If you don't give the 'where' clause -- you get the below effect..

-
Raj

On Apr 20, 2015, at 1:51 AM, [email protected] wrote:

> Hi,
> 
> Thanks for both replies. Worked perfectly.
> 
> Now I have a related doubt: Is there any way to retrieve all the tuples from 
> the API without a where clause? (Using only the time restrictions)
> 
> 
> Best regards,
> Alberto.
> 
> Quoting Raj Reddy <[email protected]>:
> 
>> Hi Alberto,
>> 
>> An example query is given below, one can get avg load every 9min for every 
>> system, using the query below.. [SUM(cpu_info.one_min_cpuload)/ 
>> COUNT(cpu_info)]
>> 
>> regards,
>> -
>> Raj
>> 
>> 
>> 
>> 
>> On Apr 14, 2015, at 1:21 PM, Megh Bhatt <[email protected]>
>> wrote:
>> 
>>> Hi Alberto,
>>> Please see inline ...
>>> 
>>> On Apr 14, 2015, at 10:34 AM, <[email protected]> <[email protected]> 
>>> wrote:
>>> 
>>>> Dear all,
>>>> 
>>>> I want to have aggregations over the statistics but I don't know how to do 
>>>> aggregations. I have a test query as follows:
>>>> {
>>>>  "end_time": "now",
>>>>  "select_fields": [
>>>>      "T", "cpu_info.cpu_share", "SUM(cpu_info.cpu_share)", 
>>>> "cpu_info.one_min_cpuload", "UUID", "COUNT(cpu_info)"
>>> [Megh]: When you select "T" you are requesting the actual samples and so 
>>> the aggregations - SUM, COUNT will only happen on the single sample and 
>>> hence. You can select "T=" which is aggregations over a specific time 
>>> period in seconds, and then the aggregations will give different values.
>>> 
>>>>  ],
>>>>  "sort_fields": [],
>>>>  "start_time": "now-60m",
>>>>  "table": "StatTable.ComputeCpuState.cpu_info",
>>>>  "where": [
>>>>      [
>>>>          {
>>>>              "name": "Source",
>>>>              "op": 1,
>>>>              "suffix": null,
>>>>              "value": "contrail",
>>>>              "value2": null
>>>>          }
>>>>      ]
>>>>  ]
>>>> }
>>>> 
>>>> I send this query and I get around 12 instances with the information I 
>>>> requested. However, the count is 1 in each instance and the SUM is equal 
>>>> to de cpu_share (as it represents only 1 value). How can I have, for 
>>>> example, an aggregation over 3 (i.e. a count equal to 3)? Is it configured 
>>>> somewhere in contrail configuration?
>>> [Megh]: As above.
>>>> 
>>>> Also, if you could provide me which kinds of aggregations are available I 
>>>> would be grateful. Are they the same as the aggregations available at 
>>>> Cassandra?
>>> [Megh]: SUM, COUNT, MAX, MIN are the aggregations available. These are 
>>> implemented on top of Cassandra and Cassandra might not offer the safe 
>>> aggregations natively.
>>> 
>>> Thanks
>>> 
>>> Megh
>>>> 
>>>> Thank you for reading.
>>>> 
>>>> Best regards,
>>>> Alberto.
>>>> 
>>>> _______________________________________________
>>>> Users mailing list
>>>> [email protected]
>>>> http://lists.opencontrail.org/mailman/listinfo/users_lists.opencontrail.org
>>> 
>>> 
>>> _______________________________________________
>>> Users mailing list
>>> [email protected]
>>> http://lists.opencontrail.org/mailman/listinfo/users_lists.opencontrail.org
>> 
>> 
> 
> 

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
Users mailing list
[email protected]
http://lists.opencontrail.org/mailman/listinfo/users_lists.opencontrail.org

Reply via email to