Opsview-webserver failed to start

Symptoms Copied

• opsview-webserver fails to start during the orchestrator-install.yml deploy playbook
• You may see an error like the following in the /var/log/opsview/opsview.log file:

Mar 1 16:40:12 orchestrator-hostname opsview-monit[109766]: 'opsview-webserver' failed to start (exit status 1) -- '/opt/opsview/webserver/bin/start.sh': nginx: [warn] duplicate MIME type "text/html" in /opt/opsview/webserver/etc/conf.d/opsview.conf:62#012nginx: [emerg] host not found in upstream "dataserver" in /opt/opsview/webserver/etc/conf.d/opsview.conf:230

Cause: Incorrect order of playbook tasks Copied

Known issue: OP-50927

There is a mistake in the orchestrator-install.yml deploy playbook whereby Nginx configuration for the opsview-dataserver component is added before the opsview-dataserver is installed. As a result, opsview-webserver fails to start because its Nginx configuration is incorrect.

The orchestrator-install.yml playbook is used by other playbooks including setup-opsview.yml and setup-everything.yml.

Solution: Install opsview-dataserver manually Copied

To work around this issue, you need to install the opsview-dataserver package manually then re-run the playbook that failed to start opsview-webserver.

1. Log in to the Opsview host as root.
2. Install the opsview-dataserver package:

CentOS/RHEL

yum install opsview-dataserver

Ubuntu/Debian

apt install opsview-dataserver