FRS with Volume Shadow Copy (VSS) - does this work?

[Chris Warwick]

I'm happily running both Symantec/Norton antivirus and Diskeeper on my FRS
replicated server volumes. This isn't a problem since I'm using current
versions of both of these products and they're FRS savvy and don't cause
unwanted FRS replication events (see Q815263).

I want to use the Volume Shadow Copy service (VSS) on the server volumes
too (well, why not, it's free!) but when I enabled VSS, FRS seemed to start
churning - so I turned VSS off again. I'm not absolutey sure that the FRS
activity was caused by VSS, or whether the replication levels caused by VSS
were going to be acceptable (I didn't want ALL my replicated folders being
synched continuously), so I thought I'd come back to this issue when I had
more time to look into it more thoroughly.

Scanning around on the web and in the FRS or VSS deployment guides, I
haven't been able to find anything that says "You cannot run VSS on volumes
hosting FRS replicas" - and I'd expect to see such a caveat if this
configuration were a problem (particularly bearing in mind the fact the
Sysvol is going to be affected if it *is* a problem).

On the other hand, I haven't seen anything that say I can run FRS and VSS
together.

SO (sorry for the verbose background!) the simple question is: can I run
VSS on volumes also hosting FRS replicas without causing FRS replica storms
- and what volumes of FRS replica traffic should I expect for each VSS
snapshot?

Thanks in advance,
Chris

 

[Jill Zoeller [MSFT]]

Chris, can you explain why you thought FRS seemed to start churning? When
did the churning occur, and how did you detect this?

I haven't seen any guidelines on not using these technologies together. In
fact, I just ran a simple test here by doing the following:

I cleared out the FRS outbound log.
I enabled shadow copies on my volume E, which took a snapshot immediately.
I checked the outbound log and no churning began (i.e., no files were
replicated)
I changed one file and checked the outbound log and the file showed up
there.
Just for fun I created another snapshot.

So, I'm not seeing any churn as a result of enabling replication. If you can
explain what you're seeing on your end, I'll investigate.

 

[Chris Warwick]

Chris, can you explain why you thought FRS seemed to start churning?
When did the churning occur, and how did you detect this?

I haven't seen any guidelines on not using these technologies
together. In fact, I just ran a simple test here by doing the
following:

I cleared out the FRS outbound log.
I enabled shadow copies on my volume E, which took a snapshot
immediately. I checked the outbound log and no churning began (i.e.,
no files were replicated)
I changed one file and checked the outbound log and the file showed up
there.
Just for fun I created another snapshot.

So, I'm not seeing any churn as a result of enabling replication. If
you can explain what you're seeing on your end, I'll investigate.

Sorry it's taken a while to get back to you on this - I was trying to find
out how to clear my FRS outbound log!

Anyway, once again I've taken the plunge (that's three times today!) and
have gone ahead and enabled shadow copies on my server volumes. Absolutely
no sign of FRS activity, so it must have just been coincidence last time I
tried it with something else replicating at the same time. I turned off
the shadow copies before because I was just about to leave the servers
alone for a bit (I have a social life too!) and didn't want FRS getting in
a state trying to deal with a massive amount of replication.

So, thanks for all your help Jill - my todo list is significantly shorter



(If you can manage to answer the CSC and Dfs question sometime as well I'll
definely post you some beer tokens!)

Regards,
Chris


[PS. How *do* you clear the outlog, btw?!]