<?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 Re: Kerberos ticket renewal for Hadoop services in Archives of Support Questions (Read Only)</title>
    <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145616#M48443</link>
    <description>&lt;P&gt;Thanks &lt;A rel="user" href="https://community.cloudera.com/users/504/kkulkarni.html" nodeid="504"&gt;@Kuldeep Kulkarni
&lt;/A&gt;&lt;/P&gt;&lt;P&gt;This has cleared some doubts about how Hadoop services auto-login even though initial user TGT shows expired.
&lt;A rel="user" href="https://community.cloudera.com/users/504/kkulkarni.html" nodeid="504"&gt;&lt;/A&gt; &lt;/P&gt;</description>
    <pubDate>Sun, 11 Dec 2016 18:45:55 GMT</pubDate>
    <dc:creator>mhdeshmukh22</dc:creator>
    <dc:date>2016-12-11T18:45:55Z</dc:date>
    <item>
      <title>Kerberos ticket renewal for Hadoop services</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145614#M48441</link>
      <description>&lt;P&gt;I'm having issues with Kerberos tickets for Hadoop services not being renewed before they expire.
E.g the ticket for Oozie, it is valid for 10h and then it takes a certain number of hours until the ticket is renewed(or recreated) again.
I'm expecting there to always be a valid ticket present for the services. &lt;/P&gt;&lt;P&gt;The effect of this is e.g that I can't list directories in HDFS as the Oozie user(in the shell), it fails with the following error message:&lt;/P&gt;&lt;PRE&gt;GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)]&lt;/PRE&gt;&lt;P&gt;I can renew the ticket again manually using the keytab which makes the HDFS listing work but i feel that it shouldn't be necessary.&lt;/P&gt;&lt;P&gt;Strangely enough there are never any service related errors in Ambari.&lt;/P&gt;&lt;P&gt;Any ideas on how to resolve this?&lt;/P&gt;</description>
      <pubDate>Fri, 16 Sep 2022 10:50:33 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145614#M48441</guid>
      <dc:creator>jonas_bissmark</dc:creator>
      <dc:date>2022-09-16T10:50:33Z</dc:date>
    </item>
    <item>
      <title>Re: Kerberos ticket renewal for Hadoop services</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145615#M48442</link>
      <description>&lt;A rel="user" href="https://community.cloudera.com/users/14823/jonasbissmark.html" nodeid="14823"&gt;@Jonas Bissmark&lt;/A&gt;&lt;P&gt;Whenever you start any service in kerberos enabled cluster, let's say Namenode service, first Ambari initiates the kerberos ticket and once service is started, it has logic to re-login and get the fresh ticket.&lt;/P&gt;&lt;P&gt;Comment from &lt;A rel="user" href="https://community.cloudera.com/users/381/cnauroth.html" nodeid="381"&gt;@Chris Nauroth&lt;/A&gt; from stackoverflow question on how Hadoop implements an automatic re-login mechanism directly inside the RPC client layer ( please read his awesome answer on &lt;A href="http://stackoverflow.com/questions/34616676/should-i-call-ugi-checktgtandreloginfromkeytab-before-every-action-on-hadoop" target="_blank"&gt;http://stackoverflow.com/questions/34616676/should-i-call-ugi-checktgtandreloginfromkeytab-before-every-action-on-hadoop&lt;/A&gt; when you get a chance )&lt;/P&gt;&lt;P&gt;####&lt;/P&gt;&lt;P&gt;&lt;STRONG&gt;The code for this is visible in the RPC Client#handleSaslConnectionFailure method:&lt;/STRONG&gt;&lt;/P&gt;&lt;PRE&gt;          // try re-login
          if (UserGroupInformation.isLoginKeytabBased()) {
            UserGroupInformation.getLoginUser().reloginFromKeytab();
          } else if (UserGroupInformation.isLoginTicketBased()) {
            UserGroupInformation.getLoginUser().reloginFromTicketCache();
          }&lt;/PRE&gt;&lt;P&gt;This explains answer to your question - "Strangely enough there are never any service related errors in Ambari."&lt;/P&gt;&lt;P&gt;##&lt;/P&gt;&lt;P&gt;The effect of this is e.g that I can't list directories in HDFS as the Oozie user(in the shell), it fails with the following error message: --&amp;gt; This is expected as ticket gets expired after 24 hours.&lt;/P&gt;&lt;P&gt;Hope this answers your question! &lt;span class="lia-unicode-emoji" title=":slightly_smiling_face:"&gt;🙂&lt;/span&gt;&lt;/P&gt;</description>
      <pubDate>Sat, 10 Dec 2016 06:44:10 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145615#M48442</guid>
      <dc:creator>KuldeepK</dc:creator>
      <dc:date>2016-12-10T06:44:10Z</dc:date>
    </item>
    <item>
      <title>Re: Kerberos ticket renewal for Hadoop services</title>
      <link>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145616#M48443</link>
      <description>&lt;P&gt;Thanks &lt;A rel="user" href="https://community.cloudera.com/users/504/kkulkarni.html" nodeid="504"&gt;@Kuldeep Kulkarni
&lt;/A&gt;&lt;/P&gt;&lt;P&gt;This has cleared some doubts about how Hadoop services auto-login even though initial user TGT shows expired.
&lt;A rel="user" href="https://community.cloudera.com/users/504/kkulkarni.html" nodeid="504"&gt;&lt;/A&gt; &lt;/P&gt;</description>
      <pubDate>Sun, 11 Dec 2016 18:45:55 GMT</pubDate>
      <guid>https://community.cloudera.com/t5/Archives-of-Support-Questions/Kerberos-ticket-renewal-for-Hadoop-services/m-p/145616#M48443</guid>
      <dc:creator>mhdeshmukh22</dc:creator>
      <dc:date>2016-12-11T18:45:55Z</dc:date>
    </item>
  </channel>
</rss>

