Fail to deploy the LMA toolchain in an environment with virt nodes
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Invalid
|
High
|
Fuel Python (Deprecated) | ||
StackLight |
Fix Released
|
Medium
|
Swann Croiset |
Bug Description
Steps to reproduce
------------------
- install the LMA plugins on the Fuel node
- Follow the instructions [1] to deploy an environment with virt nodes and enable the LMA plugins with their default configuration (eg Elasticsearch, InfluxDB and Nagios deployed locally).
Actual result
-------------
The deployment fails because there's no node with the Elasticsearch/
Expected result
---------------
The deployment should be successful.
Workaround
----------
- Enable only the LMA collector plugin in the first environment and configure it to send data to remote servers using fake names (eg elasticsearch-node, influxdb-node, alerting-node).
- Deploy the first environment with only the virt node(s).
- Provision the virtual machines as usual.
- Create a second environment using the virtual machines and enable all the LMA plugins.
- Once the second environment is deployed, edit /etc/hosts on the virt node(s) and add the proper IP addresses for elasticsearch-node, influxdb-node and alerting-node
- Restart the collectors on the virt node(s).
description: | updated |
tags: | added: area-python |
Changed in fuel: | |
milestone: | none → 9.0 |
assignee: | nobody → Fuel Python Team (fuel-python) |
importance: | Undecided → High |
status: | New → Confirmed |
Changed in lma-toolchain: | |
milestone: | none → 0.10.0 |
Changed in lma-toolchain: | |
importance: | Undecided → Medium |
summary: |
- Failt to deploy the LMA toolchain in an environment with virt nodes + Fail to deploy the LMA toolchain in an environment with virt nodes |
Changed in lma-toolchain: | |
status: | Fix Committed → Fix Released |
The deployment failed during VM provisioning because the plugins were also executed in post-deployment during VM creation.