There are some theoretically possible execution paths along those lines that could lead to the assertion failure, but they don&#39;t seem likely to be the cause of the assertion failures you&#39;re seeing.<br />
<br />
I don&#39;t think the interruption scenario is plausible.&nbsp; SIGKILL would abort the process immediately (no library unloading), as would any unhandled signal.&nbsp; If a signal arrived, was handled, and the handler returned, the initializer would continue running and would finish registering all of the keys.&nbsp; Signal handlers are not allowed to call exit().&nbsp; Signal handlers are allowed to call _exit(), but that would bypass library unloading.&nbsp; On top of that, the &quot;graceful reload or restart&quot; operation would not seem likely to send a signal to processes while they are in the middle of initializing the GSSAPI library.<br />
<br />
The failed initializer scenario actually kind of tracks, because of a bug in gssint_mechglue_init() (&nbsp;<a href="https://github.com/krb5/krb5/blob/master/src/lib/gssapi/mechglue/g_initialize.c#L106">https://github.com/krb5/krb5/blob/master/src/lib/gssapi/mechglue/g_initialize.c#L106</a>&nbsp;) where error values can be ignored.&nbsp; (If errors were correctly handled, the INITIALIZER_RAN() guard in gssint_mechglue_fini() would not consider the initializer to have run because it returned an error.)&nbsp; But I don&#39;t think this is a likely scenario for two reasons.&nbsp; First, you say you&#39;re not using krb5, and gssint_mechglue_init() doesn&#39;t actually run until a GSSAPI function is invoked.&nbsp; (It&#39;s possible that the PHP module invokes a GSSAPI function when it starts up, I guess.)&nbsp; Second, failures inside gss_krb5int_lib_init() should be vanishingly uncommon; one doesn&#39;t really expect mutex initialization to fail.<br />
<br />
I will fix the bug in gssint_mechglue_init(), but to diagnose the actual problem with any confidence, either I need to be able to reproduce the problem, or someone who reliably sees the problem needs to debug it in situ, likely by adding a bunch of instrumentation to the initializer and finalizer code.<br />
&nbsp;