Persistent reservation clarification needed

Ralph Weber ralphoweber at
Sun Jun 2 07:16:36 PDT 2002

* From the T10 Reflector (t10 at, posted by:
* Ralph Weber <ralphoweber at>
I agree with Gerry. This editorial change will appear in SPC-3 r07.



-------- Original Message --------
Subject: Re: Persistent reservation clarification needed
Date: Wed, 22 May 2002 09:37:20 -0500
From: Gerry.Houlder at
To: t10 at

* From the T10 Reflector (t10 at, posted by:
* Gerry.Houlder at

A "device server" is required to remember persistent reservations (and the
obsolete reservations) on a per LUN basis. The APTPL bit should only apply
to that specific LUN, regardless of whether it is a one or a zero. It seems
like we have an editorial problem with the APTPL=0 wording.

"Ulrich, David" wrote:

I'm looking for some background information that would help explain the
rationale for the following, found on page 135 of SPC-3 revision 7.

If the last valid APTPL bit value received by the device server is zero,
the loss of power in the SCSI target device
shall release the persistent reservation for all logical units and remove
all registered reservation keys (see

If the last valid APTPL bit value received by the device server is one, the
logical unit shall retain any persistent reservation(

s) that may be present and all reservation keys (i.e., registrations) for
all initiator ports even if power is lost
and later returned (see

Specifically, why does an APTPL value of 0 affect all luns while a value of
1 address only the lun the command was issued to?

Thanks in advance,

David Ulrich
LSI Logic Storage Systems Division

* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at

More information about the T10 mailing list