Yeah I thought about doing something like that...however I wanted to kind of 
look over the last 3 to 6 months for example...in order to account for any 
periods of high IO over the long term.

I can also look at each vm and add up the max IO of each of them and come up 
with the total that way.

Can you give me a brief explanation about how you are gathering the per vm 
stats?  Maybe I can figure out what is wrong with the cacti setup?

I have tried graphing all the hard drives (sda, sdb, etc) including dm-2 
(/var/lib/vz) but they do not seem accurate at all.

Thanks,

Shain


Shain Miley | Manager of Systems and Infrastructure, Digital Media | 
[email protected] | 202.513.3649

________________________________________
From: Dietmar Maurer [[email protected]]
Sent: Friday, January 31, 2014 10:16 AM
To: Shain Miley; [email protected]
Subject: RE: Proxmox IO graphs

> The reason I am asking is this...I have been attempting to monitor the overall
> host IO on each of our servers in order to determine what kind of performance 
> I
> would need in order to move from our current local disk storage setup...to one
> that uses shared storage (Ceph).

maybe you can simply use something like "iostat -x 5" for that purpose (package 
sysstat):




_______________________________________________
pve-user mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user

Reply via email to