Error response from daemon: Conflict. The container name "/data-registry-cron" is already in use by container "28ebca6bd35a4222b6a7343262f765d14f2dddad5aff7f6c6c64b4ea9c0d6676". You have to remove (or rename) that container to be able to reuse that name.
The referenced container no longer existed by the time I looked into this.
This name is used to run the manageprocess command every 5 minutes, as configured in salt/registry/init.sls. That command typically returns quickly. My assumption is it took longer than 5 minutes, causing this error.
The error hasn't reoccurred again. If it does, we can increase the log level to see how long the command runs for, to confirm whether that's the cause. Note that any output causes cron to send us an email.
Just a note that it does occur from time to time. I think it occurs when the server is under high load. As part of #482, I resolved some performance issues. So, I'll close this as well.
The referenced container no longer existed by the time I looked into this.
This name is used to run the manageprocess command every 5 minutes, as configured in salt/registry/init.sls. That command typically returns quickly. My assumption is it took longer than 5 minutes, causing this error.
The error hasn't reoccurred again. If it does, we can increase the log level to see how long the command runs for, to confirm whether that's the cause. Note that any output causes cron to send us an email.