Management of FIWARE Lab nodes help-desk tickets SLA resolution timne.
- Introduction
- Overall description
- Build and Install
- Running
- Access to the historical information in Monasca
- Deployment
- Testing
- Support
- License
Python script to calculate the percentage of Help-Desk tickets responsed and resolved per each FIWARE Lab node in less than 24 working hours and less than 48 working hours.
The procedure to calculate this percentages is taking all the tickets resolved since 1st January 2017 till now in Jira under the project Help-Desk and identified with the component FIWARE-LAB-HELP. This component corresponds to the tickets that were created from the corresponding mail list fiware-lab-help@lists.fiware.org.
Once that we have all the tickets, we categorized them into resolved and responded and calculate the time in which they were resolved and responded. A ticket is responded when the status of the ticket is moved from 'Open' to 'In Progress' or 'Answered'. A ticket is resolved when it is moved from 'Close' status. The tickets that are dismissed from 'Open' status are also taking into account in terms of calculate the progress timne of them.
Next step, once that we have the progressed time is calculate the different time. The process is calculating the difference in working hours. It means that every day has only 9 hours (from 8:00 to 17:00) to calculate the difference and weekends are not included. For example, if a ticket was created on Friday 16/02/2018 at 9:35:24,43 and was resolved on Monday 19/02/2018 at 8:47:43,12 the resolution time will be difference time on friday until end of working day plus the difference time from beginning of working day until the resolution time:
16/02/2018: from 09:35:24,43 to 17:00:00,00 = 07:24:35,57
19/02/2018: from 08:00:00,00 to 08:47:43,12 = 00:47:43,12
TOTAL TIME: 07:24:35,57 + 00:47:43,12 = 08:12:18,69
Therefore the resolution time in that case was 8 hours, 12 minutes and 18,69 seconds. If the ticket was created before 08:00:00, the starting point is considered 08:00:00. If the ticket is closed after 17:00:00 it is considered closed at 17:00:00. In case of tickets created during weekends, the created time is the next Monday at 08:00:00. In case of tickets closed during weekends, the closed time is the previous Friday at 17:00:00.
In this process is not considered bank holidays for each region.
Once that we have the response and resolution times we send the information to the OpenStack Monasca instance in order to keep centralized the monitoring information and make afterwards if could be neccesary some type of statistical analysis. For this purpose we store also the number of tickets for each region. Keep in mind, that it is needed a request to OpenStack Keystone service in order to recover a proper token in order to send the information to the Monasca API. In that case we are using the tenant name service due to the Monasca instance is configured to pass the monitoring information under this tenant.
The following software must be installed:
- Python 2.7
- pip
- virtualenv
The recommend installation method is using a virtualenv. Actually, the installation process is only about the python dependencies, because the python code do not need installation.
- Clone this repository.
- Define the configuration file in './config/fiware-sla.ini'
- Execute the script 'source ./deploy/config.sh'.
- With root user, execute the command 'cp ./config/fiware-sla.logrotate /etc/logrotate.d/fiware-sla'.
Once that you have installed and configured your application, you can run it just executing:
python SLAMeassurement.py
And it will be executed in order to calculate the corresponding SLA levels for all the nodes and send the information to the configured Monasca interface. Keep in mind that you have to be inside a previously defined virtual environment.
The config.sh file that you can find in the deploy folder is used in order to allow
the automatic execution of the python script just adding the corresponding header to the file
SLAMeassurement.py:
#!/usr/bin/env /env/bin/python
Where env is the name of your virtual environment.
Last but not least, the service is added into the crontab of the machine in order to execute the service every day at 02:00:00.
The service is configured in order to send the data to the FIWARE OpenStack Monasca service in order to keep a historical information about the resolution of the tickets. You can check those meassurements directly over the Monasca API but previously it is required to obtain a secure token associated to the ceilometer service requesting it to the Keystone instance:
curl -X POST http://cloud.lab.fiware.org:4730/v2.0/tokens \
-H 'Accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
"auth": {
"tenantName": "service",
"passwordCredentials": {
"username": "<ceilometer service user, one per region>",
"password": "<ceilometer service user password>"
}
}
}' | jq .access.token.idIf you do not have installed the jq tool, please download it in order to navigate inside the json response and obtain the proper token id in a easy way.
Now, it is time to remember how is managed the metrics inside Monasca. In order to send the data we have defined two dimensions in Monasca:
-
metric: a metric name to filter metrics by. By default we have defined the metrics region.ticket_resolve_time and region.ticket_response_time for the percentage of resolve and response of the Help-Desk associated to each region. You can obtain this data also requesting to Monasca API:
curl -X GET http://monasca.lab.fiware.org:8070/v2.0/metrics/names \ -H 'Cache-Control: no-cache' \ -H 'X-Auth-Token: <Ceilometer service token>' | jq .elements[].name
-
source: the application that provide this SLA data, in our case it is fixed to fiware-sla
-
region: this is the region name in which we calculate the values of SLA. You can request the list of available regions directly to Monasca through the execution of the following query:
curl -X GET 'http://monasca.lab.fiware.org:8070/v2.0/metrics?name=region.ticket_resolve_time' \ -H 'Cache-Control: no-cache' \ -H 'X-Auth-Token: <Ceilometer service token>' | jq .elements[].dimensions.region
Now to request the meassurements associated to the ticket resolve time in the Spain region starting at 15/08/2018 just execute the following query:
curl -X GET 'http://monasca.lab.fiware.org:8070/v2.0/metrics/measurements?name=region.ticket_resolve_time&start_time=2018-08-15T00:00:01Z&dimensions=region:Spain' \
-H 'Cache-Control: no-cache' \
-H 'X-Auth-Token: <Ceilometer service token>' | jqIf you want to get more details about the use of OpenStack Monsca API, please take a look to the official documentation about it in monasca-api.
There is a specific option to deploy this service in a host. Take a look to the content of deploy directory
It was defined a minimum set of tests to cover the core functionality of the
service. The tests target is used for running the unit tests in the
component. We use for those tests the tox tool.
Tox is a generic
virtualenv management and test
command line tool you can use for checking your package installs correctly
with different Python versions and interpreters running your tests in each
of the environments, configuring your test tool of choice acting as a
frontend to Continuous Integration servers, greatly reducing boilerplate and
merging CI and shell-based testing.
The configuration file can be found in tox.ini in which we have defined two
different environments:
- The first one, to test the service using nosetests.
- The second one, to check the python coding style using pycodestyle
First of all, you need to install the tool with the following commands:
pip install tox
Now, you can run the tests, simply execute the commands:
tox
The support of this service is under github. You can create your issues and they will be resolved by the development team in the following sprint.
(c) 2018 FIWARE Foundation, e.V., Apache License 2.0