Annual bug thread for 2021

[altenbach]

Continued from the Year 2020 bug thread.

Happy New Year everyone! Hopefully we can focus again on software bugs in 2021 after dealing mostly with real ones in 2020.

The rules remain the same:

• Each post here should only contain a brief description of a bug, together with a link to the thread where it is being discussed.
• One post/bug!
• One bug/post!
• No discussion of bugs. All discussions of a bug must take place in the original linked thread.

[Matthew_Kelton]

Scrolling while dragging in the project window no longer works (LV 2020 thru SP1)

[FixedWire]

A VI within one library can somehow be linked to a second library and suck that second library into the dependencies. This locks up LabVIEW as it tries to figure out what to load. This one is really bad. PM me for details.

[thols]

VI fails to compile when disabling a channel writer endpoint 

[PNR]

Potential memory corruption when (de-)serializing Sets in LabVIEW 2019 SP1 f3 (32-bit) - LabVIEW Bug...

[PNR]

Choosing "Select XNET Database" from a XNET Frame control causes application crash when no XNET Database Aliases are defined.

CAR 1385768

[FixedWire]

Found another terrible bug today.

The string output from 2 different VIs didn't yield an output at all (i.e. was blank) when called as a subVI.

These were rather simple VIs thankfully. Unit tested the code in question with no errors.

Approach:
1. tried to mass compile; didn't help

2. closed LV, opened LV and cleared the object cache; didn't help

3. rewired the connector pane; didn't help

4. deleted the string control, created a new control & connected the connector pane. AND IT WORKED.

This code is used in a validated environment. These types of bugs are really hard to find and can cause unexpected behavior. That's how it was found.This is an issue for risk mitigation.

Hopefully someone at NI has seen this before & can put a priority on it.

[wiebe@CARYA]

---

@FixedWire wrote:

Found another terrible bug today.

The string output from 2 different VIs didn't yield an output at all (i.e. was blank) when called as a subVI.

These were rather simple VIs thankfully. Unit tested the code in question with no errors.

Approach:
1. tried to mass compile; didn't help

2. closed LV, opened LV and cleared the object cache; didn't help

3. rewired the connector pane; didn't help

4. deleted the string control, created a new control & connected the connector pane. AND IT WORKED.

This code is used in a validated environment. These types of bugs are really hard to find and can cause unexpected behavior. That's how it was found.This is an issue for risk mitigation.

Hopefully someone at NI has seen this before & can put a priority on it.

---

Without steps to reproduce, this will be near impossible to fix. It might even have been a user error. I know, you're absolutely sure it isn't. Without the VIs, we'll never know.

I think this thread is for bugs acknowledged by NI though...

If you can reproduce this, please create a new thread on the forum (maybe post a link to it here). When confirmed, post the BUG number here.

The rules remain the same:

• Each post here should only contain a brief description of a bug, together with a link to the thread where it is being discussed.
• One post/bug!
• One bug/post!
• No discussion of bugs. All discussions of a bug must take place in the original linked thread.

It seems I'm breaking the rules myself...

[thols]

Delete element using CTRL+Delete only visually updates data. Actual data is not changed:

https://forums.ni.com/t5/LabVIEW/Bug-Delete-element-using-CTRL-Delete-only-visually-updates-data/m-p...

[cy...]

is it fixed?