OpenVMS VAXDRIV05_071 VAX V7.1 DUDRIVER/TUDRIVER ECO Summary

TITLE: OpenVMS VAXDRIV05_071 VAX V7.1 DUDRIVER/TUDRIVER ECO Summary
 
Modification Date:  05-MAY-1999
Modification Type:  Updated Kit:  Supersedes VAXDRIV04_071 

NOTE:  An OpenVMS saveset or PCSI installation file is stored
       on the Internet in a self-expanding compressed file.
       The name of the compressed file will be kit_name-dcx_vaxexe
       for OpenVMS VAX or kit_name-dcx_axpexe for OpenVMS Alpha.
 
       Once the file is copied to your system, it can be expanded
       by typing RUN compressed_file.  The resultant file will
       be the OpenVMS saveset or PCSI installation file which
       can be used to install the ECO.
 
Copyright (c) Compaq Computer Corporation 1999.  All rights reserved.

OP/SYS:     OpenVMS VAX 

COMPONENT:  DUDRIVER
            TUDRIVER

SOURCE:     Compaq Computer Corporation 

ECO INFORMATION:

     ECO Kit Name:  VAXDRIV05_071
     ECO Kits Superseded by This ECO Kit:  VAXDRIV04_071
                                           VAXDRIV01_071
     ECO Kit Approximate Size:  180 
     Kit Applies To:  OpenVMS VAX V7.1 
     System/Cluster Reboot Necessary:  Yes
     Rolling Re-boot Supported:  Yes
     Installation Rating:  INSTALL_3
                           3 - To be installed on all systems running
                               the listed versions of OpenVMS which
                               are experiencing the problems described.

     Kit Dependencies:

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

         None 

       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 TUDRIVER and DUDRIVER on OpenVMS VAX V7.1.  
This kit addresses the following problems: 

The two primary deliverables in this kit are:  

  1.  A number of backported fixes from the V7.2 stream which keeps 
      V6.2 and V7.1 code consistent with the V7.2 release.  This 
      proactively delivers fixes that were found during V7.2 regression 
      testing (some of these problems were also reported by customers).

  2.  The ability to mount and use Fibrechannel devices from V6.2 and 
      V7.1 systems.

Problems addressed in VAXDRIV05_071 kit:

  o  When initializing a FIBRECHANNEL (DG) device from a client,
     one of the two following errors is returned to the user (disk
     only, not tape):

       1.  INIT-F_DIAGPACK, disk is a diagnostic pack; or

       2.  INIT-W-NOBADDATA, bad block data not found on volume

     Note that an INIT of the same devices succeed on the server.

     Image(s) affected:  [SYS$LDR]SYS$DUDRIVER.EXE

  o  The following describes conditions that may cause crashes:

       1.  During a virtual circuit failover event, DUDRIVER 
           attempts to use the CDT of another connection, for 
           example, the SCS$DIRECTORY, and this results in a 
           crash.

       2.  DUDRIVER will now detect that a CDRP has a zero 
           CDRP$L_UCB field, thereby preventing a system crash.

       3.  A failover when a device has two paths (one served 
           and one local) may result in a fatal MSCPCLASS bugcheck 
           error.

       4.  When an MSCP server loses connection to the local MSCP
           device, a SHADDETINCON crash may ocur and the shadowset
           member will be remvoed from the shadow set.

     Image(s) affected:
       -  [SYS$LDR]SYS$DUDRIVER.EXE
       -  [SYS$LDR]SYS$TUDRIVER.EXE

  o  Potential performance problems may occur (high interrupt stack
     time) on low end systems in large cluster configurations, when
     drives go offline.

     Image(s) affected:
       -  [SYS$LDR]SYS$DUDRIVER.EXE
       -  [SYS$LDR]SYS$TUDRIVER.EXE

  o  DUDRIVER/TUDRIVER device allocation class processing problems
     can occur during failover and configuration.

       1.  Drives may be incorrectly configured multiple times as
           separate devices with the same name on different paths.

       2.  Invalid data structures can be seen for DUDRIVER-based
           devices with unusual (32-bit number) ALLOCLASSes, which
           are not valid for any devices on the system.

       3.  Systems may crash upon receipt of "Duplicate Unit" MSCP
           messages from a HSx controller.  The foot print will be 
           an inconsistent I/O Database or ACCVIO.

     Image(s) affected:
       -  [SYS$LDR]SYS$DUDRIVER.EXE
       -  [SYS$LDR]SYS$TUDRIVER.EXE

  o  The MOUNT and INIT commands can fail with an MEDOFL error.

     Image(s) affected:  [SYS$LDR]SYS$TUDRIVER.EXE

  o  Incorrect byte count in IRP/IOST may occur when a forced error
     on a disk is encountered.  

     Image(s) affected:  [SYS$LDR]SYS$DUDRIVER.EXE

  o  A DCL copy command to a tape drive connected to a HSD05/10
     can cause a process hang.  This problem will not happen with a
     BACKUP command and will not happen on HSJ/HSC connected tape
     drives.

     Image(s) affected:  [SYS$LDR]SYS$TUDRIVER.EXE


Problems addressed in the VAXDRIV04_071 kit:

  o  Attempting to perform a directory of a mounted tape causes the
     process to hang and "INVALID COMMAND" MSCP END MSG* errorlog
     entries to be generated.

  o  Unexpected end messages are logged on client nodes for GUS and
     AVAILABLE commands.

  o  When SCSI disks connected to a HSxx controller are set transportable, 
     the system crashes with a DISKCLASS bugcheck.

  o  DUDRIVER does not FLUSH the pending queue of SHADOWING, HBR and served  
     IOs when connection failure routines are entered.  This causes a 
     system/cluster hang.


Problems addressed in the VAXDRIV01_071 kit:

  o  A problem exists with HSJ/HSD30, 40 and 50 controllers.  After
     some event that initiates Mount Verification, a pack-ack will
     fail to complete.  The controller will report that it is
     making progress on the command, but will never finish.  This
     causes all IO to the affected devices to hang.  The problem
     has been seen mostly with the ONLINE component of a pack-ack.
     One case of a NOP hanging has been seen.  It is not known
     whether other IOs including Data Transfer IOs may be involved.

  o  Attempting to perform a directory of a mounted tape causes the
     process to hang and "INVALID COMMAND" MSCP END MSG* errorlog
     entries.

     This problem can be seen in the following scenario.  A 9-track
     tape written with a density of 6250 is mounted on a TSZ07 tape 
     drive.  The mount completes successfully and is followed by a 
     DCL "DIRECTORY" command.   When a pack-ack is issued for the 
     tape unit, the subsequent ONLINE command is issued using 
     "stale" density information.  This ONLINE command is then 
     rejected by the controller with "INVALID COMMAND, INVALID 
     FORMAT" errors and the process will hang.  The error count then 
     continues to increment as the driver keeps issuing the ONLINE 
     command.

     In the TUDRIVER START_PACKACK routine, a GUS-ONLINE-GUS sequence
     is issued to the tape unit.  The initial GUS is skipped unless
     the drive is a DLT type or if the UCB$V_TU_DENS_DETERMINED bit
     is set in DEVSTS in the UCB.  If a GUS is issued before the
     ONLINE command, the density gets updated and the ONLINE
     succeeds.


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.

Colorado Site
Georgia Site

vaxdriv05_071.README
vaxdriv05_071.CHKSUM
vaxdriv05_071.CVRLET_TXT
vaxdriv05_071.a-dcx_vaxexe
vaxdriv05_071.CVRLET_TXT