CtrlX Crash when adding Data Layer Subscription in node-red

[KevinD]

Hi,

I've a problem with node-red on Ctrlx.
When I try to add a new Data Layer Subscription, the CtrlX crash and restart when I deploy.
When the CtrlX is restarted, the added object is deleted from node-red.

I already have 59 subscriptions which work but it crashes when I add the 60th.

If I remove one and then add the new it works.

Can you help me?

Thanks,

Kevin

[CodeShepherd]

Could you give some more information?

• Are different subscription configurations added or one used for accessing several nodes?
• Which version of your system apps and Node-RED app is used?
• Could you please set your logbook to persistent and provide a system report (mentioned in second half of this post) after the crash?

[KevinD]

Hello Shepherd,

thank you for your reply.

• Are different subscription configurations added or one used for accessing several nodes?
  • There is only one subscription to a CtrlX but other configuration nodes (SQL database, HTTP requests)
• Which version of your system apps and Node-RED app is used?
  • 
  • Could you please set your logbook to persistent and provide a system report (mentioned in second half of this post) after the crash?
    • I don't have this option : 

I found a temporary workaround. If instead of using a Data Layer subscription, I use an inject node with a repetition interval every 500ms and use a Data Layer request, it works.

You will find the flow in attachment if that helps

Thanks,
Kevin.

[Sgilk]

Hi Kevin,

You don't show your system apps in the image above, but if you are not on 1.12, there is likely an incompatability. I would definitely try to upgrade your ctrlX CORE firmware to v1.20 or higher and try with the matching Node-RED version. There have been many significant updates to the system apps as well as Node-RED since version 1.12.

[CodeShepherd]

@KevinD Any news here? is this still an issue? Or can this topic be closed?

[KevinD]

Hello Shepherd,

I should update the ctrlX but it is currently running in production and having never done that before I don't want to make a mistake.

For the moment I manage to get around the problem by using data layer requests with an inject true every 500ms by detecting a rising edge instead of using the data layer subscribe

The problem is therefore on standby for the moment.
I will wait for a production shutdown. How long does an update last? Are there sometimes problems or important elements to take into account before updating?

Thanks for your help,

Kevin.

[CodeShepherd]

Time needed for updating is mainly depending on your setup.

As there were major changes between version 1.12 and 1.20 Some things need to be considered:

• You will have to ctrlX WORKS 1.20 if you like to open your PLC project and do changes afterwards
• It is recommended to update your PLC project to corresponding version
• drive integrated ctrlX CORE system app version s need to fit to the ctrlX DRIVE runtime version. See e.g. topic "Drive runtime to use for ctrlX CORE 1.20 and 1.12"
• Using an "old" configuration can lead to the need to reset the scheduler settings. See e.g. topic "Motion app hangs in Configuration/Running after booting".