Developer Portal | UUID part 2 changing

Submitted by R C on June 17, 2011 - 6:54pm

Tags:

Hello,

UUID on 2 MC75 devices changed. Only 1 digit changed in part 2 was affected. Here is the end user's log showing the change:

2011-05-26 16:01:35.0000 Device:   MOTOROLA MC75
2011-05-26 16:01:35.0000 UUID:     23FF3FA527972D9C
...
2011-05-27 08:53:11.0000 Device:   MOTOROLA MC75
2011-05-27 08:53:11.0000 UUID:     23FF3FA 427972D9C

My understanding is that the UUID is generated by the OS and can change do to either major OS updates or hardware changes.

What other factors can cause part 2 of the UUID to change?
What formula is used to calculate part 2?
What can we do to troubleshoot further?

Thanks in advance for your help.

May 4, 2019 - 10:07am

Juan-Antonio Ma...

By the way, in the staging

By the way, in the staging phase of about 2,500 MC75 for Chronexpres (Spanish Post Parcel division) partner has found that about 5% of the units have duplicate UUID (in groups of two or even three).
They use MSP to stage them, so eventually final report shows less staged devices than actually done. As we all know, MSP used UUID as identifier.

Vote:

May 4, 2019 - 10:07am

Vikram Mangeshwar

For Duplicate UUID issue, see

For Duplicate UUID issue, see SPR 20532.

UUID Part 1 is from the OS whereas UUID part 2 is from the OEM. In MC75, a unique id (8 byte) embedded in mDOC is used as UUID Part 2.

Which is changing? Part 1 or Part 2?

Vote:

May 4, 2019 - 10:07am

Nick Zeleznock

It is one digit that changed

It is one digit that changed in part 2. We saw 2 units that 1 digit in part 2 had changed.

Vote:

May 4, 2019 - 10:07am

Vikram Mangeshwar

Does the change persist? Or

Does the change persist? Or does it revert back to the original value after the next cold boot?

Vote:

May 4, 2019 - 10:07am

Nick Zeleznock

The cold boot changed the

The cold boot changed the digit back to the original. Does this make sense? Any idea what would cause this?

Vote:

May 4, 2019 - 10:07am

Gene Niles

This sounds like an SPR in

This sounds like an SPR in the making. The UUID comes from various parts in the system but I believe OS reads the collective value from RAM. What you are describing points to something writing to area that stores this value.

If you can get a reproduction, I would suggest an SPR.

Vote:

May 4, 2019 - 10:07am

Nick Zeleznock

One more thing I should have

One more thing I should have noted that the cold boot only worked after sitting all weekend. I believe the battery discharged.

Vote:

May 4, 2019 - 10:07am

Juan-Antonio Ma...

After reading SPR 20532

After reading SPR 20532 description, I think there is some sort of patch which simply creates the UUID out of ESN, left-padding with zeroes.
This patch should also be a solution for this current issue, shouldn't it?

Where can we download it? Thanks!

Vote:

May 4, 2019 - 10:07am

Nick Zeleznock

Do we have a way to check the

Do we have a way to check the mDoc chip when we see a change? I would like to determine if it is a mDoc issue or a MS issue in the way it reports it in the OS.

Vote:

May 4, 2019 - 10:07am

#10

Gene Niles

This may not be the same

This may not be the same issue. If you apply the patch, it only changes the area of the UUID that is for the mdoc part. So you may want to put the patch on and see if it affects the area where you are seeing the issue. If not, then you are looking at a different issue.

Vote:

Search form

UUID part 2 changing