This post assumes that you’re familiar with Metric Grouping Issues.
Metric Grouping Issues are most commonly discovered when a metric limit is exceeded and a ‘clamp’ is triggered.
When the count of unique metric names reporting to an application or account surpasses a set limit, a metric clamp is triggered. This prevents any new metrics from reporting.
There are two types of clamps:
- Individual Application ‘Whitelist’ - Limit default of 300K.
This is called a ‘whitelist’ because the metrics names which have so far reported to the application, up to the limit, are whitelisted. Any new metric names cannot be created until the clamp is lifted.
When the whitelist clamp is triggered, a support ticket is opened up in the account owner’s name and the admins are CC’d. When you engage in the ticket, New Relic Support can help in addressing the metric grouping issue and removing the whitelist clamp.
- Account ‘Blacklist’ - Limit default of 4 million - A much rarer occurrence. A ‘blacklist’ is the result of many metric grouping issues across many applications in an account. These cases are often caused by a systemic misuse or misconfiguration of transaction metric naming. Often there are multiple applications with the same or similar code bases that propagate improper metric naming.
This is part 2 of a series of posts on understanding and addressing MGIs, watch out for more coming soon.