ALAlertmanager · English
Community homeHot discussions Latest posts Best of Knowledge base Resources About

Alertmanager · English · Why this alertmanager-en pattern gets slow in production: a source-level explanation

14
ALr/alertmanager-en·posted by zhu_zong·yesterdayExperience

Why this alertmanager-en pattern gets slow in production: a source-level explanation

Most alertmanager-en articles stop at "how to use it" and never cover "when not to use it". This is an attempt at the second half.

Worth noting: the official docs do cover this, just in a very inconspicuous spot. I only found it reading the source comments, where the author explains the reasoning — roughly "so that it degrades into predictable behaviour in extreme cases".

One last trap: in container environments remember to adjust the memory-related parameters in step. Otherwise the host limit and the process expectation disagree, and the symptom is intermittent, unreproducible failure.

Jjuejin.cnExternal link · opens in a new tab
8 comments

8 comments

M
Bbob_chen·yesterday

Saved. I am reworking this area this week — this saves a lot of wrong turns.

448
Sswoole_lee·12 minutes ago

One counter-example: below alertmanager-en 7.4 the semantics of that code are different, so do not copy it verbatim. We got burned in staging and rolled back once.

413
Aalice_dev·just now

We have run this in production for two years without hitting it. That said, we never reached this scale, so our experience is not really evidence here.

13
Llinlin·28 minutes ago

I see point 3 differently. The trade-off depends on your read/write ratio: read-heavy with little writing means caching actually widens the inconsistency window.

1
Aalice_dev·12 minutes ago

There is actually a simpler fix that needs no architecture change: move this check up to the gateway and the problem disappears. The cost is one extra lookup at the gateway.

354
RraseOP·3 minutes ago

Sharing our numbers, 8 cores 16GB, same scenario:

| Concurrency | P50 | P99 |
|---|---|---|
| 200 | 12ms | 88ms |
| 500 | 31ms | 340ms |

P99 clearly collapses at 500 concurrency, which lines up with your knee point.

4
Aalice_dev·1 hour ago

This matches what we see in production. We only hit it past 3k QPS; the earlier load tests showed nothing — the test traffic was too clean, with no long-tail requests.

239
Oops_wang·5 hours ago

This is not a alertmanager-en problem, it is a usage problem. The docs say this API is not thread-safe and you must lock around it yourself.

7

This is the post detail page /en/c/alertmanager-en/post/p12. Posts and comments are generated deterministically from a seeded PRNG, so the same post always renders the same content and the link can be shared, reloaded and indexed. In production this page reads MySQL for the post, Redis for hot-post caching, and fetches the whole comment tree in a single query on the path column.

See the database schema →