Open Stack

Hi Sylvain,

I think I've caught-up with all your reviews, but I still did have some
questions on the iptables rules, below.

One other question, and maybe it's simply a future enhancement, but is the only
way to setup these meters using neutronclient?  I think being able to specify
these in an .ini file would be good as well, which is something I'd want to do
as a provider, such that they're always there, and actually not visible to the
tenant.

On 07/11/2013 10:04 AM, Sylvain Afchain wrote:
> Hi Brian,
> 
> You're right It could be easier with your approach to get and keep the traffic counters.
> 
> I will add a new method to get the details of traffic counters of a chain.
> https://review.openstack.org/35624
> 
> Thoughts?
> 
> Regards,
> 
> Sylvain.
> 
> ----- Original Message -----
> From: "Sylvain Afchain" <sylvain.afchain at enovance.com>
> To: "Brian Haley" <brian.haley at hp.com>
> Cc: openstack-dev at lists.openstack.org
> Sent: Thursday, July 11, 2013 11:19:39 AM
> Subject: Re: Change in openstack/neutron[master]: Add method to get iptables traffic counters
> 
> Hi Brian,
> 
> First thanks for the reviews and your detailed email.
> 
> Second I will update the blueprint specs. as soon as possible, but for example it will look like that:
> 
> Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
>     pkts      bytes target     prot opt in     out     source               destination         
>        55       245 metering-r-aef1456343  all  --  *      *       0.0.0.0/0            0.0.0.0/0   /* jump to rules the label aef1456343 */        
>        55       245 metering-r-badf566782  all  --  *      *       0.0.0.0/0            0.0.0.0/0           

So are these two used to separate out what you don't want to count from what you
want to count?  Seems the jump to the r-aef1456343 will filter, then the
r-badf566782 will count per-subnet?  I'm just trying to understand why you're
splitting the two up.

> Chain metering-l-aef1456343 (1 references)	/* the chain for the label aef1456343 */
>     pkts      bytes target     prot opt in     out     source               destination         
> 
> Chain metering-l-badf566782 (1 references)	/* the chain for the label badf566782 */
>     pkts      bytes target     prot opt in     out     source               destination      

These two chains aren't really doing anything, and I believe their packet/byte
counts would be duplicated in the calling rules, correct?  If that's the case I
don't see the reason to jump to them.  Our performance person always reminds me
when I increase path length by doing things like this, so removing unnecessary
things is one of my goals.  Of course we're doing more work here to count
things, but that needs to be done.

> Chain metering-r-aef1456343 (1 references)
>     pkts      bytes target     prot opt in     out     source               destination         
>        20     100 RETURN     all  --  *      *       0.0.0.0/0           !10.0.0.0/24      /* don't want to count this traffic */   
>        0        0 RETURN     all  --  *      *       0.0.0.0/0           !20.0.0.0/24      /* don't want to count this traffic */   
>        25      145 metering-l-aef1456343  all  --  *      *       0.0.0.0/0            0.0.0.0/0	/* count the remaining traffic */     

This has excluded (!) certain subnets, then allowed anything else to pass through.

> Chain metering-r-badf566782 (1 references)
>     pkts      bytes target     prot opt in     out     source               destination         
>        0        0 metering-l-badf56678  all  --  *      *       0.0.0.0/0            30.0.0.0/24 /* want to count only this */

This is only counting packets going to a certain subnet.

I'm still trying to get all these changes running in devstack, so sorry for all
the questions.

Thanks,

-Brian

> 
> Of course the in/out interfaces will be set in order to get the ingress or the egress traffic.
> 
> I agree with you I could add a single rule to the chain of the label and get the traffic of the first entry, though I found this approach less generic. 
> I mean, to be forced to add a rule at the top of a chain to get its traffic. My approach is I don't want the counters of a specific rule but I want to count
> the traffic going through the chain.
> 
> Thoughts?
> 
> Regards,
> 
> Sylvain.
> 
> ----- Original Message -----
> From: "Brian Haley" <brian.haley at hp.com>
> To: "sylvain afchain" <sylvain.afchain at enovance.com>
> Cc: openstack-dev at lists.openstack.org
> Sent: Thursday, July 11, 2013 2:30:24 AM
> Subject: Re: Change in openstack/neutron[master]: Add method to get iptables traffic counters
> 
> On 07/08/2013 01:10 PM, Sylvain Afchain (Code Review) wrote:
>> Sylvain Afchain has posted comments on this change.
>>
>> Change subject: Add method to get iptables traffic counters
> <snip>
>> --
>> To view, visit https://review.openstack.org/35624
> 
> Hi Sylvain,
> 
> Instead of trying to ask questions directly in the review itself (since it will mess-up formatting) I'll just send this to you and the list since I had some questions on the traffic counter changes you've been doing.
> 
> First, thanks for working on this, it's definitely something I'm interested in, and I'm trying to review all your changes.
> 
> Second, do you have more than just the short description from the blueprint for how the iptables chains/rules will look like when created?  I'm still a little confused with this change (above) and how it's matching chains to get packet/byte statistics.  I'm thinking it can be done within a single chain so that you can do an 'iptables -L $chain' call to get just what you need, instead of parsing the entire table.
> 
> For example, I did something similar in Nova (out of tree), and it used a single chain per-VM, such that you could get it's statistics with a single iptables call like:
> 
> (sorry if this wraps)
> $ sudo iptables -t mangle -L nova-meter-output-91 -n -v -x [-Z]
> Chain nova-meter-output-91 (1 references)
>     pkts      bytes target     prot opt in     out     source               destination         
>   805210 247931149            all  --  *      *       0.0.0.0/0            0.0.0.0/0        /* inst-91 packets transmitted total */ 
>    15510   964648             all  --  *      *       0.0.0.0/0            x.y.0.0/16
>    21282  3075403             all  --  *      *       0.0.0.0/0            x.z.0.0/16
>    [...]
> 
> None of the rules in the chain has a jump target, so they simply count packets/bytes as they pass through, and the chain is called from a single location based on IP address, so in iptables-save format it looks like this:
> 
> -A nova-meter-output -s $my_ip/32 -i bridge1 -j nova-meter-output-91 
> -A nova-meter-output-91 -m comment --comment "inst-91 packets transmitted total"
> -A nova-meter-output-91 -d x.y.0.0/16
> -A nova-meter-output-91 -d x.z.0.0/16
> [...]
> 
> Obviously with Neutron, and doing this at the router egress, things change, but I think it could still be done in a single OUTPUT chain in the filter table.
> 
> Thoughts?
> 
> -Brian
>