ESESP32 · Code review
Community homeHot discussions Latest posts Best of Knowledge base Resources About

ESP32 · Code review · Postmortem: how we exhausted the connection pool in esp32-review

61
ESr/esp32-review·posted by alice_dev·5 minutes agoOpen source

Postmortem: how we exhausted the connection pool in esp32-review

It took me two weeks of on-and-off digging and plenty of wrong turns. Writing the process down as it happened so the next person spends less time.

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.

What genuinely surprised me was the tail. The average looked great while P99 jumped by an order of magnitude past some threshold. The cause was not esp32-review itself but our upstream connection reuse — the load test traffic was too clean and hid the long-tail requests.

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.

24 comments

24 comments

M
Lli_ming·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.

471
Hhuang_ke·12 minutes ago

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

444
Wwinter·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.

409
Rran_bo·2 days ago

One counter-example: below esp32-review 7.4 the semantics of that code are different, so do not copy it verbatim. We got burned in staging and rolled back once.

118
Sswoole_lee·3 minutes ago

One counter-example: below esp32-review 7.4 the semantics of that code are different, so do not copy it verbatim. We got burned in staging and rolled back once.

357
Nnikic·28 minutes agoedited

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

155
Kkernel_panicOP·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.

388
Ttang_hao·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.

400
Llinlin·12 minutes agoedited

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.

16
Kkernel_panic·28 minutes agoedited

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.

336
Ttang_hao·28 minutes ago

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

52
Zzhou_yi·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.

73
Sswoole_lee·2 days 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.

56
Zzhu_zong·28 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.

25
Rrase·1 hour ago

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

152
Kkite·12 minutes ago

This is not a esp32-review problem, it is a usage problem. The docs say this API is not thread-safe and you must lock around it yourself.

16
Kkite·28 minutes ago

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

63
Lli_ming·2 days ago

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

195
Lli_ming·2 days 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.

47
Ddev_zhouMod·1 hour 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.

215
Rrase·2 days ago

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

20
Oops_wang·5 hours 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.

13
Llinlin·2 days ago

This is not a esp32-review problem, it is a usage problem. The docs say this API is not thread-safe and you must lock around it yourself.

3
Nnikic·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.

1

This is the post detail page /en/c/esp32-review/post/p2. 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 →