<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>question Impala Error: Couldn't open transport in Archives of Support Questions (Read Only)</title>
    <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Impala-Error-Couldn-t-open-transport/m-p/15804#M2363</link>
    <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;When we try to run more complex Impala queries, we often run into the following error:&lt;/P&gt;&lt;P&gt;Couldn't open transport for worker29.ourdomain.com:22000(connect() failed: Connection timed out)&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Sometimes there's only one node with that error message, sometimes there are 2-5.&lt;/P&gt;&lt;P&gt;There doesn't seem to be a network related problem - ping works, telnet to that port works, Impala debug ui works.&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;We tried setting vm.swappiness on the nodes from 60 to 0 - no positive effect. Same with switching vm.overcommit from 0 to 1.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;Our setup:&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;- around 40 nodes, i7 quad core, 2-3TB, 1Gbit NIC, located in 5 different racks&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;&lt;SPAN&gt;- nodes have around 16-48GB ram, same amount of swap, which they alsmost never use&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;&lt;SPAN&gt;- OS: Ubuntu Linux 12.04&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;- CDH 5.1.0&lt;/P&gt;&lt;P&gt;-&amp;nbsp;impalad version 1.4.0-cdh5-INTERNAL RELEASE (build e801bd8c0d134e783c2313c7dd422a5ad06591af)&lt;/P&gt;&lt;P&gt;- ~100TB HDFS storage&lt;/P&gt;&lt;P&gt;- we are using a &lt;A target="_blank" href="http://gethue.com/hadoop-tutorial-how-to-distribute-impala-query-load/"&gt;HA proxy&lt;/A&gt; which points to the nodes with &amp;gt;32GB ram&lt;/P&gt;&lt;P&gt;- "workerlogs"-table is around 6-7TB big, partitioned by year &amp;gt; month &amp;gt; day and contains apache log-data&lt;/P&gt;&lt;P&gt;- almost 100% short circuit reads&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;A target="_blank" href="http://pastebin.com/t5znFfGi"&gt;Query Profile&lt;/A&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Maybe you could give us a hint.&lt;/P&gt;</description>
    <pubDate>Fri, 16 Sep 2022 09:03:01 GMT</pubDate>
    <dc:creator>T-Man</dc:creator>
    <dc:date>2022-09-16T09:03:01Z</dc:date>
    <item>
      <title>Impala Error: Couldn't open transport</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Impala-Error-Couldn-t-open-transport/m-p/15804#M2363</link>
      <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;When we try to run more complex Impala queries, we often run into the following error:&lt;/P&gt;&lt;P&gt;Couldn't open transport for worker29.ourdomain.com:22000(connect() failed: Connection timed out)&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Sometimes there's only one node with that error message, sometimes there are 2-5.&lt;/P&gt;&lt;P&gt;There doesn't seem to be a network related problem - ping works, telnet to that port works, Impala debug ui works.&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;We tried setting vm.swappiness on the nodes from 60 to 0 - no positive effect. Same with switching vm.overcommit from 0 to 1.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;Our setup:&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;- around 40 nodes, i7 quad core, 2-3TB, 1Gbit NIC, located in 5 different racks&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;&lt;SPAN&gt;- nodes have around 16-48GB ram, same amount of swap, which they alsmost never use&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;&lt;SPAN&gt;- OS: Ubuntu Linux 12.04&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;- CDH 5.1.0&lt;/P&gt;&lt;P&gt;-&amp;nbsp;impalad version 1.4.0-cdh5-INTERNAL RELEASE (build e801bd8c0d134e783c2313c7dd422a5ad06591af)&lt;/P&gt;&lt;P&gt;- ~100TB HDFS storage&lt;/P&gt;&lt;P&gt;- we are using a &lt;A target="_blank" href="http://gethue.com/hadoop-tutorial-how-to-distribute-impala-query-load/"&gt;HA proxy&lt;/A&gt; which points to the nodes with &amp;gt;32GB ram&lt;/P&gt;&lt;P&gt;- "workerlogs"-table is around 6-7TB big, partitioned by year &amp;gt; month &amp;gt; day and contains apache log-data&lt;/P&gt;&lt;P&gt;- almost 100% short circuit reads&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;A target="_blank" href="http://pastebin.com/t5znFfGi"&gt;Query Profile&lt;/A&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Maybe you could give us a hint.&lt;/P&gt;</description>
      <pubDate>Fri, 16 Sep 2022 09:03:01 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Impala-Error-Couldn-t-open-transport/m-p/15804#M2363</guid>
      <dc:creator>T-Man</dc:creator>
      <dc:date>2022-09-16T09:03:01Z</dc:date>
    </item>
    <item>
      <title>Re: Impala Error: Couldn't open transport</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Impala-Error-Couldn-t-open-transport/m-p/15916#M2364</link>
      <description>We found cause for the error - our firewall settings were to restrictive. Interestingly smaller queries without many query fragments worked even with these restrictive settings.</description>
      <pubDate>Thu, 24 Jul 2014 10:13:26 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Impala-Error-Couldn-t-open-transport/m-p/15916#M2364</guid>
      <dc:creator>T-Man</dc:creator>
      <dc:date>2014-07-24T10:13:26Z</dc:date>
    </item>
  </channel>
</rss>

