*******************************************************************************
* *
* This is an update to an existing patch... *
* *
* Online links can be found at *
* http://ftp.support.compaq.com/patches/public/vms/axp/v7.2-1h1/dec-axpvms-vms721h1_shadowing-v0300--4.README
*******************************************************************************

TITLE: OpenVMS VMS721H1_SHADOWING-V0300 ALPHA V7.2-1H1 SHADOWING ECO Summary

New Kit Date : 08-FEB-2001
Modification Date: Not Applicable
Modification Type: Updated Kit Supersedes VMS721H1_SHADOWING-V0200
 
NOTE: An OpenVMS saveset or PCSI installation file is stored
       on the Internet in a self-expanding compressed file.
 
       For OpenVMS savesets, the name of the compressed saveset
       file will be kit_name.a-dcx_vaxexe for OpenVMS VAX or
       kit_name.a-dcx_axpexe for OpenVMS Alpha. Once the OpenVMS
       saveset is copied to your system, expand the compressed
       saveset by typing RUN kitname.dcx_vaxexe or kitname.dcx_alpexe.
 
       For PCSI files, once the PCSI file is copied to your system,
       rename the PCSI file to kitname-dcx_axpexe.pcsi, then it can
       be expanded by typing RUN kitname-dcx_axpexe.pcsi. The resultant
       file will be the PCSI installation file which can be used to install
       the ECO.
 

Copyright (c) Compaq Computer Corporation 2000. All rights reserved.

OP/SYS: OpenVMS Alpha

COMPONENT: Volume Shadowing

SOURCE: Compaq Computer Corporation

ECO INFORMATION:

     ECO Kit Name: VMS721H1_SHADOWING-V0300
                    DEC-AXPVMS-VMS721H1_SHADOWING-V0300--4.PCSI
     ECO Kits Superseded by This ECO Kit: VMS721H1_SHADOWING-V0200
     ECO Kit Approximate Size: 5760 Blocks
     Kit Applies To: OpenVMS Alpha V7.2-1H1
     System/Cluster Reboot Necessary: Yes
     Rolling Re-boot Supported: Yes
     Installation Rating: INSTALL_2
                           2 - To be installed on all systems running
                               the listed version(s) of OpenVMS and
                               using the following feature(s):

                                SHADOWING

     Kit Dependencies:

       The following remedial kit(s) must be installed BEFORE
       installation of this kit:

         VMS721H1_UPDATE-V0300

       In order to receive all the corrections listed in this
       kit, the following remedial kits should also be installed:

         None

ECO KIT SUMMARY:

An ECO kit exists for Shadowing on OpenVMS Alpha V7.2-1H1. This kit
addresses the following problems:

Problems Addressed In VMS721H1_SHADOWING-V0300 KIT:

   o The VMS721H1_SHADOWING-V0200 images would not install over the
      VMS721H1_SHADOWING-V0100 images. The shadowing images in the
      VMS721H1_SHADOWING-V0200 kit should have been the same images
      as those in the VMS721H1_SHADOWING-V0100 kit. Instead, they
      were from an earlier build than the images in the
      VMS721H1_SHADOWING-V0100 kit.

      Note that this is not an image functionality issue and no new
      image functionality is provided by the VMS721H1_SHADOWING-V0300.
      If you have installed the VMS721H1_SHADOWING-V0200 kit you
      do not need to install the VMS721H1_SHADOWING-V0300 kit

          Images Affected: [SYS$LDR]SYS$SHDRIVER.EXE (new image)
                            [SYS$LDR]SYS$PLUS_SHDRIVER.EXE (new image)
                            [SYSEXE]SHADOW_SERVER.EXE (new image)

Problems Addressed In VMS721H1_SHADOWING-V0200 KIT:

   o Correction for DCL SHOW display errors.

       After the VMS721H1_SHADOWING-V0100 TIMA kit has been
       installed, a user-mode ACCVIO or formatting display errors may
       occur with various DCL SHOWs (e.g. SHOW DEVICE, SHOW MEMORY,
       SHOW TERMINAL, SHOW WORK, SHOW NET).

          Images Affected: [SYSMSG]CLIUTLMSG.EXE

New Functionality Included in the VMS721H1_SHADOWING-V0100 Kit:

  o This kit contains new command features intended for use in
     OpenVMS clusters that span multiple sites. An example of how
     to apply the new commands will be found after they are
     described.

       + DISMOUNT

          1. /FORCE_REMOVAL ddcu:

              If connectivity to a device has been lost and the
              virtual unit is in mount verification, this switch
              causes the member to be expelled immediately from the
              shadow set. If you omit this qualifier, the device
              is not dismounted until mount verification completes.
              Note, that this qualifier cannot be used in conjunction
              with the /POLICY=MINICOPY (=OPTIONAL) qualifier.

              The device specified must be a member of a shadow set
              that is mounted on this node.

       + SET DEVICE

          1. /FORCE_REMOVAL ddcu:

              If connectivity to a device has been lost and the
              virtual unit is in mount verification, this qualifier
              causes the member to be expelled from the shadow set
              immediately.

              If the shadow set is not currently in mount
              verification, no immediate action is taken. If
              connectivity to a device has been lost but the virtual
              unit is not in mount verification, this qualifier lets
              you flag the member to be expelled from the shadow
              set, as soon as it does enter mount verification.

              The device specified must be a member of a shadow set
              that is mounted on this node.

          2. /MEMBER_TIMEOUT=xxxxxx ddcu:

              Specifies the time out value to be used for a member
              of a shadow set.

              The value supplied by this qualifier overrides the
              SYSGEN parameter SHADOW_MBR_TMO for this specific
              device. The value can be different for different
              members of the same shadow set.

              The valid range for xxxxxx is 1 to 16,777,215 seconds.

              The device specified must be a member of a shadow set
              that is mounted on this node.

          3. /MVTIMEOUT=yyyyyy DSAnnnn:

              Specifies the mount verification time out value to be
              used for this virtual unit.

              The value supplied by this qualifier overrides the
              SYSGEN parameter MVTIMEOUT for this specific virtual
              unit.

              The valid range for yyyyyy is 1 to 16,777,215 seconds.

              The device specified must be a virtual unit that is
              mounted on this node.

          4. /READ_COST=zzz ddcu:

              The valid range for zzz is 1 to 4,294,967,295 units.

              The device specified must be a member of a virtual
              unit that is mounted on this node.

              This qualifier allows you to modify the default 'cost'
              assigned to each member of a virtual unit, so that
              reads are biased or prioritized toward one member
              versus another.

              The shadowing driver assigns default READ_COST values
              to shadow set members when each member is initially
              mounted. The default value depends on the device
              type, and its configuration relative to the system
              mounting it. There are default values for a DECRAM
              device; a directly connected device in same physical
              location; a directly connected device in a remote
              location; a DECram served device; and for other served
              devices.

              The value supplied by this qualifier overrides the
              default assignment. The shadowing driver adds the
              current queue depth of the shadow set member to the
              READ_COST value and does the read operation to the
              member with the lowest value.

              Different systems in the cluster can assign different
              costs to each shadow set member. Reads requested for
              virtual units from a system at Site 1 are performed
              from a member that is also at Site 1. Reads requested
              for the same virtual unit from Site 2 can read from
              the member located at Site 2.

              If the /SITE command qualifier has been specified, the
              shadowing driver will take site values into account,
              when it assigns default READ_COST values. Note, that
              to have the shadowing driver determine if a device is
              in the category of "directly connected device in a
              remote location," the /SITE command qualifier must be
              have been applied to both the virtual unit and that
              device.

          5. /SITE={nnn, logical_name} {ddcu: DSAnnnn:}

              Qualifier indicates to the shadowing driver the site
              location of the shadow set member or of the virtual
              unit. Prior to using this qualifier, you may define
              the site location in the SYLOGICALS.COM command
              procedure, to simplify its use.

              The valid range for nnn is 1 through 255.

              The following example shows the site locations defined,
              followed by the use of the /SITE qualifier:

                $ DEFINE/SYSTEM/EXEC ZKO 1
                $ DEFINE/SYSTEM/EXEC LKG 2
                $!
                $! At the ZKO site ...
                $ MOUNT/SYSTEM DSA0/SHAD=($1$DGA0:,$1$DGA1:) TEST
                $ SET DEVICE/SITE=ZKO DSA0:
                $!
                $! At the LKG site ...
                $ MOUNT/SYSTEM DSA0/SHAD=($1$DGA0,$1$DGA1) TEST
                $ SET DEVICE/SITE=LKG DSA0:
                $!
                $! At both sites, the following would be used:
                $ SET DEVICE/SITE=ZKO $1$DGA0:
                $ SET DEVICE/SITE=LKG $1$DGA1:

          6. /COPY_SOURCE {ddcu:,DSAnnnn:}

              Controls whether one or both source members of a
              virtual unit are used as the source for read data
              during full copy operations, when a third member is
              added to the virtual unit. This only affects copy
              operations that do not use DCD operations.

              HSG80 controllers have a read-ahead cache, which
              significantly improves single disk read performance.
              Copy operations, normally alternate reads from the two
              source members, which effectively nullifies the
              benefits of the read-ahead cache.

              This qualifier lets you force all reads from a single
              source member for the copy operation.

              If the virtual unit is specified, then all reads for
              full copy operations will be performed from whichever
              disk is the current 'master' member, regardless of
              physical location of the disk.

              If a member of the virtual unit is specified, then
              that member will be used as the source of all copy
              operations. This allows you to choose a local source
              member, rather than a remote master member.

          7. /ABORT_VIRTUAL_UNIT DSAnnnn:

              To use this qualifier, the virtual unit must be in
              mount verification. When you specify this qualifier,
              the virtual unit aborts mount verification immediately
              on the node from which the qualifier is issued. This
              qualifier is intended to abort mount verification when
              it is known that the unit cannot be recovered. Note,
              that after this command completes, that the virtual
              unit must still be dismounted using the following
              command:

                DISMOUNT/ABORT DSAnnnn

  o Use of the new DISMOUNT and SET command qualifiers.

     The following diagram depicts a typical multi-site cluster
     using fibre channel, and it will be used to illustrate what
     steps need to be taken to manually recover one site, when the
     site-to-site storage interconnect fails. Note that with
     current Fibre Channel support, neither site can use the MSCP
     server to regain a path to the DGA devices.

     To prevent the shadowing driver from automatically recovering
     virtual units, from connection-related failures, two steps
     must be taken prior to any failure occurring:

       1. Every device that is a member of a multi-site shadow set
           must have its time out set to a very high value, using the
           $ SET DEVICE /MEMBER_TIMEOUT= x ddcu: command. Use of
           this command will override the SHADOW_MBR_TMO value, which
           would normally be used for a shadow set member. A value
           for x of 259200 would be a seventy-two hour wait time.

       2. Every multi site virtual unit will also have to have its
           mount verification time out raised to a high value, using
           the $ SET DEVICE /MVTIMEOUT = y command. The "y" value
           should always be greater than the "x" value. Use of this
           command will override the MVTIMEOUT value, which would
           normally be used for the virtual unit. A value for y of
           259260 would be a seventy-three hour wait time.

       3. Every virtual unit and every shadow set member must have a
           site qualifier. As was noted above, doing so will ensure
           that the read cost is correctly set. The other critical
           factor in doing this involves three member shadow sets.
           When they are being used, the site qualifier will ensure
           that the master member of the shadow set will be properly
           maintained.

     In this example, DSA42 is made up of $1$DGA1000 and $1$DGA2000

       <><><><><><><><><><><> LAN <><><><><><><><><><><>
       Site A Site B

          | |
       F.C. SWITCH <><><><> XYZZY <><><><> F.C. SWITCH
               | |
       HSG80 <><> HSG80 HSG80 <><> HSG80
               | |
       $1$DGA1000 --------- DSA42 --------- $1$DGA2000

     This diagram illustrates that systems at Site A or Site B have
     direct access to all devices at both sites via Fibre Channel
     connections. XYZZY is a theoretical point between the two
     sites. If that point were to break, each site could access
     different "local" members of DSA42 without error. For the
     purpose of this example, Site A will be the sole site chosen
     to retain access to the virtual unit. The following actions
     must be taken, as shown, to recover the virtual unit at that
     site.

       On Site A:

          $ DISMOUNT /FORCE_REMOVAL= $1$DGA2000:

       Once that command has completed, the virtual unit will be
       available for use at site A, only.

       On Site B:

          $ SET DEVICE /ABORT_VIRTUAL_UNIT DSA42:

       Once the command completes, the virtual unit status will be
       MntVerifyTimeout. Next, issue the following command to free
       up the virtual unit:

          $ DISMOUNT/ABORT DSA42:

       These steps need to be taken for all affected multi-site
       virtual units.

Problems Addressed in VMS721H1_SHADOWING-V0100:

  o Executing SET DEVICE commands can result in a crashed system
     at NSA$SIZE_NSAB due to invalid argument counts in the $CMKRNL
     calls.

       Images Affected: [SYS$LDR]SET.EXE

  o SHOW LOGICAL returns a %LIB-F-INSVIRMEM error.

       Images Affected: [SYS$LDR]SHOW.EXE

INSTALLATION NOTES:

The images in this kit will not take effect until the system is
rebooted. If there are other nodes in the VMScluster, they must
also be rebooted in order to make use of the new image(s).

If it is not possible or convenient to reboot the entire cluster at
this time, a rolling re-boot may be performed.

Install this kit with the POLYCENTER Software installation
utility by logging into the SYSTEM account, and typing the
following at the DCL prompt:

    PRODUCT INSTALL VMS721H1_SHADOWING /SOURCE=[location of Kit]

The kit location may be a tape drive, CD, or a disk directory that
contains the kit.

Additional help on installing PCSI kits can be found by typing
HELP PRODUCT INSTALL at the system prompt

All trademarks are the property of their respective owners.

---

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]