Ant wrote:
> On 23 апр, 22:14, Mike Christie <[EMAIL PROTECTED]> wrote:
>> I do not think the reo box likes obeys the immedate bit and probably
>> thinks we should be incrementing the cmdsn.
>> If you set
>> node.conn[0].timeo.noop_out_interval = 0
>> node.conn[0].timeo.noop_out_timeout = 0
>> Does this help?
> Looks very good, i transfer 1,4 Gb film into and from linux box
> without any errors in logs of linux machine.
> 1. Are we must be worried about setting this parameters to 0?

Yes and no. The initiator can work fine without this setting, but it is 
a feature, so it does provide some benefit. If you are using multipath 
you want to set those values to non-zero, so we can detect errors in the 
network and fail IO to new paths faster.

I am pretty sure this is a bug in the target, because I have seen other 
ones make a similar mistake. I would contact your vendor. Feel free to 
refer them to me, and I can work with them to get it fixed.

> 2. Am i must change back:
> node.session.cmds_max = 4
> node.session.queue_depth = 2
> To default values 128 and 32 ?

Changing these values and the nop ones are not related. Those values 
depend on how much IO your system, network and target can support.

> (asking because behaviour changed when i chenge this values, first few
> megabytes transfers very fast and then paused like cache ends, and go
> back to normal speed)
> > 

You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at

Reply via email to