Vault Basic... Redundant Server

[Rob.O]

Hi all,

I have done a lot of searching and have not found a discussion on this topic so I would like to get a discussion going about redundant servers when using Vault Basic.

In a nut shell... We have a Vault server that we highly depend on. It is a workhorse serving 30+ users on a daily basis. Since we are currently using Vault Basic, we have daily Vault backups, but no replication.  We are attached to a manufacturing plant and must continually feed the plant drawings or the company loses money.

If the single Vault server crashed, it would be devastating.  No one would be able to access the Vault files stored on that server until brought back online or another server was configured to replace it.

I am wondering how others that are using Vault Basic and a single server handle this potential risk.  Is it possible to create a redundant server that runs alongside the main server.  Maybe you have a duplicate server model sitting on the shelf just in case something like this should occur?

What are our options?

We plan to upgrade to Vault Pro, but it's not in the budget at this time.  At that point we can use replication, but until then...

Thanks for any advice!

[MarkMazour]

We have a DR (disaster recovery) plan that uses a second server with Vault installed and a few batch files on both servers. Using Task Scheduler I run the backup (via command line), copy the backup folder to the DR server. On the DR server I have a .net exe file that I wrote that gets the backup folder name (as the name changes on each run to the date and time schema) and rewrites a restore bat with the new foldername. After that I run the new restore.bat file. If there is a situatiuon where the production servers crashes or we lose a SAN all I have to do is point our users to the dr server.

Of course we would lose a day of work but this meets our RPO.

[swalton]

We run Vault Collaboration 2012 with a replicated filestore, but not database.

We run our vault on a virtual machine.  We have two physical servers that host the virtual machines and a SAN that stores the disk images.  One machine is the production host, the other is the hot spare.  Every so often, we swap machines.  We image the SAN to an offsite machine twice a day.

If we have a hardware failure on the production host, we switch to the hot spare.  If we have a hardware failure on the local SAN, we go get the offsite SAN and put in into the local server rack.