{"id":10482,"date":"2026-09-21T07:20:00","date_gmt":"2026-09-21T05:20:00","guid":{"rendered":"https:\/\/www.lukaswojcik.com\/blog\/?p=10482"},"modified":"2026-10-26T13:45:35","modified_gmt":"2026-10-26T12:45:35","slug":"why-prometheus-uses-too-much-memory-on-a-raspberry-pi-finding-excessive-time-series","status":"publish","type":"post","link":"https:\/\/www.lukaswojcik.com\/blog\/en\/raspberry-pi\/why-prometheus-uses-too-much-memory-on-a-raspberry-pi-finding-excessive-time-series\/","title":{"rendered":"Why Prometheus Uses Too Much Memory on a Raspberry Pi: Finding Excessive Time Series"},"content":{"rendered":"

The number of exporters alone is insufficient for sizing a Raspberry Pi running Prometheus. Time-series count, scrape frequency, queries and retention all matter. Each time series is identified by its metric name and the complete combination of label values.<\/p>\n

The number of distinct combinations is called cardinality. The product of the label value counts is an upper bound, not an automatically reached total: some combinations may never occur. Adding a label with one fixed value does not multiply the count. A label with many changing values can nevertheless impose substantial costs.<\/p>\n

\n\"Example\n
Example upper bound: 3 \u00d7 5 \u00d7 8 \u00d7 40 \u00d7 2,400 = 11,520,000 combinations. Only combinations actually exposed create series.<\/figcaption>\n<\/figure>\n

Value counts and turnover need separate attention<\/h2>\n

A label needs a controlled value set and a purpose. For method<\/code> and status<\/code>, actual counts depend on the application and instrumentation. HTTP specifies neither exactly five methods nor exactly twelve status values. Unknown methods can, for example, map to a common fallback value.<\/p>\n

Raw paths such as \/wp-admin<\/code>, \/.env<\/code> or \/product\/12345<\/code> can create additional series if instrumentation copies every path into a label. Another request with the same complete label combination creates no additional series. Stored data are also subject to retention rather than remaining forever.<\/p>\n
Label<\/th>Typical source of values<\/th>Cardinality control<\/th><\/tr><\/thead>
job \/ instance<\/td>Targets and configuration<\/td>Inventory and target count<\/td><\/tr>
method \/ status<\/td>Application and protocol<\/td>Allowed values and fallback<\/td><\/tr>
handler \/ route<\/td>Route templates<\/td>Fixed route set; unknown fallback<\/td><\/tr>
path (raw)<\/td>Request paths<\/td>Normalization at the source<\/td><\/tr>
user_id \/ email<\/td>Individual identifiers<\/td>Omission from aggregate metrics<\/td><\/tr>
container_id \/ pod<\/td>Runtime instances<\/td>Current count and turnover<\/td><\/tr><\/tbody><\/table>\n

Churn means series appearing and disappearing, for example as pod identifiers change. If ten old series disappear and ten new series appear, the total stays the same despite twenty changes. A constant number of simultaneously present series can hide high churn. Old series remain relevant in the affected blocks during retention; with a stable turnover rate and bounded retention, the total does not have to keep growing every week.<\/p>\n

Assessing resources without a fixed per-series allowance<\/h2>\n

Many series consume memory, CPU, storage and network capacity. Current series data and indexes, processing the write-ahead log (WAL) used for recovery, and large queries all contribute. A fixed kilobyte allowance per series is not a reliable capacity guarantee: label sizes, samples, versions and query loads vary. RAM, CPU or storage can become the first constraint on a Pi.<\/p>\n

Memory exhaustion can cause an out-of-memory (OOM) termination and a restart loop. This is a possible failure sequence, not a diagnosis established by a broken dashboard. Process memory, host headroom, restarts, logs and changes in series count need examination together.<\/p>\n

Targeted queries for unusual series counts<\/h2>\n

The following PromQL examples initially limit the investigation to the example job app<\/code>. An unrestricted scan across all metrics can add load to an already struggling Pi. The second query counts only present, nonempty path<\/code> values; with no matching series, this PromQL expression returns no result rather than an automatically supplied zero.<\/p>\n

Text<\/span>