Ok, this isn't really a scenario as much as it actually happened to me. All I know is that I did something wrong since my recovery did not go anywhere near as planned.
My DP server is a standalone 1U server with raid storage. It had hardware issues with the raid (sync errors) and the filesystem became corrupt. There was no chance of recovering the database off the disk.
I have 2 backup to disk servers and a 24 slot tape library. The IDB was set to backup twice daily to disk followed by a copy from disk to tape. There are 16 tapes in the daily rotation pool. We do another copy on the weekends for offsite.
I brought up another server, patched the OS, and installed DP to the same patch level as my prior server. This is where I realized I had a probelm -- I had no idea which tape the IDB was on in the daily pool.
I ended up doing a tape scan of all 16 tapes and checking the first reported session on the tape, then cancelling and reading the next tape. I took my best 2 candidates and let them run for 3 hours to import the catalog. The whole process took about 4-5 hours with 2 tape drives until the IDB was identified and catalogued.
Once I restored the IDB I was back and running again, but this has me wondering what I should have done to make sure I could identify where the IDB was backed up. I'm open to suggestions and discussion of backing up the backup server itself.
