With apologies to Arrested Development fans out there, the Google Cloud Platform community has amassed a whole host of posts about logs recently — enough for a Bob Loblaw log blog.
Maybe you don’t think logging is the most exciting feature, but that doesn’t make it any less important. That’s why Cloud Platform’s integrated monitoring suite, Google Stackdriver, includes logging as part of the base package. Check out Google Developer Advocate Aja Hammerly’s explanation of what Stackdriver Logging can do, how to set it up and what to do when it’s no longer enough (hint: export the logs into BigQuery).
Stackdriver Logging is turned on by default for applications running on Google App Engine, but it can also be used with other apps and services. For example, former Googler Joe Beda uses Stackdriver Logging to augment Docker logs — to prevent them from filling up a disk, and from being deleted along with their associated container. Likewise, John Hamminck, chief evangelist at Treasure Data, tackles streaming logging with Fluentd, Kubernetes and GCP. And then there’s Ido Shamun, co-founder for @ElegantMonkeys, who streams the logs from his hapi.js stack running on App Engine to GCP Logging. “Now all the logs are fully structured in GCP Logging and I can create custom metrics and get insights out of my logs,” he writes.
Logs are also a crucial debugging tool. For the uninitiated, Romin Imani offers this tutorial on using Stackdriver Monitoring to debug a running application using the Debugger and Trace features, analyzing application logs and adding log points to an application. Oh, and about those log points…. You can now add log points on the fly, to a running application, with no need to restart.
To quote Buster, “This party is going to be Off. The. Hook.”