Data recovery from distributed SAN-systems
The idea to store data in a distributed manner was fulfilled in a Storage Area Network (SAN). SAN-systems create a huge amount of disk space achieved by integrating several servers into one system and allow getting fast parallel access to data from different servers. They are mainly used by corporate users for a reliable centralized data storing. SANs implement special file systems and particular controllers for access to storage. Several servers integrated into one system and the distributed data storing method make SAN a system with high fault-tolerance and great reliability. Nevertheless, seldom outages of the device still happen. Failures of SAN-controller, inappropriate reaction of the controller to a SAN-component failure and what more often occurs is an operator's error resulting in data deletion.
SAN-systems are unique with respect to data recovery for their non-transportability, thus excluding the possibility to take such system to a service center. Besides, these systems usually store a considerable amount of crucial information. In case of metadata outages or failures of the controller metadata the data can be recovered quite rapidly through the reconstruction of a file system on the basis of metadata and its further copying to a safer place. If a SAN-component has failed, it's possible to recover the data lost from this component only. If vitally important data has been deleted, the system must be switched to a read-only mode at once and you should immediately take copies of metadata and the journal of the SAN-system (usually 1-10% of the whole amount of a file system). Data recovery from these systems is available on remote-service basis offered by SysDev Laboratories. You need to know the file system type of SAN and send us the metadata image of SAN, so we can test the possibility of data recovery (the metadata do not contain any corporate information and rather inform about data distribution and file names on the SAN-system).