BSBurp Suite · Job board
Community homeHot discussions Latest posts Best of Knowledge base Resources About

Burp Suite · Job board · burpsuite-jobs under load on an 8-core 16GB box — full benchmark scripts included

2.1K
BSr/burpsuite-jobs·posted by tang_hao·2 days agoDiscussionLocked

burpsuite-jobs under load on an 8-core 16GB box — full benchmark scripts included

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

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 burpsuite-jobs itself but our upstream connection reuse — the load test traffic was too clean and hid the long-tail requests.

Performance44%
Maintainability28%
Ecosystem and community17%
Hiring difficulty11%

3577 votes total

246 comments

246 comments

· first 120 loaded
M
Ddev_zhou·just now

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

470
NnikicOP·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.

425
Kkernel_panic·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.

19
Ttang_hao·just now

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.

455
Llinlin·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.

448
Aalice_dev·2 days ago

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

448
Aalice_dev·2 days ago

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

427
Llinlin·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.

426
Hhuang_ke·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.

441
Sslow_query·2 days agoedited

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

318
Oops_wang·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.

200
Hhuang_ke·2 days ago

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

460
Rrase·12 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.

1
Hhuang_ke·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.

35
Hhuang_ke·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.

37
Lli_ming·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
Sswoole_lee·2 days agoeditedLevel 6

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.

55
Aalice_dev·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.

18
Sswoole_lee·2 hours 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.

35
Nnikic·2 days ago

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

412
Rran_bo·2 days ago

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

411
Zzhu_zong·2 days 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.

75
Nnikic·2 days ago

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

59
Oops_wangMod·2 days ago

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

368
Kkite·12 minutes agoedited

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.

343
Sswoole_lee·5 hours ago

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

392
NnikicOPMod·3 minutes ago

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

235
Aalice_dev·1 hour 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.

135
Zzhou_yi·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.

179
Lli_mingOP·1 hour ago

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

24
Sslow_query·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.

195
Hhuang_ke·2 days ago

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

367
Sslow_query·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.

57
Rran_bo·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.

5
Nnikic·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.

1
Oops_wang·5 hours agoedited

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

1
Kkernel_panic·2 days agoedited

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

4
Mmike_xu·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.

8
Bbob_chen·3 minutes ago

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

340
Wwinter·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.

325
Sswoole_lee·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.

324
Bbob_chen·3 minutes ago

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

16
Aalice_dev·just nowedited

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.

320
Zzhou_yi·just nowedited

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

73
Llinlin·3 minutes ago

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

103
Kkernel_panic·2 days agoedited

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.

132
Llinlin·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.

15
Sslow_queryOP·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.

275
Sswoole_lee·2 days ago

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

11
Sswoole_lee·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.

42
Sswoole_lee·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.

459
Wwinter·2 days ago

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

14
Oops_wangOP·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.

255
Llinlin·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.

312
Bbob_chen·2 days ago

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

311
Nnikic·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.

307
Kkernel_panic·2 days ago

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

288
Sslow_query·2 days ago

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

286
Sswoole_leeOP·2 days ago

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

247
Nnikic·3 minutes agoedited

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

454
Bbob_chen·2 days ago

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

66
Kkite·12 minutes 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.

60
Ddev_zhou·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.

270
WwinterOPMod·2 days ago

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

208
Ttang_hao·2 days ago

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

250
Oops_wang·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.

489
Kkite·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.

246
Kkite·2 days ago

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

220
Bbob_chen·2 days ago

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

16
Zzhu_zong·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.

214
Ttang_hao·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.

12
Lli_ming·28 minutes ago

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

208
Ttang_hao·2 days ago

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

184
Hhuang_ke·2 days ago

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

181
Aalice_dev·12 minutes ago

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

177
Rrase·2 days ago

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

156
Wwinter·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.

133
Bbob_chenOP·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.

71
Zzhou_yi·2 days ago

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

111
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.

94
Wwinter·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.

4
Ddev_zhou·12 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.

91
Lli_mingMod·2 days ago

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

91
Cchen_dev·2 days ago

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

89
Sswoole_lee·5 hours 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.

73
Cchen_dev·yesterday

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

70
Rran_bo·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.

48
Sslow_query·28 minutes ago

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

1
Zzhou_yi·2 days ago

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

48
Nnikic·2 days ago

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

37
Zzhu_zong·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.

36
Zzhou_yi·2 days agoedited

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

34
Hhuang_ke·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.

31
Mmike_xu·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.

123
Hhuang_ke·2 days ago

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

417
Zzhou_yi·3 minutes ago

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

12
Ttang_hao·2 days ago

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

77
Lli_ming·2 days ago

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

15
Mmike_xu·2 days ago

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

19
Kkite·2 days ago

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

2
Nnikic·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.

118
Zzhu_zong·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.

49
Kkernel_panicMod·2 days agoedited

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

17
Nnikic·2 days ago

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

16
Zzhou_yi·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.

15
Ddev_zhou·1 hour ago

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

8
Bbob_chen·2 days ago

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

17
Llinlin·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.

364
Rrase·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.

88
Lli_ming·12 minutes agoedited

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.

8
Sslow_queryMod·2 days 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.

7
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.

5
Ttang_hao·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.

3
Ttang_hao·28 minutes ago

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

3
Nnikic·just now

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

118
Hhuang_ke·2 days agoedited

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

2
Sswoole_lee·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.

1
Wwinter·2 days ago

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

505
Rran_boOP·1 hour ago

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

87
Llinlin·2 days ago

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

1

This is the post detail page /en/c/burpsuite-jobs/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 →