The following post contains a few brief notes compiled from a day at the start of the month. Whenever I have attended an event, completed a project... post-event and when time is permitting, I like to review the things I should have known, things I need to remember better, things I have learnt, and find answers to questions raised. Enough waffle – on with the notes!
Understanding the role of NVRAM with NetApp Controllers
Source – https://communities.netapp.com/thread/13289 and Eugene Kashpureff
“In a single head environment all of NVRAM is utilized to log write operations. NVRAM is split into two pages - p0 and p1.
In an HA Pair the NVRAM on each member is split, and all writes are logged on the local and partner NVRAM, for both p0 and p1.
NVRAM is good for about three days.”
“Data ONTAP uses NVRAM to log network transactions as a data integrity measure. In case of a system or power failure, Data ONTAP uses the contents of NVRAM to restore network data to disk.”
“NetApp provides a "MetroCluster" option, that uses "SyncMirror" to provide a geo-cluster or active/active cluster between two sites up to 160 km apart (DOT 8.1.1 – previously 100km.)”
A MetroCluster consists of two controllers – one controller in Site A, the second controller in Site B.
Synchronous Replication introduces latency, see http://cosonok.blogspot.com/2012/12/synchronously-mirrored-sans
“Disk drives from different manufacturers may differ slightly in size even though they belong to the same size category. Right sizing ensures that disks are compatible regardless of manufacturer. Data ONTAP right sizes disks to compensate for different manufacturers producing different raw-sized disks."
Flash Pool allows you to cache both random reads and random writes.
Controller Failover Considerations for NFS and CIFS
NFS if a stateless protocol
CIFS is a stateful protocol
Source – Active/Active Controller Configuration Overview and Best Practice Guidelines (PDF from 2007)
“For clients or applications using stateless connection protocols, I/O requests are suspended during the takeover/giveback period, but can resume when the takeover/giveback process completes. For CIFS, sessions are lost, but the application may—and generally will—attempt to re-establish the session.”
DTA2800 Data Transport Appliance
The DTA2800 can be used for offline and online data migration over Fibre Channel and iSCSI, from legacy storage to NetApp storage.
Image: DTA2800 and NetApp Data Migration Service
Migrating CIFS Data
In a scenario where you want to consolidate many Windows Fileservers, many of which have the same share names, leverage NetApp MultiStore!
“MultiStore provides secure multi-tenancy capability to the NetApp Data ONTAP® operating system, allowing you to:
> Host multiple customers and/or departments on a single Data ONTAP storage system
> Consolidate many file servers, thus reducing cost and increasing staff productivity
> Perform simple and fast data migration
> Simplify your disaster recovery strategy”
Migrating Exchange Data
Typically migrating Exchange data is a case of enabling circular logging (so the transaction logs don’t balloon in size) and performing migrations. With NetApp you can leverage SnapManager for Exchange and take regular snapshots whilst migrating the Exchange data so as to keep the transaction logs down in size, whilst at the same time protecting the data being migrated, and not requiring store restarts to enable/disable circular logging.
Migrating Oracle Data (on Red Hat)
With Oracle Data on say Red Hat Enterprise Linux, an option for migration from legacy Fibre Channel storage is to migrate to NFS.