MOMonitoring · Advanced
Community homeHot discussions Latest posts Best of Knowledge base Resources About

Monitoring · Advanced · Cut our monitoring-pro build from 4 minutes to 40 seconds with these 5 changes

316
MOr/monitoring-pro·posted by rase·3 days agoTutorial

Cut our monitoring-pro build from 4 minutes to 40 seconds with these 5 changes

Some background first. Our setup is monitoring-pro plus three downstream services, seven figures of daily requests, peaking around nine in the evening.

Order of investigation, by return on effort: 1. Check downstream latency first — usually it is not your problem 2. Then pool hit rate and wait-queue length 3. Only then GC and allocation 4. Suspect the framework last

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.

We also fixed monitoring along the way: replaced average-based alerts with percentiles and split them per endpoint. False alerts dropped by about seventy percent and the on-call rotation visibly cheered up.

On trade-offs, my view is this: if nobody on the team owns this area long-term, do not introduce a second mechanism. With two coexistence you first have to work out which one is even in play when things break, and that costs far more than the performance you saved.

36 comments

36 comments

M
Cchen_dev·28 minutes ago

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.

485
Aalice_dev·12 minutes ago

Thanks for sharing real numbers — far more useful than the articles that only cover concepts.

2
Rran_boOP·28 minutes agoedited

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

84
Ddev_zhouOP·12 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.

417
Kkernel_panic·2 days ago

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

245
Aalice_dev·28 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.

231
Ttang_hao·2 days 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.

203
RraseOP·2 days ago

A question: what changes in a container with a 512Mi memory limit? That is how we run it in production.

231
Sslow_query·2 days ago

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

2
Cchen_dev·yesterday

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.

371
Rrase·2 days ago

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

430
Ttang_hao·3 minutes ago

Agreeing with the above. One addition: with this option enabled the GC count in your metrics doubles, so adjust the alert threshold at the same time or it will keep firing.

13
Sswoole_lee·2 days ago

Can you give a minimal reproduction? I ran it locally for ten minutes and could not reproduce on macOS with the latest version.

441
Rrase·12 minutes ago

A question: what changes in a container with a 512Mi memory limit? That is how we run it in production.

196
Cchen_dev·2 days 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.

1
Hhuang_ke·2 days 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.

216
Cchen_dev·5 hours ago

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.

171
Sswoole_lee·3 minutes ago

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

107
Wwinter·2 days ago

Has anyone run a controlled experiment? I did, reducing it to a single variable, and the difference was 4% — within noise. So I suspect the main cause is something else.

85
Bbob_chenMod·12 minutes agoedited

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.

67
Rran_bo·1 hour ago

Thanks for sharing real numbers — far more useful than the articles that only cover concepts.

25
Mmike_xu·2 days ago

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

393
Lli_ming·2 days 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.

447
Cchen_devMod·28 minutes ago

Worth learning from this debugging approach. We went straight at the logs and took a much longer route.

95
Kkernel_panic·just now

Can you give a minimal reproduction? I ran it locally for ten minutes and could not reproduce on macOS with the latest version.

113
Ddev_zhou·28 minutes ago

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

49
Llinlin·2 days agoedited

Agreeing with the above. One addition: with this option enabled the GC count in your metrics doubles, so adjust the alert threshold at the same time or it will keep firing.

37
Zzhu_zong·12 minutes ago

A question: what changes in a container with a 512Mi memory limit? That is how we run it in production.

28
Oops_wang·3 minutes ago

I just read the monitoring-pro source — the author actually explains the reasoning in a comment, roughly "so that it degrades into predictable behaviour in extreme cases".

6
Bbob_chenOP·2 days 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.

511
Ddev_zhou·just now

Worth learning from this debugging approach. We went straight at the logs and took a much longer route.

468
Mmike_xu·28 minutes ago

Has anyone run a controlled experiment? I did, reducing it to a single variable, and the difference was 4% — within noise. So I suspect the main cause is something else.

5
Cchen_dev·yesterday

I just read the monitoring-pro source — the author actually explains the reasoning in a comment, roughly "so that it degrades into predictable behaviour in extreme cases".

1
Aalice_dev·2 days agoedited

Worth learning from this debugging approach. We went straight at the logs and took a much longer route.

19
Lli_ming·2 days ago

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

8
Bbob_chen·2 days agoedited

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.

8

This is the post detail page /en/c/monitoring-pro/post/p3. 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 →