[neutron] neutron-metering-agent track router and floating IP traffic

Florian Engelmann florian.engelmann at everyware.ch
Thu Apr 18 11:51:06 UTC 2019


it looks like every notification is bound to the label:

grep -A6 bandwidth meters.yaml
   - name: 'bandwidth'
     event_type: 'l3.meter'
     type: 'delta'
     unit: 'B'
     volume: $.payload.bytes
     project_id: $.payload.tenant_id
     resource_id: $.payload.label_id

Wouldn't it make sense to change this to:

   - name: 'bandwidth'
     event_type: 'l3.meter'
     type: 'delta'
     unit: 'B'
     volume: $.payload.bytes
     project_id: $.payload.tenant_id
     resource_id: $.payload.label_id + "_" + $.payload.tenant_id




Am 4/16/19 um 12:14 PM schrieb Florian Engelmann:
> Hi,
> 
> I configured our neutron meters like:
> 
> labels:
> openstack network meter create --description "shared 
> bandwidth-public-out" --share bandwidth-public-out
> openstack network meter create --description "shared 
> bandwidth-public-in" --share bandwidth-public-in
> 
> rules:
> openstack network meter rule create --egress --remote-ip-prefix 
> 0.0.0.0/0 bandwidth-public-out
> openstack network meter rule create --ingress --remote-ip-prefix 
> 0.0.0.0/0 bandwidth-public-in
> 
> This created the iptables rules in all namespaces and collects data:
> 
> openstack metric measures show bc6c670f-cbb5-4fe1-974c-10455448fa27
> +---------------------------+-------------+---------------+
> | timestamp                 | granularity |         value |
> +---------------------------+-------------+---------------+
> | 2019-04-15T16:00:00+02:00 |      3600.0 | 581854.954545 |
> | 2019-04-15T17:00:00+02:00 |      3600.0 | 23169398.5625 |
> | 2019-04-15T18:00:00+02:00 |      3600.0 |      586508.9 |
> | 2019-04-15T19:00:00+02:00 |      3600.0 | 1276372.84375 |
> | 2019-04-15T20:00:00+02:00 |      3600.0 | 507482.911765 |
> [...]
> 
> But how to know which project caused which amount of traffic?
> 
> It looks like only one metric is created for each label?
> 
> openstack metric list | grep bandwidth
> | 0f9f20a4-1c60-442f-81cb-48285e3f70da | ceilometer-ew-low | bandwidth 
>                      | B       | efae8040-85d7-455a-958d-213a81ef98ef |
> | 1a058a79-965f-4a0a-a9ee-d482f57892ab | ceilometer-ew-low | bandwidth 
>                      | B       | 837ed34a-65d2-42ef-924e-fd3c7e5aaed6 |
> 
> 
> What does "shared" mean? I was not able to find any relevant documentation.
> Do I have to create a meter for each project or a rule or both? Is it 
> possible to define a "default" meter and rule for every existing project 
> and all future projects?
> 
> All the best,
> Flo
> 

-- 

EveryWare AG
Florian Engelmann
Senior UNIX Systems Engineer
Zurlindenstrasse 52a
CH-8003 Zürich

tel: +41 44 466 60 00
fax: +41 44 466 60 10
mail: mailto:florian.engelmann at everyware.ch
web: http://www.everyware.ch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5230 bytes
Desc: not available
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20190418/107b43a8/attachment-0001.bin>


More information about the openstack-discuss mailing list