Dear HP, why was patch VMS83A_LOGIN V2.0 (aka DEC-AXPVMS-VMS83A_LOGIN-V0200) released for the second

[Jan van den Boogaard_1]

Dear HP, why was patch VMS83A_LOGIN V2.0 (aka DEC-AXPVMS-VMS83A_LOGIN-V0200) released for the second time ??? first time was June 10, 2009. Second time was July 22, 2009.

Thanks.

[Ian Miller.]

You enjoyed so much that you wanted to do it again, again :-)

Are you referring to an announcement of the patch and if so was it on the famous openvms.org alerts list or the official HP notification?

[Volker Halle · Accepted solution]

Jan,

according to my OpenVMS image database, the 2 images in this kit (LOGINOUT.EXE and SETP0.EXE) have the same link date/time (if you believe the patch release notes) as in the previsouly (11-JUN-2009) released VMS83A_LOGIN-V0200 kit.

The checksums of the kits are DIFFERENT - as reported in the release notes.

The new kit just showed up on the HP FTP site on 23-JUL-2009 12:19.

This is a very bad practise. When re-releasing a kit the version of the patch kit MUST be incremented - otherwise you'll never know which kit you have really installed.

Volker.

[Graham Burley]

Not excusing anybody/thing but I'm pretty sure I've seen this before when the release notes have been changed.

According to the .txt files published on the ftp server the change added the following text:

"This kit do not support external authentication."

[Graham Burley]

I did a quick check of VMS83A patches in our ECO conference on EISNER: https://mail.encompasserve.org/anon/htnotes/title?f1=ECO&f3=vms83a

At some time the following patches have all been re-released with revised checksums:
VMS83A_ACMELDAP-V0300
VMS83A_ACRTL-V0100
VMS83A_DRIVER-V0200
VMS83A_LOGIN-V0200

My interpretation is that this is due to an amendment to the release notes and re-kitting, at least that's what I hope is going on.

[Kris Clippeleyr]

Hi,

It is me who might be the culprit.
I officially filed a case that the various LOGIN patchkits caused a regression of functionality when the ACMELOGIN kits where previously installed.
So engineering decided to NOT merge the ACMELOGIN functionality into the mainstream LOGINOUT.EXE image, but instead to change the installation procedure of the LOGIN cq. UPDATE patchkits so that is checked (and signalled to the installer) if an SYS$ACM enabled LOGINOUT.EXE image is present.
Hence the repackaging.

Kris (aka Qkcl)

[Vivek Sharanappa]

Hi,

The new kit has a change to the release notes and hence re-kitting was done.

There is no change to the objects.

Thanks,
Vivek

[Jan van den Boogaard_1]

Hi,

I just checked it myself , first installed the 1st patch, then the 2nd patch. Both release notes say

7 FILES PATCHED OR REPLACED:


o [SYSEXE]LOGINOUT.EXE (new image)

Image Identification Information

image name: "LOGINOUT"
image file identification: "X-54"
image file build identification: "XBCA-0080070007"
link date/time: 3-FEB-2009 16:22:16.40
linker identification: "A13-03"
Overall Image Checksum: 1823929423

o [SYSEXE]SETP0.EXE (new image)

Image Identification Information

image name: "SETP0"
image file identification: "X-54"
image file build identification: "XBCA-0080070020
link date/time: 14-JAN-2009 14:42:26.04
linker identification: "A13-03"
Overall Image Checksum: 3323757728

the only differences is in the release notes of 22 july:

in the text

5.2.1 ACMELDAP LOGINOUT.EXE Conflicts

5.2.1.1 Problem Description:

The VMS83A_LOGIN-V0200 patch kit is being released in
conjunction with the VMS83A_ACMELDAP-V0300 patch kit.
Previously, VMS83A_ACMELDAP kits contained a copy of the
VMS83A_LOGIN and VMS83A_ACMELOGIN patch kits. These kits
include versions of LOGINOUT.EXE that are functionally
different. This functional difference has led to
conflicts that have resulted in customers experiencing
regression of LDAP ACME login capability. This
regression occurs after installing a VMS83A_LOGIN or
VMS83A_UPDATE kit that includes a non-ACMELDAP
LOGINOUT.EXE.

The VMS83A_LOGIN patch kit has been removed from the
VMS83A_ACMELDAP kit and will now only ship as a separate
VMS83A_LOGIN patch kit. It will now check for the
presence of the ACMELOGIN kit and, if found, will warn
the user of the regressions risk and ask if they wish to
continue the installation. This kit do not support
external authentication.


the last line "This kit do not support external authentication." is new.

Problem solved, thanks all for yor replies.

[Jan van den Boogaard_1]

Closed.