<?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: Ignite - SCSI port mismatch. in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415995#M764786</link>
    <description>David,&lt;BR /&gt;&lt;BR /&gt;The make_recovery portion of the Ignite-UX product was intended for recovery of &lt;BR /&gt;vg00 on the same system that it was created for. It is understandable that &lt;BR /&gt;customers choose to use this functionality to make ignite images of an existing &lt;BR /&gt;"base" system to install new systems with this same "base". However, you need &lt;BR /&gt;to very careful when doing so. Since, make_recovery is expecting to do its &lt;BR /&gt;restoring on the exact same system, it expects to be dealing with the same &lt;BR /&gt;hardware paths and the same hardware addresses (even in networking). For this &lt;BR /&gt;reason, it is recommended that using make_recovery tapes on other systems &lt;BR /&gt;should be reserved only for those systems of the same (or very similar) &lt;BR /&gt;hardware architectures. You also need to make sure that the networking &lt;BR /&gt;parameters are not included or else that will result in duplicate MAC Addresses &lt;BR /&gt;and/or IP Addresses on your network. &lt;BR /&gt;&lt;BR /&gt;If you are planning to use make_recovery to ignite another system this will &lt;BR /&gt;require a lot of planning and foresight. You will also probably need to run &lt;BR /&gt;make_recovery with the -p option to allow you to preview and to manually edit &lt;BR /&gt;the files /var/opt/ignite/recovery/arch.include and &lt;BR /&gt;/var/opt/ignite/recovery/config.recover before resumimg with make_recovery -r &lt;BR /&gt;and actually creating the tape. This can be quite involved. You can consult the &lt;BR /&gt;man page for make_recovery for more info. If this is not satisfactory then you &lt;BR /&gt;may consider assistance from a HP Consultant or attending the HP sponsored &lt;BR /&gt;3-day class "Managing Software with Ignite-UX".</description>
    <pubDate>Tue, 16 Nov 1999 09:00:39 GMT</pubDate>
    <dc:creator>Jason Luginbuhl_1</dc:creator>
    <dc:date>1999-11-16T09:00:39Z</dc:date>
    <item>
      <title>Ignite - SCSI port mismatch.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415994#M764785</link>
      <description>We recently received a HP K580 and on booting the machine we were getting these &lt;BR /&gt;errors: &lt;BR /&gt;lvlnboot: Warning: couldn't query physical volume /dev/dsk/c1t6d0&lt;BR /&gt;The path of the physical volume refers to a device that does not exist, or is &lt;BR /&gt;not configured into the kernal.&lt;BR /&gt;volume group "/dev/vg00" has been successfully changed.&lt;BR /&gt;.&lt;BR /&gt;.&lt;BR /&gt;.&lt;BR /&gt;and it goes on. The suggestion is that the correspondence between physical &lt;BR /&gt;volumes and volume groups has been lost. &lt;BR /&gt;&lt;BR /&gt;I've been looking into this problem since 11/11, and&lt;BR /&gt;around 11/12 we had an answer. The original operating&lt;BR /&gt;system was "burned" so that it expected these devices&lt;BR /&gt;on these SCSI channels:&lt;BR /&gt;&lt;BR /&gt;internal disks: channel 1&lt;BR /&gt;DVD-ROM:        channel 2&lt;BR /&gt;disk array      channels 5 and 6&lt;BR /&gt;&lt;BR /&gt;The system was delivered with the SCSI channels on:&lt;BR /&gt;&lt;BR /&gt;internal disks: channel 3&lt;BR /&gt;DVD ROM:        channel 4&lt;BR /&gt;disk arrays:    channels 1 and 2&lt;BR /&gt;&lt;BR /&gt;What's amazing to me, the system is actually booting! But until this problem is &lt;BR /&gt;ironed out, I'm not doing anything with this system, especially anything with&lt;BR /&gt;disks (I have no clue what kind of damage could happen&lt;BR /&gt;if you mirrored a system with mis-mapped physical&lt;BR /&gt;volumes - anyone know?).&lt;BR /&gt;&lt;BR /&gt;So we've called HP hardware to fix this (I'm assuming&lt;BR /&gt;it's just a cable swap).&lt;BR /&gt;&lt;BR /&gt;David.</description>
      <pubDate>Mon, 15 Nov 1999 08:17:54 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415994#M764785</guid>
      <dc:creator>David Myers_3</dc:creator>
      <dc:date>1999-11-15T08:17:54Z</dc:date>
    </item>
    <item>
      <title>Re: Ignite - SCSI port mismatch.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415995#M764786</link>
      <description>David,&lt;BR /&gt;&lt;BR /&gt;The make_recovery portion of the Ignite-UX product was intended for recovery of &lt;BR /&gt;vg00 on the same system that it was created for. It is understandable that &lt;BR /&gt;customers choose to use this functionality to make ignite images of an existing &lt;BR /&gt;"base" system to install new systems with this same "base". However, you need &lt;BR /&gt;to very careful when doing so. Since, make_recovery is expecting to do its &lt;BR /&gt;restoring on the exact same system, it expects to be dealing with the same &lt;BR /&gt;hardware paths and the same hardware addresses (even in networking). For this &lt;BR /&gt;reason, it is recommended that using make_recovery tapes on other systems &lt;BR /&gt;should be reserved only for those systems of the same (or very similar) &lt;BR /&gt;hardware architectures. You also need to make sure that the networking &lt;BR /&gt;parameters are not included or else that will result in duplicate MAC Addresses &lt;BR /&gt;and/or IP Addresses on your network. &lt;BR /&gt;&lt;BR /&gt;If you are planning to use make_recovery to ignite another system this will &lt;BR /&gt;require a lot of planning and foresight. You will also probably need to run &lt;BR /&gt;make_recovery with the -p option to allow you to preview and to manually edit &lt;BR /&gt;the files /var/opt/ignite/recovery/arch.include and &lt;BR /&gt;/var/opt/ignite/recovery/config.recover before resumimg with make_recovery -r &lt;BR /&gt;and actually creating the tape. This can be quite involved. You can consult the &lt;BR /&gt;man page for make_recovery for more info. If this is not satisfactory then you &lt;BR /&gt;may consider assistance from a HP Consultant or attending the HP sponsored &lt;BR /&gt;3-day class "Managing Software with Ignite-UX".</description>
      <pubDate>Tue, 16 Nov 1999 09:00:39 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415995#M764786</guid>
      <dc:creator>Jason Luginbuhl_1</dc:creator>
      <dc:date>1999-11-16T09:00:39Z</dc:date>
    </item>
    <item>
      <title>Re: Ignite - SCSI port mismatch.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415996#M764787</link>
      <description>Jason,&lt;BR /&gt;&lt;BR /&gt;I thank you for your thoughtful response, but both operating systems (Ignite) &lt;BR /&gt;were done by HP, the hardware is essentially out of the box, the HP-PB &lt;BR /&gt;addresses are the same, the machines have the same hardware, but the SCSI &lt;BR /&gt;channel addresses are different. However, the Ignite was made assuming a &lt;BR /&gt;certain match between channels and HP-PB addresses and only one of the machines &lt;BR /&gt;follows these addresses. This is the comparison:&lt;BR /&gt;&lt;BR /&gt;                   Machine 1     Machine 2&lt;BR /&gt;internal disks     c1t5d0        c3t5d0&lt;BR /&gt;                   c1t6d0        c3t6d0&lt;BR /&gt;DVD-ROM            c2t2d0        c4t2d0&lt;BR /&gt;30/FC array        c5t0d0        c1t0d0&lt;BR /&gt;                   c5t0d1        c1t0d1&lt;BR /&gt;                   c5t0d2        c1t0d2&lt;BR /&gt;                   c6t0d0        c2t0d0&lt;BR /&gt;                   c6t0d1        c2t0d1&lt;BR /&gt;                   c6t0d2        c2t0d2&lt;BR /&gt;&lt;BR /&gt;I don't recall all the HP-PB addresses offhand,&lt;BR /&gt;but *both* internal disks are addressed as 10/0.5&lt;BR /&gt;and 10/0.6 respectively. Examination of /etc/lvmtab&lt;BR /&gt;with strings yields the following correspondence:&lt;BR /&gt;&lt;BR /&gt;/dev/vg00&lt;BR /&gt;/dev/dsk/c1t6d0&lt;BR /&gt;&lt;BR /&gt;So machine 2 yields a slew of boot errors: (example)&lt;BR /&gt;.... (booting)&lt;BR /&gt;/sbin/bcheckrc&lt;BR /&gt;lvlnboot: Warning: couldn't query physical volume&lt;BR /&gt;/dev/dsk/c1t6d0&lt;BR /&gt;The path to the physical volume refers to a device&lt;BR /&gt;that does not exist, or is not configured into the&lt;BR /&gt;kernel.&lt;BR /&gt;Volume group "/dev/vg00" has been successfully&lt;BR /&gt;changed.&lt;BR /&gt;vgdisplay: Warning: couldn't query physical volume&lt;BR /&gt;/dev/dsk/c1t6d0. The specified path does not &lt;BR /&gt;correspond to physical volume attached to this&lt;BR /&gt;volume group.&lt;BR /&gt;vgdisplay: Warning: couldn't query all of the &lt;BR /&gt;physical volumes.&lt;BR /&gt;...(continues boot process and comes up 'fine')&lt;BR /&gt;&lt;BR /&gt;I'm not a hardware jock on HP-UX (though I've worked&lt;BR /&gt;with it as a user and an administrator since 1990) and&lt;BR /&gt;exactly how this correspondence between HP-PB addresses and SCSI addresses &lt;BR /&gt;works, I don't know. If it were in&lt;BR /&gt;software or firmware, I'd be happy. But that &lt;BR /&gt;essential bit of mapping eludes me at present.&lt;BR /&gt;&lt;BR /&gt;David.</description>
      <pubDate>Wed, 17 Nov 1999 11:43:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/ignite-scsi-port-mismatch/m-p/2415996#M764787</guid>
      <dc:creator>David Myers_3</dc:creator>
      <dc:date>1999-11-17T11:43:51Z</dc:date>
    </item>
  </channel>
</rss>

