"ESEutil /d /p" will have a slightly different result. The "/d" tells ESEutil to defragment the designated database. The "/p" option used with the "/d" instructs ESEutil to leave the newly created defragmented database in the temporary work area, and not to overwrite the original database.

Now slightly modify the command to "ESEutil /p" and the actions taken on the designated database are extremely different. The "/p" evokes the Exchange "Repair" mode. At first glance this sounds like a great thing to do, and it couldn't hurt to try because repairing the database should be beneficial right? Wrong!

This command actually invokes a "Hard Repair" mode of ESEutil. This means that ESEutil will attempt to repair corrupt pages, but it makes no attempt to put the database in a consistent state. If it finds problems that cannot be corrected, those pages will be discarded. Each page contains data, therefore each discarded page represents data loss. Discarding certain pages of the database can actually render it useless. In other words, wave goodbye to your data.

Sometimes, using the repair mode is the only way to fix a database. In the vast majority of situations, however, it should be avoided except as a last resort, and there are specific steps that should be taken pre and post use of "Repair /p" mode.

About ISinteg
The purpose of the Microsoft ISinteg utility is to inspect and fix weaknesses within the IS (Information Store). ISinteg looks at the mailboxes, public folders, and other parts of the IS, checking for anything that appears to be out of place. ISinteg scans the tables and B-trees that organize the ESE pages into their logical structures. In addition, the tool looks for orphaned objects, or objects that have incorrect values or references.

Because ISinteg focuses on the logical level rather than physical database structure, it can repair and recover data that ESEutil can't. When looking at the physical database level, ESEutil might find the data to be valid because it looks for things such as page integrity and B-Tree structure. Data that appears valid to ESEutil from a physical view of the database might not be valid from a logical view.

For example, data for various IS tables like the message, folder, or attachments table may be intact, but the relationships among tables, or records within tables, may be broken or incorrect because of corruption in the logical structure. This corruption can render the database unusable.

Logical corruption of your Exchange Server databases is problematic, and much more difficult to diagnose and repair than physical corruption. The user and administrator are, typically, unaware of a logical corruption occurrence. No specific symptoms identify logical corruption. Often, when an administrator discovers the logical corruption, it's too late for any repairs to take place.

ISinteg can be run in one of two ways. The Default mode, in which the tool runs the tests you specify and reports its findings, or the Fix mode, where you specify optional switches, instructing ISinteg to run the specified tests and attempt to fix whatever it can.

The most important thing about running ISinteg is to run the command until it no longer reports any problems. Just running the command once doesn't guarantee the Information Store is functioning properly. Depending on the size of the Information Store, the process can take a long time. However, it ensures the databases are properly functional. For additional information on ISinteg, please refer to the GOexchange FAQ on our Web site--Microsoft Isinteg at http://www.goexchange.com/faq_GEvsMStools5.html.

Troy Werelius is CEO of Lucid8 LLC at http://www.lucid8.com/, the creators of GOexchange, the automated maintenance solution for Microsoft Exchange 5.5, 2000 and 2003 servers. GOexchange prevents disasters, repairs problems, and accelerates performance. Visit www.goexchange.com for a free demo copy of GOexchange.