ISCSI -> Disk or Net throttle -> best practice?

May 16, 2013
25
0
21
hi, we have 5 nodes in a cluster configuration. All uses ISCSI over 2 x 1 GB in multipath configuration (ISCSI hosts are open-E DSS7 in cluster setup, connected to 2 x Raid5 volumes ( 6 x SAS 2TB Seagate) JBod via 6Gib/s SAS). At the moment we have ~30 KVM VMs active. Some with very low I/O, some with a lot of (NFS/Samba/LDAP/TFTP). As I started with Proxmox I never used any kind of limits. So I want to change it. Last week I saw a lot of veth0 RX drops in a few VMs with high I/Os (NFS/LDAP) and I think "unlimited" in the VM config is the problem. Now I asking for good practices of limiting the VMs:
  • list NFS Server: -> Used for $HOMES with ~few hundreds of active users per week (sometimes 100 per day)
  • list TFTP: -> Used for 100 diskless (NFS root) clients. ~30-40 active per day
  • list LDAP Server: -> Used for nearly everything Keep in mind, local disks aren't used, only ISCSI.
any suggestions? ps. sorry the format of the text is strange ... all newlines are removed ... why ever ..
 
Last edited:

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!