Deploy of Influxdb node fails on slow env

[Ilya Shakhat]

Fuel deployment failed with: "Deployment has failed. Method granular_deploy. Failed to execute hook 'lma_collector-0.9.0' Puppet run failed. Check puppet logs for details". Puppet logs from influxdb node contain the following error: "(/Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns) change from notrun to 0 failed: /usr/local/bin/configure_influxdb.sh returned 7 instead of one of [0]"

The issue occurred on virtual environment (Fuel on KVM). It appears that while influxdb service is running it did not start listening on port 8086 in time. According to logs:

2016-01-12 21:47:14 +0000 Puppet (debug): Executing '/usr/sbin/service influxdb start'
2016-01-12 21:47:14 +0000 /Stage[main]/Influxdb::Service/Service[influxdb]/ensure (notice): ensure changed 'stopped' to 'running'
.....
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script] (info): Starting to evaluate the resource
2016-01-12 21:47:14 +0000 Exec[configure_influxdb_script](provider=posix) (debug): Executing '/usr/local/bin/configure_influxdb.sh'
2016-01-12 21:47:14 +0000 Puppet (debug): Executing '/usr/local/bin/configure_influxdb.sh'
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns (notice): + INFLUXDB_URL=http://127.0.0.1:8086
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns (notice): + curl -G http://127.0.0.1:8086/query --data-urlencode 'q=CREATE USER root WITH PASSWORD '\''admin'\'' WITH ALL PRIVILEGES'
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns (notice): % Total % Received % Xferd Average Speed Time Time Time Current
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns (notice): Dload Upload Total Spent Left Speed
2016-01-12 21:47:14 +0000 /Stage[main]/Lma_monitoring_analytics::Influxdb/Exec[configure_influxdb_script]/returns (notice): ^M 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0curl: (7) Failed to connect to 127.0.0.1 port 8086: Connection refused
2016-01-12 21:47:14 +0000 Puppet (err): /usr/local/bin/configure_influxdb.sh returned 7 instead of one of [0]

[guillaume thouvenin (guillaume-thouvenin)]

We merged new code that separate the creation of the InfluxDB database and its configuration. This new modifications should solve the issue you are seeing. Can you give it a try?

https://github.com/openstack/fuel-plugin-influxdb-grafana/commit/a876bbc80e7cac63f4d18f2522bc560708196d77

[Varun (varun-utagikar)]

I faced the same issue when deploying Fuel 8 on bare metal. will try the fix and post back.