Re: Documentation on the undocumented CCFLAGS parameter

Jacob, We are still not documenting the CCFLAGS --- because they change from release to release. They should only be used at the direction of tech support, and many have a significant impact on performance. For instance 0x400000 is used to check memory free references (v10). What it does is to do some extra checks with SAPI memory (i.e. UDR memory). "Jacob" <spamntrapf@yahoo.com> wrote in message news:1145910792.435496.311110@u72g2000cwu.googlegroups.com... > Hi family. > > We have recently discovered that several of our older servers have the > parameter CCFLAGS set (to 0x400000). I could find no reference to this > parameter in the Admin Reference or Guide so I went about searching > this group and found this gem, posted by Madison Pruett in Nov. 2002: > > >CFLAGS has existed (undocumented) since 5.x. > >The purpose of CCFLAGS is to do additional consistancy checking. These other checks > >include things such as memory scribbeling, memory checking, etc. Most of the setting > >of CCFLAGS have a significant impact on performance. It was a convienent way to do > >set the old page archive logic, but personally, I would have prefered that some other > >means besides setting a CCFLAGS bit had been chosen. It's too easy to set a wrong > >flag in CCFLAGS that can cause an adverse affect on the engine. > > Well, that explains why I couldn't find it. The postings also > indicated that nobody knows what that setting actually does, aside from > that it helped backups finish in their accustomed time after an upgrade > caused a degradation in backup performance. > > But from all the postings I did see, it seems the only setting anyone > knows about is 0x400000. Hey, there are 32 bit positions we could play > with! Why settle for just that one? Which brings me to point of my > question: > > Does anyone actually know what behaviors are controlled by the various > flags in CCFLAGS? And is that anyone willing to share this with us? > There must be bazoons of optimizations and debugging capabilities > hiding among those flags! (Probably dangers as well..) > > I'd LOOOOOVE to know this stuff. :-) > > Thanks mucho. > > -- Jacob >