Steve Hay wrote:
This allows the test to run successfully in isolation for me, but "nmake
test" now hangs when it reaches this test.
Even reverting this change I still find that "nmake test" hangs when it
gets to that test. Must be something else that's changed since Friday...
Same here.
apr_brigade_cleanup(void * 0x06370718) line 45
brigade_cleanup(void * 0x06370718) line 34
run_cleanups(cleanup_t * * 0x0636c080) line 1951 + 13 bytes
apr_pool_destroy(apr_pool_t * 0x0636c070) line 730 + 12 bytes
apr_pool_clear(apr_pool_t * 0x00a0f258) line 690 + 12 bytes
mpm_recycle_completion_context(CompContext * 0x009ce040) line 82
winnt_get_connection(CompContext * 0x009ce040) line 639
worker_main(long 1) line 694 + 9 bytes
_threadstartex(void * 0x0028d2a0) line 227 + 13 bytes
KERNEL32! 7c80b50b()
This is the infinite loop:
/*
* Bah! We can't use APR_RING_FOREACH here because this bucket has
* gone away when we dig inside it to get the next one.
*/
while (!APR_BRIGADE_EMPTY(b)) {
e = APR_BRIGADE_FIRST(b);
apr_bucket_delete(e);
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
