<?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>topic Re: splunkd causes OutOfMemory when searching data with large event size in Getting Data In</title>
    <link>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756845#M120045</link>
    <description>&lt;BLOCKQUOTE&gt;&lt;HR /&gt;1. 1MByte events are... huge. Whether it is kv-json or plain regex-based extractions, it's gonna be heavy.&lt;HR /&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;I think search-time extraction (aka KV_MODE) assumes a lot of contingencies so it holds a lot more data in memory. &amp;nbsp;That is causing the memory pressure. &amp;nbsp;After taking away KV_MODE, there is no problem in search. &amp;nbsp;I then apply | spath inline. &amp;nbsp;No problem at all. &amp;nbsp;It is actually very performant.&lt;/P&gt;&lt;P&gt;I am starting to learn some difference between implied actions and some evaluation actions. (See a Slack thread about tojson and eval. &amp;nbsp;In this case, spath behaves like eval.) &amp;nbsp;I am guessing that there is a good reason why certain implied actions consumes so much more resource. &amp;nbsp;Maybe that's why those large-event default sourcetypes use index-time extraction instead.&lt;/P&gt;&lt;P&gt;In the end, index-time extraction and inline spath are the only options for such sourcetypes.&lt;/P&gt;</description>
    <pubDate>Tue, 30 Dec 2025 00:49:24 GMT</pubDate>
    <dc:creator>yuanliu</dc:creator>
    <dc:date>2025-12-30T00:49:24Z</dc:date>
    <item>
      <title>splunkd causes OutOfMemory when searching data with large event size</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756841#M120043</link>
      <description>&lt;P&gt;I am onboarding a JSON dataset whose event size is very close to 1MB. &amp;nbsp;I have to increase TRUNCATE to 1000000 (from default of 10000).&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;TRUNCATE = 1000000
KV_MODE = json&lt;/LI-CODE&gt;&lt;P&gt;But when I perform a search on this sourcetype, splunkd's memory demand sky-rockets and causing oom_killer to kill random processes, effectively bringing down the OS.&lt;/P&gt;&lt;P&gt;I see some default sourcetypes (&lt;SPAN&gt;etc/system/defaut/props.conf) already have TRUNCATE=1000000. &amp;nbsp;But they use INDEXED_EXTRACTIONS=json. &amp;nbsp;Is index-time extraction the only/recommended way to handle these exceedingly large events? &amp;nbsp;Is there some formula to determine search time memory need based on event size?&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 29 Dec 2025 20:13:46 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756841#M120043</guid>
      <dc:creator>yuanliu</dc:creator>
      <dc:date>2025-12-29T20:13:46Z</dc:date>
    </item>
    <item>
      <title>Re: splunkd causes OutOfMemory when searching data with large event size</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756842#M120044</link>
      <description>&lt;P&gt;1. 1MByte events are... huge. Whether it is kv-json or plain regex-based extractions, it's gonna be heavy.&lt;/P&gt;&lt;P&gt;2. As a side note - if splunkd brings down whole OS it might be the time to tweak the VMM parameters. (swappiness, zram, oom killer priorities...)&lt;/P&gt;</description>
      <pubDate>Mon, 29 Dec 2025 20:47:04 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756842#M120044</guid>
      <dc:creator>PickleRick</dc:creator>
      <dc:date>2025-12-29T20:47:04Z</dc:date>
    </item>
    <item>
      <title>Re: splunkd causes OutOfMemory when searching data with large event size</title>
      <link>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756845#M120045</link>
      <description>&lt;BLOCKQUOTE&gt;&lt;HR /&gt;1. 1MByte events are... huge. Whether it is kv-json or plain regex-based extractions, it's gonna be heavy.&lt;HR /&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;I think search-time extraction (aka KV_MODE) assumes a lot of contingencies so it holds a lot more data in memory. &amp;nbsp;That is causing the memory pressure. &amp;nbsp;After taking away KV_MODE, there is no problem in search. &amp;nbsp;I then apply | spath inline. &amp;nbsp;No problem at all. &amp;nbsp;It is actually very performant.&lt;/P&gt;&lt;P&gt;I am starting to learn some difference between implied actions and some evaluation actions. (See a Slack thread about tojson and eval. &amp;nbsp;In this case, spath behaves like eval.) &amp;nbsp;I am guessing that there is a good reason why certain implied actions consumes so much more resource. &amp;nbsp;Maybe that's why those large-event default sourcetypes use index-time extraction instead.&lt;/P&gt;&lt;P&gt;In the end, index-time extraction and inline spath are the only options for such sourcetypes.&lt;/P&gt;</description>
      <pubDate>Tue, 30 Dec 2025 00:49:24 GMT</pubDate>
      <guid>https://community.splunk.com/t5/Getting-Data-In/splunkd-causes-OutOfMemory-when-searching-data-with-large-event/m-p/756845#M120045</guid>
      <dc:creator>yuanliu</dc:creator>
      <dc:date>2025-12-30T00:49:24Z</dc:date>
    </item>
  </channel>
</rss>

