Question

CuurentActivityContext has been garbage collected

V
Rank
Badge +3

Our users get this error message when using lobbies that are connected to forms/tbws . This appears every day, sometimes several times a day for some users. We tried to look in to this and the answer we got from the end users is that it could have something to do when many work on the same lobby simultaneously OR that the lobby wasn´t updated which means that the information somewhere in the lobby is old. We also got from our users that it appeared at the same time during the day (for example around 10 a clock).

 

 

 

10 replies

N
Rank
Badge +3

Hi @Vinoy Heshan,

I wanted to check in and see if you've made any progress in identifying the cause of the issue or finding a solution.

Thanks & Best Regards,

nikllk

V
Rank
Badge +3

​​​@nikllk 

Thanks for checking up, still we do not have a straight forward answer for this matter.

Regards,

Vinoy

D
Rank
Badge +2

We have the same error on our Shipments page where they have an embedded lobby. We’ve had the problem since we went live in May 2022. Our IFS partners had no idea of how to fix it, and we’ve had an open ticket with IFS since Jan 2023 - over a year now. 

IFS have asked us to replicate it in a non-prod environment because they can’t create it in their environment, but the problem is we don’t know what conditions make it happen. I know they have multiple windows open at once to gather information to build consolidated shipments, they return to the page and get the error. It’s not all of the time though so it’s impossible to recreate it on demand.

Given the fact that the message relates to CurrentActivityContext - if they are swapping in and out of screens would it be better to use ApplicationContext or would this cause other memory leak issues?

H
Rank
Badge +3

Hi, 

we have the similar problem! several times during the day → lobbies are just gone, not possible to use.

after reboot of IFS, they are working again normally. 

 

How we can fix this problem? What to check and fix?

Those elements are very important for users.

 

@Vinoy Heshan, @nikllk, @DAMS Sue 

 

 

N
Rank
Badge +3

Hi @Hanna V,

Do you get the same CurrentActivityContext has been garbage collected error? In your situation what actions lead this error to occur? Have you reported this to the IFS support?

Thank you & Best Regards,

nikllk

H
Rank
Badge +3

@nikllk

yes, we get that error after 1-2 hours usage of lobby elements quite randomly.. then those lobby elements just disappears. they are not even behind the lock when going to check which lobby elements I want to be visible in selected view.

This problem has been for several months.

we are using IFS 10 update 15. 

 

 

We have not managed to find the root reason, currently trying to get some notes from users when this happens, does it happen to colleague at the same time and so on.

We have not reported this to IFS, yet. 

We try to find help from here Community first. If we do not find any explanation → we will report this error to IFS

 

-Hanna

N
Rank
Badge +3

Hi @Vinoy Heshan @Hanna V @DAMS Sue,

Following a meeting with our internal product development team, it has been identified that the CurrentActivityContext has been mapped as a weak reference. Transitioning this to a strong reference could potentially resolve the issue; however, given the complexities of the application's architecture, this change might introduce unforeseen challenges.

We acknowledge that this situation is less than ideal, particularly considering the occurrence of garbage collection errors experienced by several customers. Consequently, our next step is to conduct further investigation to identify potential solutions without compromising the integrity of the application's architecture.

To facilitate this investigation, we kindly request more specific information regarding the occurrence of the issue. Could you please provide details on the application windows users had open when the problem arose? Additionally, information on the sequence of windows the customer navigated between during the occurrence of the issue would be invaluable.

We understand that this request may disrupt your workflow, and we sincerely apologize for any inconvenience caused. However, obtaining detailed information is crucial for us to identify patterns and devise an effective solution.

Thank you for your cooperation and understanding.

Best Regards, Nimesh

H
Rank
Badge +3

Hi @nikllk 

We have lobby elements in. same element is in many views, because elements contains important info about qty in stock and customer order data (importan info for production planner).

  • shop order requisition release
  • Batch balance
  • Shop order/materials
  • Inventory part/planning data
  • Planning details for inventory parts
  • Inventory parts
  • Shop Orders

First two are heavily used views, the most important ones.. 

We have not notified any certain pattern, when the error occurres. Quite often it is around 8.00 EET. 

→ can some backround job affect ? 

 

-Hanna

H
Rank
Badge +3

HI @nikllk 

This is weird problem, 

Example today with one user, lobby elements went down about 8.15, then he rebooted IFS and elements worked again. But 9.35 elements went down again… elements worked little over one hour… 

and in some day, elements might work whole 8 hours working day.

 

What can affect? What we should check? 

 

-Hanna

D
Rank
Badge +2

Hi @nikllk 

Would changing it to a Strong Reference risk Memory Leaks?

Would a Soft Reference be any better?

https://stackoverflow.com/questions/9809074/java-difference-between-strong-soft-weak-phantom-reference

Thanks

Sue

Reply


Terms & ConditionsCookie settings