We have a setup on Enonic Cloud with separate nodes for admin/editor traffic and separate nodes for visitor traffic. Our security is partially based on that no one being able to do any editorial work on the visitor traffic domain. This means that the log data for the visitor traffic node(s) is unavailable to anyone but Enonic employees.
On rare occasions it would be extremely useful for us to be able to follow the log browser for the visitor traffic node(s) live. On close to no occasion is it useful for us to follow the log browser for the admin node.
Since we very rarely use this, we would prefer not to always stream/copy the traffic log. It would generally just be a huge waste of both disk space and bandwidth.
The feature request is that we would like to use the log browser on our admin/editor node to look at the log from at least one of the traffic nodes. If we could switch between the various traffic nodes we run on, that would be even better.
Cluster wide log browsing will be part of the Self Service cloud setup. Most customer in your scenario use external log/metrics systems like Elastic, Datadog and Splunk - and we feed data to them.
That said, I guess there is a potential to rewrite the log browser app so it would be able to extract the feed from any cluster node, like a server/client approach - but not sure about the scope of this?
Yes, when the self-service thing is available, we will use that. In the meantime, if it’s a quick thing on your side to point the log browser to observe the log file on a different node than the one it’s running on, it seems a like a quick fix that will fulfill our needs.
We do use Datadog already for other and wide-ranging purposes, but as part of thinking sustainability, we’d rather not keep a copy of log files that we very rarely will look at or otherwise have any use for.
Hi Bjørn.
As expected this is not as simple as it may sound. The solution is log shipping as initially proposed. We will also offer access to clusterlogs from the new cloud console when this is ready.