segmentation fault when reusing instances of tessreract

[Martinez78]

I tried to use Tesseract via threads in parallel. I created three instances of Tesseract with the same settings, and create Callable objects referencing always one of those three for each file in a batch of PDFs and submitted the “callable” objects to a FixedThreadPool. This way these “callable” Objects do reuse always one of the same three Tesseract instances previously created when processing one file. In each callable object doOCR is called with the corresponding input file. Processing a lot of PDF documents with theses same Tesseract instances works well, but with some PDFs after a while the JVM crashes with a SIGSEV.

A fatal error has been detected by the Java Runtime Environment:

SIGSEGV (0xb) at pc=0x00007ff6e4302f3e, pid=20667, tid=0x00007ff72939a700

JRE version: Java(TM) SE Runtime Environment (8.0_102-b14) (build 1.8.0_102-b14)
Java VM: Java HotSpot(TM) 64-Bit Server VM (25.102-b14 mixed mode linux-amd64 compressed oops)
Problematic frame:
C [liblept.so.5+0xfef3e] pixDestroy+0xe

I also saw the message " Please call Init before attempting to set an image.corrupted size vs. prev_size while consolidating", in my opinion pointing to a memory alignment / structure mapping problem
hs_err_pid20667.log
.

The error does not occur when each callable uses its own freshly created instance of Tesseract for each and every input file.