Working with a customer trying to help him re-index his database. Database is a little over a terabyte and I gave him a re-indexing script to run on his SQL Server 2005 Enterprise server.
FYI
It had been recommended by a 3rd party load testing company that all of our re-indexing be done in "offline" mode so the re-indexing was done in offline mode for this customer.
The first reindex attempt stopped with a "severe" error message in the "query window" when sql reached the largest table in the system, which I found out contains about 29 milliion rows. The previous indexes had been successfully rebuilt or reorganized based
on the percentage of fragmentation.
I asked for the sql error logs and found the "severe" error message written to the log around the time of a "sql server restart". I asked the customer and he admitted to restarting sql server. I suspected that the reindex script had not completed but could
not be sure.
I decided to approach it next by rebuilding the index on the huge table only. He called me the next day and told me the rebuild had failed. He sent me the error logs and he also informed me he had killed the index query after it had been running for over
24 hours.
I found some IO errors in the sql error logs referring to "15 seconds" had been exceeded a
View Complete Post