interview question and answer

July 6, 2014

Loose Truncation In Exchange 2013 Sp1

I just happened to read about what is loose truncation in Exchange 2013 SP1 and it seems to be very interesting.Microsoft is coming up with all new features in every release of exchange and I am very eager to know what will be the feature that is available in the next release of exchange..
Loose Truncation
Lets see as to what this Loose Truncation got to do in exchange 2013. In early versions of exchange, Log truncation in a DAG environment was bewildering.Lets take this scenario where I have a two member DAG and in which one of the node is suspended for some reason. I am backing up my database, and I find that the database gets backed up but the logs remain unchanged because they are not yet shipped to the passive. If am administrator does not notice that the backup has unaltered the logs the logs keep growing and fill the drive and eventually there will be a database dismount due to less space.Now in exchange 2013 SP1 they have introduced a new feature that is called loose truncation where active copy will check the free space in the drive and deletes it own logs…
This feature is not enabled by default and I was wondering if there is any PowerShell command to enable it.Then I felt that Microsoft has gone to windows again rather than PowerShell. The reason I say this is  the process of enabling Loose truncation is in the registry and few DWORD values in a particular registry. The path to enable loose truncation isHKLM\Software\Microsoft\ExchangeServer\v15\BackupInformation
we have to create 3 registry key manually to enable this type of truncation
LooseTruncation_MinCopiesToProtect
This key is used to enable loose truncation. It represents the number of passive copies to protect from loose truncation on the active copy of a database. Setting the value to 0 disables loose truncation.In my case I have a 2 node DAG and I would specify 1 as the value as I have one passive node to protect.
LooseTruncation_MinDiskFreeSpaceThresholdInMB
This talks about the available disk space in MB. So its at this mentioned amount of free space where the Loose truncation will trigger and start the deletion of the logs. By default its 200GB and if we want we can either decrease or increase the value depending upon our requirement
LooseTruncation_MinLogsToProtect
This Registry says the amount of logs that needs to be protected from the truncation. This applies to both passive and active but in passive the number is calculated depending on the last available log.
Let`s say that the loose truncation has done it job on the active node but in some point in time passive node has come back up. At this time the passive node will remain failedandsuspended. In order to get the passive node to healthy state we either have to manually reseed the database or enable Autoreseed.

No comments: