Reporting a LABVIEW bug

[4RangN1]

I am using LABVIEW 2018 without a service program. It seems there is no way for me to talk to NI.  NI should be very kind and supportive. Nowaday NI is very very NOT until I buy the service program. Very costly for a private programmer. How to report a bug and get a fix from NI?

[RavensFan]

Post it in the forum as a message.  It will give others a chance to test to see if you really have a bug or if it is programmer error.

Generally, NI monitors the forums as well and will check on things considered a bug and will file a bug report if they concur.

[Bob_Schor]

Even if you did have a Service Program, when you make a Service Request, there are three ways to report it:  by Phone, by e-mail, or through the Forums.  Even without a Service Request, you can (and should) report it here, where your fellow LabVIEW Enthusiasts can look at your situation and help you get it corrected.

Please provide an explanation of the nature of the "bug", code illustrating the bug (be sure it is executable, i.e. that we can download it, run it, and see the problem), and what you think the correct behavior should be.  I know I'd be interested, and I sure my fellow Knights (and other LabVIEW Enthusiasts who frequent the Forums) would be happy to respond, either "explaining" the behavior (I'd say well over 75% of the "bugs" that I think I found turn out to be misunderstandings on my part, sometimes due to careless, or even absent, reading of the Detailed Help) or acknowledging a possible Bug, which we can then bring to NI's attention.

Bob Schor

[altenbach]

Fixing a real bug will take time and depends on the severity (crash, inconvenience, incorrect behavior, cosmetic, etc.). Only very serious bugs will be applied to older version via a patch release.

Start a new thread and post a detailed description of the bug and some example code.

Once the bug is confirmed, we can add an entry in the annual bug thread.

Once we collectively see the code, we will:

1. First determine if it really is a bug. Sometimes it is expected behavior and correct. Sometimes it is due to ignorance (race conditions, comparing DBLs, etc.).
2. Determine if the bug still exists in LabVIEW 2019. (maybe it is already fixed!)
3. Determine if it also existed in older versions (If it existed for decades, it's probably not serious).
4. We will figure out if the bug is already known. (NI has a database of all known bugs)
5. We will figure out a workaround so you can keep working.

The best you can hope for in the short to medium term is a workaround to avoid the bug, and we can help you with that.

So:

• Attach some simplified code that still reproduces the bug.
• Tell us exactly what the values of all controls are and how you interact with the running code and exactly in what order (e.g. press button A, change value B to 5, press button C, indicator D is now 6 instead of the expected 7. etc.).
• Tell us what result/behavior you see and what result/behavior you expect to see instead.

[4RangN1]

My Labview program is very big and complicated. It crashed while I am working on development, Fortunately it is restored when restarted.It happens a couple of times a week.  I am glad that it does not crash while running.

When crashed, it automatically ask to report to NI. I did it many times, but no response from NI. It just forced to quit; no way to investigate. I don't understand the nature of the crash.

[GerdW]

Hi Rang,

---

@4RangN1 wrote:

It crashed while I am working on development, Fortunately it is restored when restarted.

---

I hope you are using any kind of SCC tool to keep a history of your code instead of relying on code restoring capabilities of NI/LabVIEW crash detector?

[thols]

LabVIEW crashing does not in itself indicate that there is a bug in LabVIEW. It is possible to write code that can crash any environment.

Try to break the code into smaller pieces and projects and work on them without loading everything at once and see if that helps. Then you can start tracking down the cause.

Try to remember when the crash started to happen and checkout a version from your SCC before that and see if that crashes too. 

Update LabVIEW to the latest SP.

Try on another PC and see if the issue is the same.

Is there anything saved in %UserProfile%\Documents\LabVIEW Data\LVInternalReports\LabVIEW\<lv-version> or in %UserProfile%\DocumentsLabVIEW Data\lvfailurelog?

See here how to capture a crash if LabVIEW doesn't record it.

I'm sure there are lots of threads how to investigate a crash. Search for them too.

[wiebe@CARYA]

---

@4RangN1 wrote:

When crashed, it automatically ask to report to NI. I did it many times, but no response from NI. It just forced to quit; no way to investigate. I don't understand the nature of the crash.

---

The intent of the error reporter is for NI to collect crash report and perform statistics on them. It's not a way to get in NI to contact you. They will only do that if your crash is interesting enough for them to need you, and of course only if you supply your email.

You might want to remove your email from the crash report if you post it here, BTW. I wouldn't be surprised if those webbots look in zip files to collect new spamees.

I never heard from anyone that NI contacted them as a response to a crash report. That doesn't rule out that it ever happened of course, but I wouldn't wait for it to happen.