<?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: Can too many stale PV extents crash system? in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859319#M937151</link>
    <description>Certainly no limit that I'm aware of, but you have to consider the fact that the longer the disk goes unreplaced, the more time there is for the working disk to develop a fault, which may not in itself cause a crash, but will definitely be a problem.</description>
    <pubDate>Fri, 06 Dec 2002 16:48:19 GMT</pubDate>
    <dc:creator>Chris Wilshaw</dc:creator>
    <dc:date>2002-12-06T16:48:19Z</dc:date>
    <item>
      <title>Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859315#M937147</link>
      <description>Just wondering... L2000, failed mirrored root disk.  Will replace disk in two days.  Is there any type of LVM limit on the number of stale extents allowed on a PV?  The stale extend count is currently 303 on this particular PV which has been failed for 3 days, so I am assuming I am not going to hit some kind of limit.  But what if I let this bad disk sit here forever - I assume I will eventually hit some type of limit on stale extents.  The answer to this question is particulary to me because we have some disks that are hit very hard, and if we loose one of those mirrored disks I would like to know how long we have before terrible LVM things start happening if we hit a PV limit on stale extents.</description>
      <pubDate>Fri, 06 Dec 2002 16:40:02 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859315#M937147</guid>
      <dc:creator>Byron Myers</dc:creator>
      <dc:date>2002-12-06T16:40:02Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859316#M937148</link>
      <description>No, theres no limit ive ever heard of. They can remain stale forever until you fix them or remove them. Ive had stales for many many weeks and they wont cause a crash (as long as your patches are uptodate)&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 06 Dec 2002 16:44:21 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859316#M937148</guid>
      <dc:creator>Stefan Farrelly</dc:creator>
      <dc:date>2002-12-06T16:44:21Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859317#M937149</link>
      <description>Hi:&lt;BR /&gt;&lt;BR /&gt;I don't know of any limit or trigger-point for the crash you fear, but why wait to replace the failed disk(?).  I'd want to fix it and remirror before the *other* copy started to fail [Murphy's Law].&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt;&lt;BR /&gt;...JRF...</description>
      <pubDate>Fri, 06 Dec 2002 16:45:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859317#M937149</guid>
      <dc:creator>James R. Ferguson</dc:creator>
      <dc:date>2002-12-06T16:45:47Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859318#M937150</link>
      <description>No, there's no limit on the number of stale extents.  You can have an entire disk go bad and thus have ALL extents on a disk as stale.&lt;BR /&gt;&lt;BR /&gt;This is the whole purpose of mirroring.  If there were some limit on the number of stale extents, why bother mirroring.&lt;BR /&gt;&lt;BR /&gt;I, too, have had an LV running with a bad disk for several weeks.  If your patches are up to date you shouldn't have a problem.</description>
      <pubDate>Fri, 06 Dec 2002 16:47:34 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859318#M937150</guid>
      <dc:creator>Patrick Wallek</dc:creator>
      <dc:date>2002-12-06T16:47:34Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859319#M937151</link>
      <description>Certainly no limit that I'm aware of, but you have to consider the fact that the longer the disk goes unreplaced, the more time there is for the working disk to develop a fault, which may not in itself cause a crash, but will definitely be a problem.</description>
      <pubDate>Fri, 06 Dec 2002 16:48:19 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859319#M937151</guid>
      <dc:creator>Chris Wilshaw</dc:creator>
      <dc:date>2002-12-06T16:48:19Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859320#M937152</link>
      <description>Hi Byron,&lt;BR /&gt;&lt;BR /&gt;Only time I had problem with a failed disk causing problem to the system was that an unmirrored secondary swap was defined on the mirror disk by the previous administrator.&lt;BR /&gt;&lt;BR /&gt;You may want to check if you had such a configuration.&lt;BR /&gt;&lt;BR /&gt;-Sri</description>
      <pubDate>Fri, 06 Dec 2002 16:50:31 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859320#M937152</guid>
      <dc:creator>Sridhar Bhaskarla</dc:creator>
      <dc:date>2002-12-06T16:50:31Z</dc:date>
    </item>
    <item>
      <title>Re: Can too many stale PV extents crash system?</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859321#M937153</link>
      <description>Thanks for the replies.  Patrick's comment about LVM allowing all extents in a failed PV to go stale moved my thinking in the right direction.  We always schedule down time when replacing mirrored disks even though we can do this online.  These systems are offsite, and I am counting on HW support to correctly identify the correct failed disk.  HW support is very good, but its a balancing act on replacing now or waiting a few days.  Another Murphy-ism is that the good disk could get replaced.  &lt;BR /&gt;</description>
      <pubDate>Fri, 06 Dec 2002 16:57:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/can-too-many-stale-pv-extents-crash-system/m-p/2859321#M937153</guid>
      <dc:creator>Byron Myers</dc:creator>
      <dc:date>2002-12-06T16:57:38Z</dc:date>
    </item>
  </channel>
</rss>

