Re: Weirder and weirder, Re: CEE0810: I think I have something fighting with itself, or with my own code

Last week, it wasn't throwing an exception on the V6 box if *any* memory manager is explicitly in place, even the default, even if I were to set QIBM_MALLOC_TYPE to some nonsense value.

NOW, it's throwing an exception with the debug memory manager in place.

And I have a stack dump, generated by the debug memory manager (irrelevant columns and extra space squeezed out in hopes of keeping it from being mangled):

MODULE NAME PROCEDURE STATEMENT#
QC2ALLOC _C dump stack 32
QC2ALLOC send_message__FiPvUi 62
QC2ALLOC send_C2M1212__FPvUi 2
QC2ALLOC do_free_debug__FPv 15
QSOGSKSY gsk_attribute_get_cert_info 1085
COMMSSLR4 SSL_DEBUG_CERT_INFO 6250
COMMSSLR4 COMMSSL_UPGRADE 5199
HTTPAPIR4 HTTP_PERSIST_OPEN 6294
HTTPAPIR4 HTTP_URL_GET_RAW 5697
GEOLOOKUP GEOCODE 3647
GEOLOOKUP GEOLOOKUP 3508
GEOLOOKUP _QRNP_PEP_GEOLOOKUP

Does this tell us anything? The "SSL_DEBUG_CERT_INFO" suggests that maybe my hypothesis about the Google web service SSL/TLS negotiations to get something a V6 box can handle are too lengthy, and that I can therefore, with essentially zero likelihood of this application ever being run on a V6 box in the real world, ignore the problem.



--
JHHL