>>>Nope..doesn't make sense. If you have some sort of write cache on the hardware, then I could see it, but that's not an FP issue.
>>
>>I used to update the key field with the order tag active and index corruption was frecuent, since I left this practice everything is ok.
>
>
>Interesting exchange. Does anyone else have an opinion on this?
>
>Emmanuel: Don't update an indexed field when that index is active.
>
>Craig: Should not matter.

I have index corruption occasionally, due to busy table usage and network problems. I suspect that though Craig is generally correct, the probability of corruption may be slighty higher when an index is in use and being updated at the same time. The tag may be already pulled across the network, and slightly more vulnerable in that situation. Just a guess, but with a little evidence, also. One tag I use a lot on several tables and that has frequent editing updates goes bad more than other tags.

But it's not something I'd be too concerned about, it would cause more trouble to try to avoid tag active-updates that it'd be worth, I think...Occasional rebuilding of indexes is really the best solution for all indexes (unless the tables/indexes can be kept in readonly server directories).