Disk IO Graph

[ctradmin]

Hi,

We have a Proxmox 3.0 Cluster Running with Ceph. For each VM Proxmox generates 4 rrd graphs.

CPU, Memory, Network and Disk IO.

My Problem with Disk IO is that at some VMs theres a P to measure the amount of IO, and at others is a T or a k. So what does they mean?


Thanks

 

[ctradmin]

bump

 

[dietmar]

That is the amount of data read/written in bytes, scaled by following units:

k (Kilo, 10e3)
M (Mega, 10e6)
G (Giga, 10e9)
T (Terra, 10e12)
P (Peta, 10e15)

 

[spirit]

Hi, P is petabyte.

but it's a bug in proxmox.
I have sometime this bug when I stop/start a vm.

The io value is a delta between old total io amount tranferred and new total io amount transferred.
I need to look at it. (I think they are already a bugreport in bugzilla)

 

[Raymond Burns]

Has there been any advancement with this?
I get this a lot on Hardware RAID

 

[dietmar]

Has there been any advancement with this?

I think this is already fixed in git (but will only apply to newly generated RRD DBs).