Description of problem: node-agent rpm loads tendrl-specific collectd plugins in /usr/lib/collectd whereas the collectd plugin path is /usr/lib64/collectd hence the collectd shows the error: python plugin: Error importing module "tendrl_gluster". Unhandled python exception in importing module: ImportError: No module named tendrl_gluster Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Import gluster cluster into Tendrl 2. Observe grafana brick dashboard Actual results: No collectd stats in grafana. It shows "No Datapoints" Expected results: All applicable stats pushed from either collectd or monitoring-integration should be available on grafana Additional info: node_agent rpm is installing collectd plugins at path /usr/lib/collectd instead of collectd plugin path /usr/lib64/collectd which is causing issue that our cusom collectd plugins are not pushing data to graphite. This is possibly bcoz of difference in value of _libdir in rhel vs centos/fedora in https://github.com/Tendrl/node-agent/blob/master/tendrl-node-agent.spec#L47 of spec file. So we probably will need to hard code path as /usr/lib64/collectd in lines 47, 48 and 57 and 85.
This issue is fixed with latest builds
Moving into right component.
Collectd plugins from tendrl-node-agent package are in /usr/lib64/collectd path: # rpm -ql tendrl-node-agent | grep lib64/collectd /usr/lib64/collectd/__init__.py /usr/lib64/collectd/__init__.pyc /usr/lib64/collectd/__init__.pyo /usr/lib64/collectd/gluster /usr/lib64/collectd/gluster/__init__.py /usr/lib64/collectd/gluster/__init__.pyc /usr/lib64/collectd/gluster/__init__.pyo /usr/lib64/collectd/gluster/heavy_weight /usr/lib64/collectd/gluster/heavy_weight/__init__.py /usr/lib64/collectd/gluster/heavy_weight/__init__.pyc /usr/lib64/collectd/gluster/heavy_weight/__init__.pyo /usr/lib64/collectd/gluster/heavy_weight/tendrl_gluster_heal_info.py /usr/lib64/collectd/gluster/heavy_weight/tendrl_gluster_heal_info.pyc /usr/lib64/collectd/gluster/heavy_weight/tendrl_gluster_heal_info.pyo /usr/lib64/collectd/gluster/heavy_weight/tendrl_glusterfs_profile_info.py /usr/lib64/collectd/gluster/heavy_weight/tendrl_glusterfs_profile_info.pyc /usr/lib64/collectd/gluster/heavy_weight/tendrl_glusterfs_profile_info.pyo /usr/lib64/collectd/gluster/low_weight /usr/lib64/collectd/gluster/low_weight/__init__.py /usr/lib64/collectd/gluster/low_weight/__init__.pyc /usr/lib64/collectd/gluster/low_weight/__init__.pyo /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_brick_utilization.py /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_brick_utilization.pyc /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_brick_utilization.pyo /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_health_counters.py /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_health_counters.pyc /usr/lib64/collectd/gluster/low_weight/tendrl_glusterfs_health_counters.pyo /usr/lib64/collectd/gluster/tendrl_gluster.py /usr/lib64/collectd/gluster/tendrl_gluster.pyc /usr/lib64/collectd/gluster/tendrl_gluster.pyo /usr/lib64/collectd/gluster/tendrl_gluster_brick_disk_stats.py /usr/lib64/collectd/gluster/tendrl_gluster_brick_disk_stats.pyc /usr/lib64/collectd/gluster/tendrl_gluster_brick_disk_stats.pyo /usr/lib64/collectd/gluster/utils.py /usr/lib64/collectd/gluster/utils.pyc /usr/lib64/collectd/gluster/utils.pyo Also there is no Import Error in the logs as described in the description and Grafana dashboards contains the required data. # rpm -q tendrl-node-agent tendrl-node-agent-1.6.3-9.el7rhgs.noarch NOTE: it is fixed also in the previous released version tendrl-node-agent-1.5.4-16.el7rhgs.noarch. >> VERIFIED
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2018:2616