Model create error (invalid time delta value)

[el-tanito]

Hello, I am having an error (invalid time delta value) while creating a baseline.

Am I doing something wrong?

[vsergeyev]

Hello, It may be a bug.

Could you please also send a screenshot of panel settings. E.g. metric, group by, fill.

Thank you for reporting this issue!

V.

21 лип. 2020 р. о 14:03 el-tanito notifications@github.com пише:

 Hello, I am having an error (invalid time delta value) while creating a baseline.

Am I doing something wrong?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[el-tanito]

@vsergeyev, below are the configuration details.

If I change the configuration parameters (E.g., group by, fill) another error message appears.

[vsergeyev]

Thank you,

It looks like a bug. I’ll look for a way to fix it.

As a workaround - you may choose some value in seconds/minutes in GROUP BY time()

This will force it to use selected value. Ex. - 10s or so.

V.

21 лип. 2020 р. о 19:39 el-tanito notifications@github.com пише:

 @vsergeyev, below are the configuration details.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[el-tanito]

@vsergeyev , I tried to force the group by time variable, but it doesn't work.

Inside the configuration file (loudm.yml) I have defined the two buckets. On grafana I then selected database name. Is it correct?

loudml.yml

 buckets:

 # Input bucket
 - name: welding
   type: influxdb
   addr: 127.0.0.1:8086
   database: Marelli
   retention_policy: autogen
   measurement: loudml

 # Output bucket
 - name: loudml
   type: influxdb
   addr: 127.0.0.1:8086
   database: loudml_db
   retention_policy: autogen
   measurement: loudml
   annotation_db: loudmlannotations

storage:
  path: /var/lib/loudml

server:
  listen: 127.0.0.1:8077

docker-compose.yml

This is the mapping between host and container within the LoudML service

 volumes:
  - loudml-storage:/var/lib/loudml
  - ./loudml/loudml.yml:/etc/loudml/loudml.yml:ro

[vsergeyev]

Hello @el-tanito ,

Thank you for giving it a try.

For Input bucket it should be used "welding" For Output bucket it should be used "loudml"

Docker settings looks good for me.

V.

[el-tanito]

Hi @vsergeyev , By changing some configuration parameters, I managed to generate the baseline.

Unfortunately, when I perform the "refresh models and jobs from Loud ML server", the job fails because it does not find the bucket (name = 'Marelli'). As shown in the picture below :

Here are the configuration parameters used:

• grafana
• loudml.yml

loudml.yml

 # Output bucket

 - name: loudml_db
   type: influxdb
   addr: http://influxdb:8086
   database: loudml_db
   retention_policy: autogen
   measurement: "current_output_analysis"
   annotation_db: loudmlannotations

 # Input bucket

 - name: Marelli
   type: influxdb
   addr: http://influxdb:8086
   database: Marelli
   retention_policy: autogen
   measurement: "current_output"

What could be the problem?