Application hang event 1002

Hi All,

Does anyone know how to fix application hang event 1002?

One of our customer is on Windows Server 2019 and sometime the application will be hanged several times a day.

Thanks

Louie

Hi Louie.

The only similar experience that I had, was when the connection to MSSQL in our backend app was lost, and Clarion doesn’t seem to graciously reconnect. My solution was to change the connection to ADO where I have more control. It was a bit drastic to change it, but it proved to be effective.

Thys

I think you will need to add logging to your application to see what is happening when the error occurs.

Can you tell me how ? I have the same issue. all .tps and all parts of the application are on a fileserver. After migrating to a new fileserver the program ā€˜hangs’ after a while when the user does nothing. How can I check this hanging?

Piet, I believe you will need to disable opportunistic locking and what ever level of SMB is on your server (SMB2 or SMB3, likely). I have not worked with a TPS database over a network in 20 years, so I’m not much help.

1 Like

Dit not expect this: Normally the user runs the application with a shortcut on the desktop (the application is on a fileserver). After a while ( I think when the screen goes blanc) the program hangs. Now I let the user doubleclick the .exe on the server: after a while…no problem anymore.

We’re experiencing the issue as well. Clicking the exe did not solve the issue for us. We’re able to reproduce the issue consistently, but the timing of the hang is not consistent. The hang happens when we leave a list and a form open and do other work outside our Clarion app. The Clarion app will eventually hang, CPU for the process will jump to 12%, memory usage is low. The app does not return from the hanging state no matter how long we leave it. We have to end task using task manager.

Our app is written in Clarion 11.0.13401. Backend is SQL 2016.

Hoping someone has more detail on how to address this issue.

Check in SQL Server Management Studio if your app is still connected to the server. It’s possible the connection has been dropped? You might be able to run a ā€œkeep aliveā€ query on the app frame to ensure the connection never gets dropped?

1 Like

Can you check to see if it’s deadlocked or waiting on a transaction?

I checked with both SQL profiler and using a statement to return open cursors and they are still reporting as alive on the SQL side. When the hang occurs, nothing is being caught on the SQL side. Almost implying that a call to SQL was not made during that time.

Thanks, Seanh. This was my initial feeling as well. Using SQL profiler and a few statements, I was able to see any deadlocks. When the hang occurs, there doesn’t appear to be a call that makes it to SQL. Almost as if it’s more related to the application code.

Not sure if you are aware of this Store procedure:

It allows you to run a command like this, for example.

exec sp_WhoIsActive
  @get_locks = 1, 
  @get_plans = 1,  
  @get_additional_info = 1, 
  @find_block_leaders = 1

It has helped us in the past to find locks from other users that influenced our connection.

2 Likes

Thanks for pointing this handy SP out. I’ll be sure to take a look and see if it presents any additional clues.

After extensive troubleshooting, I’ve identified the issue as related to modern Windows and MDI (Multiple Document Interface) in our Clarion app.

Current Status:
We’re running version 11.0.13401
Version 11.0.13622 includes fixes for this issue: Clarion 11.13622 Sept. 4, 2020

Related Issue:
This thread more accurately describes what we’re experiencing: Moving beyond Clarion 10 11897 (Cross-Thread Deadlock Silent GPF's)

Options Under Consideration:

  • Update to version 11.0.13622
  • Update to version 11.1.xxx
  • Disable Run-Time Translation as a mitigation measure
  • Migrate the app and eliminate parent-child windows in MDI

We’re evaluating these options and will provide an update. If anyone has additional suggestions, please share them. Thanks in advance.

I’d suggest the last of the 11.1 series 13855 I think. It had a number of good fixes in it

1 Like

Resolution found:

Our issue was caused by having a TIMER call on a child window that updates the date and time on the status bar of the MDI App. This was not intended. Removing that call resolved the hanging.

Thank all!

2 Likes