# 4710 Creating Indexes and Other Batch Operations Cause Corruption with Global File I/O Buffering Enabled

---

I have a ElevateDB Server running as service in Windows 10 with this configuration in edbsrvr.ini :

Cache Modules=1
Buffered File IO=1
Buffered File IO Settings=
Buffered File IO Flush Check Interval=60
in ElevateDB Manager I have this configuration (other as default) :

If I create a new index in a big table with a lot of indexes (I haven't tried in another table) all works perfecte and if I verify the Table all is perfect.

But if I stop the service and start again when I verify the Table an error like this appears :

Verification error for the index idx_Prova in the Articles table file D:\Bases Locals\ElevateDB\UniCode\AndoFarma_Gestio0002\Articles.EDBIdx (Invalid next index page 81802 pointer (1859193118)) - Fixed

When it's repair all works great again; but if I stop and start the service again, another time the same error.

The only way to stop this cycles is modify the Buffered File IO=0 repair the table an set Buffered File IO=1 and all works perfect until I create a new index (stoping and starting the service every time I changed Buffered File IO).