Monty and Alex,
maybe Monty discovered the source of the problem: Innobase indeed
modifies the sigmask of the thread which does the database intialization.
Maybe other threads then inherit the modified sigmask.
I had disabled the use of native Unix aio from Innobase, but forgot
the following segment of code into /innobase/os/os0file.c,
about line 920 in function os_aio_init:
.................................
}
#ifdef POSIX_ASYNC_IO
/* Block aio signals from the current thread and its children:
for this to work, the current thread must be the first created
in the database, so that all its children will inherit its
signal mask */
sigemptyset(&sigset);
sigaddset(&sigset, SIGRTMIN + 1 + 0);
sigaddset(&sigset, SIGRTMIN + 1 + 1);
sigaddset(&sigset, SIGRTMIN + 1 + 2);
sigaddset(&sigset, SIGRTMIN + 1 + 3);
pthread_sigmask(SIG_BLOCK, &sigset, NULL);
#endif
}
.................................
Alex, try commenting out those lines, and check if the signal 14 error
still occurs.
Regards,
Heikki
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
