Bug #5062
closed
exception after successful XSL transform/
0%
Description
Hi, In a single-threaded application we are getting the following exception after successful XSL/XML transform: "JET RUNTIME HAS DETECTED UNRECOVERABLE ERROR: system exception at 0x00007fd03eb0ece9 Please, contact the vendor of the application. Core dump will be piped to "/usr/share/apport/apport %p %s %c %d %P %E" Extra information about error is saved in the "jet_err_20387.txt" file."
How can we prevent it? especially we ant to prevent the creation of this core dump since it is slowing our application.
Files
Updated by Michael Kay over 3 years ago
- Assignee set to O'Neil Delpratt
Thanks for reporting it. I'm assigning it to O'Neil despite the fact that he's away this week, as he knows far more about this area than anyone else and I think that will lead to a quicker response.
I suspect though that he'll need more information about exactly what you are doing: what API methods you are calling, and what platform you are running on. The ideal is always to have a repro - that is, a script that we can run and that demonstrates the error, preferably simplified to remove all extraneous complications.
Meanwhile, if you search this site for the phrase JET RUNTIME HAS DETECTED UNRECOVERABLE ERROR you may find other issues with the same root cause, some of which have patches or workarounds (most of them seem to be Python - you haven't said what API you are using).
Updated by O'Neil Delpratt over 3 years ago
Hi, are you able to provide a small repo? If you are using linux then the gdb debugger will give better insight to the cause of the crash.
Updated by ofer benoliel over 3 years ago
as can be seen in the attached error file, the platform is Ubuntu 16.04 and is the C API in interface. (Saxon/C) the title also shows: [Saxon/C - Bug #5062] exception after successful XSL transform We have previously sent you the interface wrapper that we use. We use it in the most simple way. What is required to be done to stop the creation of the dump file from JAVA code?
Updated by O'Neil Delpratt over 3 years ago
Hi,
I took a look at the wrapper here: https://saxonica.plan.io/attachments/52921
There seems to be a seg error somewhere in the program. I will try to reproduce it from the wrapper program.
Do you call proc->release() at the end of the program? I could not see it as it is required.
Updated by O'Neil Delpratt over 3 years ago
- Status changed from New to AwaitingInfo
Please can you confirm the call of proc->release() in the program at the end?
Let me know if the problem goes away.
Updated by ofer benoliel over 3 years ago
Adding a call to the Release() function didn't make any change.
Updated by O'Neil Delpratt about 1 year ago
- Status changed from AwaitingInfo to Closed
This issue is over years old. SaxonC 12 is now built using Graalvm a redesign of the internals. This includes the error handling. Please try the latest product and reopen the bug issue if the problem still exists.
Please register to edit this issue
.