Process 1698729 (zbarimg) crashed in _IO_peekc_locked.cold()

[gosssamer]

I'm using zbarimg with spamassassin and amavisd on fedora40 and occasionally see this message in my system logs. I also see error 15 issues when scanning images, but I'm not sure there's a direct correlation.

Apr 9 00:49:13 iceman amavis[2672435]: (2672435-01) SA info: extracttext: [2672676] (/usr/bin/zbarimg) error: exit 15
I'm assuming exit 15 is SIGTERM and is perhaps related to low memory? This does sometimes happen at a time when the same email with an image to be scanned is received for multiple recipients, but not always.

Here is my ExtractText config:

# QR-code decoder
extracttext_external    zbar            /usr/bin/zbarimg -D --verbose=4 {}
extracttext_use         zbar            .jpg .png .pdf .webp image/(?:jpeg|png) application/pdf
add_header              all             ExtractText-Uris _EXTRACTTEXTURIS_

I traced one of them to an email with a large PDF attachment. I extracted the PDF from the email and ran the above zbarimg command on the file directly, and it exited with error 4 - no QR code found.

There are also coredumps available and this info in my system logs:

Apr  1 12:25:21 iceman systemd-coredump[2239236]: Process 2239234 (zbarimg) of user 980 dumped core.#012#012Module libbrotlicommon.so.1 from rpm brotli-1.1.0-3.fc40.x86_64#012Module libpcre2-8.so.0 from rpm pcre2-10.44-1.fc40.
x86_64#012Module libicudata.so.74 from rpm icu-74.2-1.fc40.x86_64#012Module libgraphite2.so.3 from rpm graphite2-1.3.14-15.fc40.x86_64#012Module libGLdispatch.so.0 from rpm libglvnd-1.7.0-4.fc40.x86_64#012Module libGLX.so.0 fr
om rpm libglvnd-1.7.0-4.fc40.x86_64#012Module libXau.so.6 from rpm libXau-1.0.11-6.fc40.x86_64#012Module liblzma.so.5 from rpm xz-5.4.6-3.fc40.x86_64#012Module liblz4.so.1 from rpm lz4-1.9.4-6.fc40.x86_64#012Module libcap.so.2

I also have the coredumps available for review, if someone can share how to go about using gdb to review it.