Skip to content

Log Storage

Logs and metrics accumulate — this page is where you see how much room they’re taking up and decide how long to keep them. It covers your current usage against quota, a burn-down trend, and the retention settings that control how old data rolls off.

Open Configure → Logs and Events and select the Log Storage tab (shown as Storage on the page itself).

At the top, GridNMS shows current usage against your quota, split into two tiers:

  • Hot — logs and metrics on fast, immediately searchable storage. This is what Log Search and your dashboards query day to day.
  • Cold — older data that’s aged out of the hot tier per your retention settings below, kept for longer but no longer on fast storage.

Each bar shows usage as a percentage of quota, and turns into a clear over-quota message if you’ve exceeded it. The quota itself depends on your deployment: a Cloud plan’s included allowance, an amount an administrator has explicitly configured, or — for a self-hosted instance with nothing configured — the data disk’s own capacity. See Subscription & Billing for what happens on GridNMS Cloud when you go over.

Once there’s enough history, a Storage trend chart shows up to 90 days of hot and cold usage, with a reference line marking your quota. Below it, GridNMS gives you a plain-language read on where you’re headed — for example, “at the recent rate, hot storage is projected to reach quota around [date]” — or tells you there isn’t enough history yet, or that usage is flat or shrinking.

Where you have more than one, a Breakdown by site & collector table shows hot and cold usage per site and per collector. This is visibility only — there’s no separate cap per site or collector, just the one quota above.

On a self-hosted deployment, a Storage quota control lets you cap total logs-plus-metrics storage in GB. Set it to 0 to auto-size the quota to the data disk’s own capacity instead of a fixed number.

The Retention section controls how long data stays before it’s moved and eventually removed. Data rolls off by age, or by a size cap, whichever is hit first.

Field What it controls
Hot (days) How long data is kept on fast storage.
Move to cold (days) When it moves from hot to the cold storage tier.
Delete after (days) When it’s deleted for good. Set to 0 for never.
Size cap (GB) An overall cap on kept storage. Set to 0 for no cap; if you hit it, the oldest data is evicted first regardless of age.

Click Save retention to apply your changes.

Turn on Archive to your own cold storage to have GridNMS copy cold data to a bucket you own — Amazon S3 or any S3-compatible service — before it’s deleted. Your own copy is independent of GridNMS’s retention: nothing about archiving changes when GridNMS deletes its own copy, which is still controlled entirely by the settings above. Note that data leaving GridNMS’s cloud this way is billed as egress.

Configure the destination:

  • Bucket and Prefix — where in your bucket the archive is written.
  • Endpoint — leave blank for AWS S3, or set it for an S3-compatible service.
  • Region, Access key ID, and Secret access key — credentials for your bucket. The secret key is write-only; it’s masked once saved and never shown again.
  • Archive older than (days) — only cold data (data that’s already aged past the hot tier) is copied out.
  • Format — the file format written to your bucket: Parquet, JSONEachRow(.gz), CSV, or TSV.

Saving with archiving turned on runs a live test write to your bucket — a bad bucket name, endpoint, region, or credential fails the save immediately rather than failing silently later. Once archiving is running, the card shows the last successful run time, or the last error if one occurred.

docs built 2026-09-26 · 195c6d00