RecordSave() crashes
Aldevinas Katkus
Operating system: Windows
version 8.5 2021-04-18 (build 85360453)

I cannot figure out why
NewContactCUDsmRemote() crashes in my import procedure, but it does not if I create subcontact manually.
I am using this in the Periodic Task.

I digged out to the CUVcRecordDefaults() which uses
wn = CurWindow;
if (wn!=5) then begin
CUr.CustCat = CSb.DefCustCat;

I thought, CurWindow does not work in the background thread and If I use RecordClear(ContactCUr) instead of RecordNew(ContactCUr) in the NewContactCUDsmRemote() I would get rid of this crash, but it did not help. Export-import did not help either.

updating function Boolean NewContactCUDsmRemote(record CUVc CUr,var record CUVc ContactCUr)
lCUr.Code = ContactCUr.Code;
if (ReadFirstMain(lCUr,1,true)==false) then begin
LogText(0,"NewContactCUDsmRemote 3 will store ContactCUr.Code=" & ContactCUr.Code);
if (RecordStore(ContactCUr,false)) then begin
LogText(0,"NewContactCUDsmRemote store success");//never executes

end else begin
LogText(0,"NewContactCUDsmRemote fail");//never executes

and the results in the log are:

2022-01-27 18:28:10 /THREAD(14) NewContactCUDsmRemote 3 will store ContactCUr.Code=AK5
2022-01-27 18:28:10 /THREAD(14) Damaged database: D4DiskStore dvp!=gp->MasterDVp
2022-01-27 18:28:10 /THREAD(14) D4DiskStore
2022-01-27 18:28:10 /THREAD(14) dump of call stack
2022-01-27 18:28:10 /THREAD(14) 0000000008769C90 + 2452 halpatch\CUVcWActionTool.hal: NEWCONTACTCUDSMREMOTE
2022-01-27 18:28:10 /THREAD(14) Damaged database: D4DiskStore failed
2022-01-27 18:28:10 /THREAD(14) fide = CUVc
2022-01-27 18:28:10 /THREAD(14) D4SegmentStore failed

same with 2021-12-28 version, except the last line in the log is
2022-01-27 18:48:13 /THREAD(13) holding wrong DV

Aldevinas Katkus
the same procedure works if I use tecnics>>settings>>Timed Operations>>Background tasks>>Hal routine and crashes if I use NewPeriodicTask2() to run the same procedure every 60/180 seconds (interval does not matter, crashes on the first run).
What I can see, RecordSave() is beging executed in both cases, but RecordStore(ContactCUr,false) fails with crash in Periodic task.

Any ideas why this happens?

With tecnics>>settings>>Timed Operations>>Background tasks>>Hal routine:
2022-01-28 12:12:22 MyNewContactCUDsmRemote 2 GetNextCustNr=AK5

2022-01-28 12:12:22 MyNewContactCUDsmRemote 3 will store ContactCUr.Code=AK5
2022-01-28 12:12:22 CUVcRecordSave AK5
2022-01-28 12:12:22 MyNewContactCUDsmRemote store success

With periodic tasks:
2022-01-28 11:44:37 /THREAD(13) MyNewContactCUDsmRemote 2 GetNextCustNr=AK5
2022-01-28 11:44:37 /THREAD(13) MyNewContactCUDsmRemote 3 will store ContactCUr.Code=AK5
2022-01-28 11:44:37 /THREAD(13) CUVcRecordSave AK5
2022-01-28 11:44:37 /THREAD(13) Damaged database: D4DiskStore dvp!=gp->MasterDVp

2022-01-28 11:44:37 /THREAD(13) D4DiskStore

2022-01-28 11:44:37 /THREAD(13) dump of call stack

David Delač
Hi Aldevinas,

Please create a bug report containing text backup, halcust and steps to reproduce the issue and let me know when It's done. I have to debug it myself to get better idea of what could be wrong.

Best Regards,
David D.
Aldevinas Katkus
I could not repeat it in a separate hal.

I am not sure what I did so it does not crash anymore. It could be I made a queued function call like this:

external updating procedure IVImportIn();

procedure DoIVImportTask(string args)

queued.IVImportIn; //call a function which has some qupdating function calls


procedure InitIVImportTask()
//used to crash in NewContactCUDsmRemote() when used as Periodic Task: Damaged database: D4DiskStore dvp!=gp->MasterDVp D4DiskStore

Leave Comment
You can subscribe to notifications for this post by selecting the 'star' icon on the top right corner of the post.
Back to the list
Latest Posts
Andis Blicāns, Grasko
The above applies to the window when it is opened for the first time. If the window is already open and I choose Window/Restore Position, the window takes the correct position. Namely, the upper left ...
00:43 18 July 2024
Version: 2024-03-16 Server: Linux Client: MacOS Sonoma I am seeing multiple users in different locations getting logged off many times per day when idle sometimes after a few minutes of being awa...
20:09 17 July 2024