Hi
In our environment we have LUNs from our SAN that are presented to our file server via in guest mapping
fileserver and vbr server are virtual machines
fileserver is a Microsoft failover clustered running with 2 nodes (not using CSV)
datastores are vmfs6, repository is a ReFS 64kb volume on another SAN
I'm fairly new to Veeam, and the way that I have set it up is using a protection group for the cluster, and then a volume based agent backup on that to get at the actual LUNs
We use an EQL SAN meaning that I cant integrate to that with Veeam and grab the LUNs at the storage level
I'm finding that when backing up the LUNs from the fileserver agent job that it causes a few traffic problems and a dip in our network speeds
I am looking at the various throttles etc within Veeam, but In my mind 250 mb/s over a 10gb/s network shouldn't cause an impact so I may have something wrong in my understanding
is anyone else doing anything similar?
-
- Influencer
- Posts: 10
- Liked: never
- Joined: Jul 29, 2019 9:42 am
- Full Name: Mr Gary Dufley
- Contact:
-
- VP, Product Management
- Posts: 7081
- Liked: 1511 times
- Joined: May 04, 2011 8:36 am
- Full Name: Andreas Neufert
- Location: Germany
- Contact:
Re: backing up iSCSI LUNs
Usually non of the Veeam components are the bottleneck themself.
You have correctly assumed that in Guest mounts of volumes (including the pass through volumes) can only be backed up with a Veeam Agent.
Backup speed is based on Source and Target disk speed. As well the network and such are a limitating factor. You can copy a big file over SMB to the target server from the in guest volume to compare speed. It should be roughly same speed and you can work on your bottlenecks that way without Veeam in the mix.
You have correctly assumed that in Guest mounts of volumes (including the pass through volumes) can only be backed up with a Veeam Agent.
Backup speed is based on Source and Target disk speed. As well the network and such are a limitating factor. You can copy a big file over SMB to the target server from the in guest volume to compare speed. It should be roughly same speed and you can work on your bottlenecks that way without Veeam in the mix.
Who is online
Users browsing this forum: Google [Bot] and 12 guests