chore(security): a DoS and resource-limit review, its tests, and SECURITY.md for 0.1
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 15/100
Research direction
This is a design-level review, not a single fix. Start by reading SECURITY.md and the QUIC, HPACK and QPACK code paths the issue names, and map each resource limit (streams, header and table sizes, reassembly buffers, ACK ranges, connection IDs) to where it is enforced. Done means a written internal review, a test per DoS class run against the server, DoS tests wired into CI, and a current SECURITY.md. Confirm scope with maintainers before starting.
Written by the indexing model from the issue text.
Description
SECURITY.md says the code hasn't had an external security review. That doesn't have to block 0.1, but a network stack shipping its own QUIC, HPACK and QPACK in C should do two things.
Before the tag:
- An internal review pass over the attack surface. Resource limits per connection: streams, header and table sizes, reassembly buffers, CONTINUATION floods, QPACK blocked-stream memory, ACK range counts, and connection-ID tables. The known HTTP/2 and HTTP/3 DoS classes: rapid reset (CVE-2023-44487), CONTINUATION flood, HPACK bomb, ping and settings floods, zero-window stalls. Anti-amplification and Retry token handling.
- Tests for each DoS class, run against our server.
- The release notes and SECURITY.md say plainly what has and hasn't been reviewed, and how to report a vulnerability.
After 0.1: an external review, and an application to OSS-Fuzz once the targets from #55 exist.
Done when: the internal pass is written up, every finding is fixed or filed, the DoS tests are in CI, and SECURITY.md is current.
- Dominant language
- C
- Stars
- 0
- Forks
- 0
- Avg merge
- 20h 4m
- Merged PRs (30d)
- 25
Getting set up
- No Dockerfile or Docker Compose file
- No pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from public/cursed-http
-
P0: release blocker
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
public/cursed-http#56 ·
Maintainers usually reply within 1 day
-
P0: release blocker
Difficulty 4/5 3-5 days Newbie friendliness 35/100
public/cursed-http#60 ·
Maintainers usually reply within 1 day
-
documentation P0: release blocker
Difficulty 3/5 1-2 days Newbie friendliness 42/100
public/cursed-http#59 ·
Maintainers usually reply within 1 day
-
P0: release blocker
Difficulty 4/5 3-5 days Newbie friendliness 35/100
public/cursed-http#58 ·
Maintainers usually reply within 1 day
-
P0: release blocker
Difficulty 5/5 Over a week Newbie friendliness 15/100
public/cursed-http#57 ·
Maintainers usually reply within 1 day
All issues in public/cursed-http
Similar issues
-
triage
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
xerial/sqlite-jdbc#1515 ·
Maintainers usually reply within 1 day
-
macOS 菜单栏托盘图标保留橙色,与系统单色图标风格不一致Possibly taken @xhugoliu claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
YizhengWw/HachimoDock#56 · 1 comment ·
-
compiler-rt:asan
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
llvm/llvm-project#242921 ·
Maintainers usually reply within 1 day
-
Missing gcc __builtin_stdc_bit_width/floor/ceil signature makes intellisense wrongly report error 70Open
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
microsoft/vscode-cpptools#14874 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
FujiNetWIFI/fujinet-firmware#2401 ·
Maintainers usually reply within 1 day