Even though if shows two available volumes it never uses them. It just continues with the repeating errors that the DB is failed.
Any clues?
Thanks!!

Even when manually reseeding and specifying the source server, any error seems to cause the reseed to revert to the active database copy as the source. I resorted to activating the database copy within the data-center where the reseed was taking place after several attempts to manually specify the source failed.

Rob,

I’m noticing the same thing you indicate in your post about the spare volume being renamed to a database after reseed and have also seen the throttling errors as well. I was wondering if the volume eventually repaired itself over that weekend in March?

Thanks,
Jerry

For clarification, it is my spare volume label that is renamed as part of the reseed process.

I’ve also noticed that if I attempt to simulate more than one disk failure within a few hours (by marking the disk offline), the Auto Reseed Manager goes into a throttled state (event 1117 in the Seeding crimson log). One server has been in this throttled state for 2 days so far – even with restarting the Exchange Replication Service & the spare being listed. I’m going to let it bake through the weekend & see if eventually completes. I’ve read on TechNet that the Disk Reclaimer will only run once every 24 hours (restarting the Replication Service seems to reset this timer) but it makes no mention of this task being throttled or limits on the number of disk failures that its able to process per day, etc..
I haven’t found any web articles to indicate that either of these is expected behavior so I thought I’d ask about your observations. Thanks!

It looks a bit weird the first time you do it. Using multiple folder paths to point to the same underlying volume is not all that common in Windows Server admin, so a lot of Exchange admins just haven’t seen it done before.

I guessed so. I used VMware workstation to build my lab and the underlying volume used is basically the same for the mount points. I’m trying to follow your implementation as I’m currently watching your videos on Pluralsight in preparation for the exam (70-345). I was wondering and I watched and re-watched your demos, in case I’m missing something. I also purchased your ebook on same subject and read autoseed section to ensure i fully understood the technique and process.

I appreciate your quick response.

What you’re seeing is normal and expected if you’ve mounted the same underlying volume into two different folder paths. Each folder is basically a view of the same volume, so you’ll see both sets of db and log folders because it’s the same location.

Whereas, DB01 should only contain DB01.db and DB01.log
If I delete DB02.db and DB02.log from DB01 mount point, they also get deleted in DB02 mount point.

Please, help.