02-11 17:28:23 +01:00
$ psql -c &#39;SELECT * from pg_stat_activity;&#39; | grep idle | grep -c cgspace
78
2019-05-05 15:45:12 +02:00
2018-02-11 17:28:23 +01:00
"/>
2019-06-25 19:10:57 +02:00
<meta name="generator" content="Hugo 0.55.6" />
2018-02-11 17:28:23 +01:00
<script type="application/ld+json">
{
"@context": "http://schema.org",
"@type": "BlogPosting",
"headline": "November, 2015",
2019-04-13 11:15:55 +02:00
"url": "https:\/\/alanorth.github.io\/cgspace-notes\/2015-11\/",
2018-04-30 18:05:39 +02:00
"wordCount": "798",
2019-04-13 11:15:55 +02:00
"datePublished": "2015-11-23T17:00:57\x2b03:00",
"dateModified": "2018-03-09T22:10:33\x2b02:00",
2018-02-11 17:28:23 +01:00
"author": {
"@type": "Person",
"name": "Alan Orth"
},
"keywords": "Notes"
}
</script>
<link rel="canonical" href="https://alanorth.github.io/cgspace-notes/2015-11/">
<title>November, 2015 | CGSpace Notes</title>
<!-- combined, minified CSS -->
2019-02-13 17:47:17 +01:00
<link href="https://alanorth.github.io/cgspace-notes/css/style.css" rel="stylesheet" integrity="sha384-G5B34w7DFTumWTswxYzTX7NWfbvQEg1HbFFEg6ItN03uTAAoS2qkPS/fu3LhuuSA" crossorigin="anonymous">
2018-02-11 17:28:23 +01:00
2019-04-14 15:59:47 +02:00
<!-- RSS 2.0 feed -->
2018-02-11 17:28:23 +01:00
</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>
2018-12-19 12:20:39 +01:00
2018-02-11 17:28:23 +01:00
<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>
2018-12-19 12:20:39 +01:00
2018-02-11 17:28:23 +01:00
<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/2015-11/">November, 2015</a></h2>
<p class="blog-post-meta"><time datetime="2015-11-23T17:00:57&#43;03:00">Mon Nov 23, 2015</time> by Alan Orth in
<i class="fa fa-tag" aria-hidden="true"></i>&nbsp;<a href="/cgspace-notes/tags/notes" rel="tag">Notes</a>
</p>
</header>
<h2 id="2015-11-22">2015-11-22</h2>
<ul>
<li>CGSpace went down</li>
<li>Looks like DSpace exhausted its PostgreSQL connection pool</li>
2019-05-05 15:45:12 +02:00
<li><p>Last week I had increased the limit from 30 to 60, which seemed to help, but now there are many more idle connections:</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | grep idle | grep -c cgspace
78
2019-05-05 15:45:12 +02:00
</code></pre></li>
</ul>
2018-02-11 17:28:23 +01:00
<ul>
<li>For now I have increased the limit from 60 to 90, run updates, and rebooted the server</li>
</ul>
<h2 id="2015-11-24">2015-11-24</h2>
<ul>
<li>CGSpace went down again</li>
<li>Getting emails from uptimeRobot and uptimeButler that it&rsquo;s down, and Google Webmaster Tools is sending emails that there is an increase in crawl errors</li>
2019-05-05 15:45:12 +02:00
<li><p>Looks like there are still a bunch of idle PostgreSQL connections:</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | grep idle | grep -c cgspace
96
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>For some reason the number of idle connections is very high since we upgraded to DSpace 5</p></li>
2018-02-11 17:28:23 +01:00
</ul>
<h2 id="2015-11-25">2015-11-25</h2>
<ul>
<li>Troubleshoot the DSpace 5 OAI breakage caused by nginx routing config</li>
2019-05-05 15:45:12 +02:00
<li><p>The OAI application requests stylesheets and javascript files with the path <code>/oai/static/css</code>, which gets matched here:</p>
2018-02-11 17:28:23 +01:00
<pre><code># static assets we can load from the file system directly with nginx
location ~ /(themes|static|aspects/ReportingSuite) {
2019-05-05 15:45:12 +02:00
try_files $uri @tomcat;
2018-02-11 17:28:23 +01:00
...
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>The document root is relative to the xmlui app, so this gets a 404—I&rsquo;m not sure why it doesn&rsquo;t pass to <code>@tomcat</code></p></li>
<li><p>Anyways, I can&rsquo;t find any URIs with path <code>/static</code>, and the more important point is to handle all the static theme assets, so we can just remove <code>static</code> from the regex for now (who cares if we can&rsquo;t use nginx to send Etags for OAI CSS!)</p></li>
<li><p>Also, I noticed we aren&rsquo;t setting CSP headers on the static assets, because in nginx headers are inherited in child blocks, but if you use <code>add_header</code> in a child block it doesn&rsquo;t inherit the others</p></li>
<li><p>We simply need to add <code>include extra-security.conf;</code> to the above location block (but research and test first)</p></li>
<li><p>We should add WOFF assets to the list of things to set expires for:</p>
2018-02-11 17:28:23 +01:00
<pre><code>location ~* \.(?:ico|css|js|gif|jpe?g|png|woff)$ {
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>We should also add <code>aspects/Statistics</code> to the location block for static assets (minus <code>static</code> from above):</p>
2018-02-11 17:28:23 +01:00
<pre><code>location ~ /(themes|aspects/ReportingSuite|aspects/Statistics) {
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>Need to check <code>/about</code> on CGSpace, as it&rsquo;s blank on my local test server and we might need to add something there</p></li>
<li><p>CGSpace has been up and down all day due to PostgreSQL idle connections (current DSpace pool is 90):</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | grep idle | grep -c cgspace
93
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>I looked closer at the idle connections and saw that many have been idle for hours (current time on server is <code>2015-11-25T20:20:42+0000</code>):</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | less -S
datid | datname | pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | xact_start |
-------+----------+-------+----------+----------+------------------+-------------+-----------------+-------------+-------------------------------+-------------------------------+---
20951 | cgspace | 10966 | 18205 | cgspace | | 127.0.0.1 | | 37731 | 2015-11-25 13:13:02.837624+00 | | 20
20951 | cgspace | 10967 | 18205 | cgspace | | 127.0.0.1 | | 37737 | 2015-11-25 13:13:03.069421+00 | | 20
...
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>There is a relevant Jira issue about this: <a href="https://jira.duraspace.org/browse/DS-1458">https://jira.duraspace.org/browse/DS-1458</a></p></li>
<li><p>It seems there is some sense changing DSpace&rsquo;s default <code>db.maxidle</code> from unlimited (-1) to something like 8 (Tomcat default) or 10 (Confluence default)</p></li>
<li><p>Change <code>db.maxidle</code> from -1 to 10, reduce <code>db.maxconnections</code> from 90 to 50, and restart postgres and tomcat7</p></li>
<li><p>Also redeploy DSpace Test with a clean sync of CGSpace and mirror these database settings there as well</p></li>
<li><p>Also deploy the nginx fixes for the <code>try_files</code> location block as well as the expires block</p></li>
2018-02-11 17:28:23 +01:00
</ul>
<h2 id="2015-11-26">2015-11-26</h2>
<ul>
<li>CGSpace behaving much better since changing <code>db.maxidle</code> yesterday, but still two up/down notices from monitoring this morning (better than 50!)</li>
<li>CCAFS colleagues mentioned that the REST API is very slow, 24 seconds for one item</li>
2019-05-05 15:45:12 +02:00
<li><p>Not as bad for me, but still unsustainable if you have to get many:</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ curl -o /dev/null -s -w %{time_total}\\n https://cgspace.cgiar.org/rest/handle/10568/32802?expand=all
8.415
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>Monitoring e-mailed in the evening to say CGSpace was down</p></li>
<li><p>Idle connections in PostgreSQL again:</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | grep cgspace | grep -c idle
66
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>At the time, the current DSpace pool size was 50&hellip;</p></li>
<li><p>I reduced the pool back to the default of 30, and reduced the <code>db.maxidle</code> settings from 10 to 8</p></li>
2018-02-11 17:28:23 +01:00
</ul>
<h2 id="2015-11-29">2015-11-29</h2>
<ul>
<li>Still more alerts that CGSpace has been up and down all day</li>
2019-05-05 15:45:12 +02:00
<li><p>Current database settings for DSpace:</p>
2018-02-11 17:28:23 +01:00
<pre><code>db.maxconnections = 30
db.maxwait = 5000
db.maxidle = 8
db.statementpool = true
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>And idle connections:</p>
2018-02-11 17:28:23 +01:00
<pre><code>$ psql -c 'SELECT * from pg_stat_activity;' | grep cgspace | grep -c idle
49
2019-05-05 15:45:12 +02:00
</code></pre></li>
2018-02-11 17:28:23 +01:00
2019-05-05 15:45:12 +02:00
<li><p>Perhaps I need to start drastically increasing the connection limits—like to 300—to see if DSpace&rsquo;s thirst can ever be quenched</p></li>
<li><p>On another note, SUNScholar&rsquo;s notes suggest adjusting some other postgres variables: <a href="http://wiki.lib.sun.ac.za/index.php/SUNScholar/Optimisations/Database">http://wiki.lib.sun.ac.za/index.php/SUNScholar/Optimisations/Database</a></p></li>
<li><p>This might help with REST API speed (which I mentioned above and still need to do real tests)</p></li>
2018-02-11 17:28:23 +01:00
</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">
2019-08-04 21:49:04 +02:00
<li><a href="/cgspace-notes/2019-08/">August, 2019</a></li>
2019-04-01 08:02:18 +02:00
2019-04-13 11:15:55 +02:00
<li><a href="/cgspace-notes/posts/">Posts</a></li>
2019-08-04 21:49:04 +02:00
<li><a href="/cgspace-notes/2019-07/">July, 2019</a></li>
2019-07-01 11:22:43 +02:00
<li><a href="/cgspace-notes/2019-06/">June, 2019</a></li>
2019-06-02 09:57:51 +02:00
<li><a href="/cgspace-notes/2019-05/">May, 2019</a></li>
2018-02-11 17:28:23 +01:00
</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>