Trouble replicating disk containing Fslogix profiles

[unsichtbarre]

Happy Friday,
(Case #05608496)

I am working on a situation where the client has a file server with three disks: OS (100 GB C:\) FIleshare (D:\ 1.5 TB) and FSlogix Profiles (E:\ 1.2 TB) running on VMware vSAN RAID 5 Storage Policy.

Recently, the FSlogix disk started falling badly out of Replication RPO. Local backups of the fileserver as well as all other servers seem unaffected for both replication and backup jobs.

We went so far as to ship the client a USB disk and manually re-seeded the entire file server (including FSlogix disk) to the replica site where the digests for all disks calculated remarkably quickly (about 4 hours total!) and one replication happened successfully, however the replica immediately started falling badly off RPO on the FSlogix disk again. When I say falling off RPO, I mean from 1 hour RPO to over 16 hours for the FSlogix disk alone!!

I have checked fragmentation and other disk properties and the FSlogix disk does present some fragmentation (3%), however that will occur within days of fully defragmenting the drive. Veeam identifies the replica Proxy as the bottleneck (in hotadd mode) - although it does not struggle for any other jobs including other large file server type replicas. Also, when I look at vSAN performance when jobs are running in hotadd mode, the vSAN latency seems high (greater than 15ms.), however maybe that is to be expected when mounting multiple disks to a single VM and reading from all of them at once.

THX,
-John

[PetrM]

Hi John,

It does not look like a known issue so you may contact our support team and reopen the case. I'm pretty sure that our engineers will find out the root cause and possible solutions very soon. Let's wait for the results of their research.

Thanks!