-
Notifications
You must be signed in to change notification settings - Fork 160
Troubleshooting
Typically, Dagda works fine, but some scenarios can cause problems. So here are some quick solutions to help get back on track.
When Syslog is your Docker Logging Driver, the error you get might be like this:
<2017-08-29 11:40:11,966> <ERROR> <DagdaServer> <docker_driver> <Docker logging driver is not set to be 'json-file' or 'journald'>
The Docker-py .logs() function raises this error if your dockerd log driver was set to syslog.
For fixing this issue, you must set your dockerd log driver like --log-driver=json-file or like --log-driver=journald.
Dagda is integrated with Falco for monitoring runtime docker containers to detect anomalous activities, so if you have not installed the kernel headers in the host OS, the error you get in logs might be like this:
<2017-01-13 17:06:21,072> <ERROR> <DagdaServer> <dagda_server> <...>
<2017-01-13 17:06:21,073> <WARNING> <DagdaServer> <dagda_server> <Runtime behaviour monitor disabled.>
For fixing this issue, the kernel headers must be installed in the host operating system, before running Dagda.
This can usually be done on Debian-like distributions with: apt-get -y install linux-headers-$(uname -r)
Or, on RHEL-like distributions: yum -y install kernel-devel-$(uname -r)
After that, run the command /usr/lib/dkms/dkms_autoinstaller start is recommended for avoiding the next Sysdig Falco error trace:
rmmod: ERROR: Module sysdig_probe is not currently loaded
Important to note: In some distributions it has been detected that the previous error persits even though the previous steps have been done. In this case, you must install Sysdig manually for fixing it, so if you need instructions for Sysdig installation, see the How-to install Sysdig for Linux page.