[dpdk-dev] [PATCH v3 1/2] doc: live migration of VM with Virtio and VF

[Iremonger, Bernard]

Hi John,

<snip>

> > Subject: [PATCH v3 1/2] doc: live migration of VM with Virtio and VF
> >
> > This patch describes the procedure to be be followed to perform Live
> > Migration of a VM with Virtio and VF PMD's using the bonding PMD.
> >
> > It includes sample host and VM scripts used in the procedure, and a
> > sample switch configuration.
> 
> Hi Bernard,
> 
> Thanks for the doc. It is a complicated process so it is good to have this
> detailed how-to. Some comments below.
> 
> 
> > +
> > +Live Migration of VM with SR-IOV VF:
> > +====================================
> 
> 
> I wouldn't include a colon in the heading. Something like this may be better:

Ok, I will remove colons from headings.
 
>      Live Migration of a VM with SR-IOV VF
>      =====================================
> 
> 
> > +
> > +Live Migration overview for VM with Virtio and VF PMD's:
> > +--------------------------------------------------------
> 
> 
> Also, I wouldn't include the "Live Migration ... for VM with Virtio ..."
> in each heading. The context should be clear from the main sections of the
> doc. I this case I would just call the heading "Overview"

Ok, will change.
 
> 
> > +
> > +It is not possible to migrate a Virtual Machine which has an SR-IOV
> > Virtual Function.
> 
> Add (VF) in brackets at the first mention of Virtual Function here. Also for PF
> below.

Ok, I will add (VF) and (PF)

> 
> > +To get around this problem the bonding PMD is used.
> 
> I would add to the end of this paragraph:
> 
>     The following sections show an example of how to do this.

Ok, will add.
> 
> 
> Also I think the "Test Setup" header from below should be moved up to this
> point.
> 
> > +
> > +A bonded device is created in the VM.
> > +The virtio and VF PMD's are added as slaves to the bonded device.
> > +The VF is set as the primary slave of the bonded device.
> > +
> > +A bridge must be set up on the Host connecting the tap device, which
> > +is the backend of the Virtio device and the Physical Function device.
> > +
> > +To test the Live Migration two servers with identical operating
> > +systems
> > installed are used.
> > +KVM and Qemu 2.3 is also required on the servers.
> > +
> > +The servers have Niantic and or Fortville NIC's installed.
> > +The NIC's on both servers are connected to a switch which is also
> > +connected to the traffic generator.
> 
> I would prefix this paragraph with: "In this example, the servers have Niantic
> and .."

Ok, will do.
> 
> 
> > +The switch is configured to broadcast traffic on all the NIC ports.
> > +
> > +Live Migration with SR-IOV VF test setup:
> > +-----------------------------------------
> 
> Just "Test Setup" would be better. And move up, I think.

Ok, will do.
> 
> 
> > +
> > +Live Migration steps for VM with Virtio and VF PMD's:
> > +-----------------------------------------------------
> 
> Again "Live Migration steps" would be fine.

Ok, will change.

> 
> 
> Also I would change this paragraph to link forward to the scripts like:
> 
> 
>     The sample scripts mentioned in the steps below can be found in the
>     :ref:`lm_bond_virtio_sriov_host_scripts`
>     and :ref:`lm_bond_virtio_sriov_host_scripts` sections.
> 
> 
> And then put the targets in the relevant sections of the docs.
> 
>     .. _lm_bond_virtio_sriov_host_scripts:
> 
>     .. _lm_bond_virtio_sriov_vm_scripts:
> 
> 
Ok will do.

> > +
> > +The host is running the Kernel Physical Function driver (ixgbe or i40e).
> > +
> > +The ip address of host_server_1 is 10.237.212.46
> > +
> > +The ip address of host_server_2 is 10.237.212.131
> > +
> > +The sample scripts mentioned in the steps below can be found in the
> > +host_scripts and vm_scripts sections.
> > +
> > +On host_server_1: Terminal 1
> > +^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> 
> The documentation guidelines say to use ~~~~~~ as the third level heading
> underline. It doesn't really matter since the RST parsers figure it out.
> Up to you if you want to change it.

I will leave as is.

> 
> http://dpdk.org/doc/guides/contributing/documentation.html#section-
> headers
> 
> > +The following command only works with kernel PF for Niantic
> > +
> > +.. code-block:: console
> > +
> > +    testpmd> mac_addr add port 1 vf 0 AA:BB:CC:DD:EE:FF
> > +
> > +create bonded device(mode) (socket)
> 
> Should be a space after device.

Ok, will add a space.

> Also these lines describing the state or state
> change should be written have capital letters and full stops so they read like
> sentences. Also, if describing a state change you could add "now" so the
> reader is aware of the change. For example:
> 
>   - primary is port 1, 2 active slaves
>   + Primary is now port1. There are 2 active slaves.

Ok, will change.

> 
> > +
> > +Setup Virtual Machine on host_server_1
> > +
> > +.. code-block:: sh
> > +
> > +  #!/bin/sh
> > +
> 
> 
> I thought that the code block had to be indented 3 spaces but the RST
> parsers didn't complain. Might be worth changing them anyway to a
> consistent 3 or 4 space indent.

I think 2 spaces are enough (if I remember correctly)

> 
> John

Thanks for the review.

Regards,

Bernard.