Re: QoS for iSCSI target?

[Chris Friesen <chris.friesen@windriver.com>]

On 03/11/2016 01:45 AM, Nicholas A. Bellinger wrote:
> On Thu, 2016-03-10 at 23:30 -0800, Christoph Hellwig wrote:
>> On Thu, Mar 10, 2016 at 04:24:25PM -0600, Chris Friesen wrote:
>>> Hi,
>>>
>>> I'm looking for information on whether the iSCSI target in the kernel offers
>>> any way to do QoS between traffic driven by different initiators.
>>>
>>> I'm trying to make sure that one initiator can't do a denial-of-service
>>> attack against others.
>>>
>>> Does the kernel target have this sort of thing built-in, or do I need to
>>> look at network traffic-shaping to achieve this?
>>
>> It doesn't right now, but it shouldn't be hard to integrate it with
>> blk cgroups.
>
> For iscsi-target application QoS, the per session command sequence
> number window depth (CmdSN) exists to enforce per InitiatorName limits
> via TPG default_cmdsn_depth + se_node_acl->queue_depth configfs
> attributes.
>
> Note these values can be changed on the fly for iscsi-target using
> explicit se_node_acl->acl_group, but currently require a se_session
> reinstatement event for updated ExpCmdSN + MaxCmdSN to take effect.

On a slightly different note, is there any way to throttle or limit the overall 
bandwidth consumed by the iSCSI target in the kernel?  I'd like to ensure that 
the iSCSI traffic doesn't completely swamp the host accesses to the same block 
device.

I suppose I could do networking-based traffic shaping, but are there any 
controls in the block IO subsystem?

Chris