What are you doing with patches!?!

[LMSTAR_USER]

We have released a test system to 68 sites around the world with LabWindows/CVI version 7.1.1. One of our users discovered an issue that he was unable to run in an interactive window getting an error "...debugging information for iw.exe is out of date. Unable to debug this DLL/EXE..."

NI tech support supplied the following:
"The problem you are experiencing with the Interactive Execution Window was discovered shortly after the 7.1.1 patch was released. Therefore, our R&D team diligently worked to track down the source of the error, and it was fixed in the most recent build of 7.1.1 (build 315). The new version can be found on the following website: http://digital.ni.com/softlib.nsf/websearch/361DFA46AE788E6A86256F56004D90DB?opendocument&node 2060_US."

This new patch, however, is also labeled 7.1.1 (build 314). There is no way to differentiate these two patches. They even have the same release notes. We have also seen that we are in possession of 7.1.1 (build 313). I haven’t seen anything on the NI website stating that there is anymore then 1 version of 7.1.1. This is very unclear and causing us trouble now, and it will in the future when you release another 7.1.1 without notifying anybody or making it clear that it is a different and newer patch. This costs our users and us money to debug these problems that are already fixed, and would be easily done if we knew about them. With three digits at your disposal, would it hurt to increment the numbers???

[TiWi]

Hi,
I have downloaded the patch at the link you provided and when I install it on the evaluation version it says 314. Where is the 315 available???

[Wendy L]

Hello all,

The note about build 315 was a typographical error on my part. The build is actually 314.

Thanks.

[LMSTAR_USER]

I guess thats my main problem, how can you release 7.1.1 (build 314) twice!?! It is very costly for us to upgrade systems, and even more to debug them. I downloaded the 7.1.1 patch awhile ago, and even though I was diligently checking for new patches before releasing our software, I missed it because you didn't say anywhere you had a new version. You still don't say anywhere you have a new version. Tell me how a user is supposed to know if he downloaded the 7.1.1 patch between 1/25/2005 and 2/17/2005 that he should download it because it is a new patch? Your version numbers are the same, and even your readme files are the same.

[LuisG]

Hello,

I realize that this issue is potentially confusing, and I apologize for the trouble it's given you. I will attempt to clarify the history of these versions:

We released version 7.1.1 (the full version name is actually 7.1.1.313) earlier this year. This was a maintenance release that replaced version 7.1 (7.1.0.306). At the time we shipped 7.1.1, we also made a patch available that would modify an existing 7.1 installation and "promote" it to 7.1.1.313.

Unfortunately, what happened then, is that shortly after the patch was made available, we realized that it was missing a critical file, the absence of which would result in the error your user found. This error affected the patch only, not the regular CVI 7.1.1 installer. We immediately updated the patch, which then became 7.1.1.314. The last digit in a version number is the build number. This number needed to be higher than 7.1.1.313 so that the new patch would replace the files from the previous patch, in case the old patch had already been applied.

So the way things stand now is that if you install 7.1.1 from the CD, you get version 7.1.1.313, and if you instead apply the 7.1.1 patch from our website, you end up with version 7.1.1.314. I realize this can be confusing, although this numbering scheme is necessary in order to handle all possible permutations of the order in which these can be installed. The important point that I want to emphasize is that the 7.1.1.313 CD and 7.1.1.314 patch are functionally equivalent. The installed files are absolutely identical.

Sorry again for the confusion, and for the incovenience of having to deal with the error your user found.

Luis
NI

[BrianM]

This situation is not potentially confusing, it IS confusing!!!

I hate to argue the point, but the 7.1.1 (313) CD is NOT functionally equivalent to applying the patch 7.1.1 (314 on the CVI splash panel, but actually 7.1.1.315 if you look at the CVI_711_Patch.exe File Properties!!)

We had a project that wouldn't compile using the 7.1.1 CD (on a clean XP installation) but would only compile after applying the 7.1.1 NEW patch (7.1.1.315.0). This has cost one of our projects at least a week in effort trying to track the problem down.

I think the main point is that NI failed to mention that there were differences AT ALL between the 7.1.1 patches. If we had known up-front then we would have saved a lot of effort.

If you look at the thread at

http://forums.ni.com/ni/board/message?board.id=180&message.id=16449

you will see that it cost our engineer (and NI) a considerable time investigating a non-existent problem. (More to the point, the NI engineer never mentioned that there had been TWO 7.1.1.patches)

(This is part of a common trait within NI, you only hear about the bugs in the current software when a new release of the software is available for sale.)

Be more forthcoming about problems and inform users of any updates (however trivial) NI has gone to the trouble of now requesting users of their products to get license files from NI. As most, if not all, of this activity is via e-mail, is it too much to ask that NI use this information to notify users of ANY change to their licensed product(s)!!!!!!!!!!

[LMSTAR_USER]

Actually I beg to differ on a point that you made Luis. I have in my possession, not one, but two 7.1.1.314 patches. So there has been a 7.1.1.313, 7.1.1.314, and another 7.1.1.314. I think BrianM is right when he says it is not potentially confusing, it IS confusing. In addition, if I go to the website right now to download the patch, it says 7.1.1 period. Nothing about the build number. If the build number were listed on the site, and unique for every patch I have, I would not be having any problem. But since you are re-using version numbers, build numbers, and they aren't even included on your website ... It is problematic at best. If you need my two patch 7.1.1.314 installers I would be happy to send them to you to prove they exist.

[test_man]

Here is my 2 cents!

1. NI should inform its registred user about updates like BrianM said. Lot of people in my group do not use this usergroup and they were still using 7.0 until I told them.
2. Version updates should be 7.1.1, 7.1.2 etc.

Sheetal

[cosmo]

I will only add to this by saying that I get the NI News email, probably like most on this board, and perhaps similar to a whole lot more users who don't use this board regularly. Why can't NI add a link to the bottom of the NI News email that references all of the latest patch released since the last mailing, for every type of hardware and software and links to what they fix. Don't you think this might save NI some unnecessary phone support calls, just for users to find out that some misbehavior has already been fixed by a patch?
I would even suggest that NI consider posting 'soon to be released' patch-in-progress issues. For example, I found a bug in DataSockets earlier this year that will be fixed in the next release. How many other users would be interested in knowing about that bug, to help them decide if it is worth investigating if it affects their own code. It might save them hours of testing, retesting, and then a call to NI just to find out that yes, NI knows about it, they are fixing the problem, and here is the workaround until the patch is released.

Orlan

[LuisG]

Brian,

As I posted in the other thread, I can't explain why your engineer solved the problem when he replaced 313 with 314. Like I said earlier, 314 *is* functionally equivalent to 313. Please trust me on this. At this point, my best guess is that some other installation-related factor was accounting for his problem still happening in 313, and that it somehow disappeared when 314 was installed. It's hard to know for sure without being able to debug the problem on site.

But the main point is that there was only one 7.1.1 patch.

You mention that the patch your engineer installed was 315, not 314. Actually, that's not quite true. There was no 315 patch. I realize that the patch executable itself is versioned 7.1.1315.0. But this is *not* the CVI version number, it's only a number assigned by the installer to the patch binary (and no, it's not a coincidence that the numbers are so similar -- it's a result of having built the installer for the 314 patch twice).

In any case, I grant you your point that this is not just potentially confusing, but actually confusing. I stand corrected 🙂

Finally, I do agree with yours and Sheetal's comment about letting all users know when a patch is available, not just posting it in our website and it this forum. I will make sure that we do a better job in this respect with any future patches.

I apologize for the trouble that this problem caused you and your engineer.

Luis