zend_mm_heap corrupted

[pkhadka56]

I'm using wordpress:6.6.2-php8.3-apache in Kubernetes and started seeing the following error on the logs, the pods are restarting due to this. I suspect the reason for this is opcache. Any help on how to resolve this?

[mpm_prefork:notice] [pid 1:tid 1] AH00170: caught SIGWINCH, shutting down gracefully
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted
zend_mm_heap corrupted

[LaurentGoderre]

From what I can find, it could be that something is sending a signal to apache to restart and then the shutdown is experiencing some issue.

https://stackoverflow.com/a/76049084 https://stackoverflow.com/a/35212026

[yosifkit]

I think the zend_mm_heap corrupted mostly unrelated is just the result of the container stopping when getting its stop signal. The STOPSIGNAL defined in the Dockerfile comes from the parent php:8.3-apache image and is SIGWINCH. So, this is likely a regular signal coming from Kubernetes (maybe because of failed liveness probe checks or similar).