Recently, the UTS Messaging team implemented a disaster recovery (DR) solution for Listserv. Disaster recovery was necessary for Listserv due to its extensive use (currently over 6,500+ lists) and usage for critical emergency notifications. Previously, Listserv did not have a disaster recovery solution in the event that the server or data center went down. The server was always backed up with the Tivoli Storage Manager (TSM), but a restoration from TSM would be slow due to the small but numerous files that Listserv contains.
Currently, the Listserv resides in VMware and is running at the Whitestreet data center. I worked with Mike Lewis (Infrastructure) from the Systems Team and David Hauenstein (Infrastructure) from the Storage Team to setup a redundant storage solution utilizing Symmetrix Remote Data Facility (SRDF – a family of EMC products that facilitates the data replication from one Symmetrix storage array). A restoration without SRDF would potentially take days, while SRDF allows a full server restoration within an hour(s).
This solution provides a real-time, file-system-level copy of the entire server. In the event of a failure at the Whitestreet data center, the Listserv will be brought back online at the NDB data center. This solution provides much better resiliency for the Listserv in the event of a critical failure.
To access Listserv, or create a new list to use, please visit http://listserv.cc.emory.edu.
Leave a Reply