{"containers":{"cna":{"affected":[{"collectionURL":"https://repo.hex.pm","cpes":["cpe:2.3:a:benoitc:hackney:*:*:*:*:*:*:*:*"],"defaultStatus":"unaffected","modules":["hackney_conn","hackney"],"packageName":"hackney","packageURL":"pkg:hex/hackney","product":"hackney","programFiles":["src/hackney_conn.erl","src/hackney.erl"],"programRoutines":[{"name":"hackney_conn:read_full_body/2"},{"name":"hackney_conn:maybe_decompress_body/2"},{"name":"hackney_conn:decompress_gzip/1"},{"name":"hackney_conn:decompress_deflate/1"},{"name":"hackney:request/5"},{"name":"hackney:body/1"}],"repo":"https://github.com/benoitc/hackney","vendor":"benoitc","versions":[{"lessThan":"4.9.0","status":"affected","version":"2.0.0-beta.1","versionType":"semver"}]},{"collectionURL":"https://github.com","cpes":["cpe:2.3:a:benoitc:hackney:*:*:*:*:*:*:*:*"],"defaultStatus":"unaffected","modules":["hackney_conn","hackney"],"packageName":"benoitc/hackney","packageURL":"pkg:github/benoitc/hackney","product":"hackney","programFiles":["src/hackney_conn.erl","src/hackney.erl"],"programRoutines":[{"name":"hackney_conn:read_full_body/2"},{"name":"hackney_conn:maybe_decompress_body/2"},{"name":"hackney_conn:decompress_gzip/1"},{"name":"hackney_conn:decompress_deflate/1"},{"name":"hackney:request/5"},{"name":"hackney:body/1"}],"repo":"https://github.com/benoitc/hackney","vendor":"benoitc","versions":[{"lessThan":"493e8bf83a391548e5101792276ab3fc428fa76c","status":"affected","version":"dc41b79339ca0c41e16d3f2dd17172bc2de855d6","versionType":"git"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:a:benoitc:hackney:*:*:*:*:*:*:*:*","versionEndExcluding":"4.9.0","versionStartIncluding":"2.0.0-beta.1","vulnerable":true}],"negate":false,"operator":"OR"}],"operator":"AND"}],"credits":[{"lang":"en","type":"finder","value":"Martin Karrer"},{"lang":"en","type":"finder","value":"Quinn Wilton"},{"lang":"en","type":"reporter","value":"Quinn Wilton"},{"lang":"en","type":"remediation developer","value":"Martin Karrer"},{"lang":"en","type":"remediation developer","value":"Benoit Chesneau"},{"lang":"en","type":"coordinator","value":"Jonatan Männchen / EEF"}],"dateAssigned":"2026-10-26T09:42:31.000Z","datePublic":"2026-10-25T00:00:00.000Z","descriptions":[{"lang":"en","supportingMedia":[{"base64":false,"type":"text/html","value":"

Allocation of Resources Without Limits or Throttling vulnerability in hackney_conn in benoitc hackney allows a malicious HTTP server to exhaust client memory via an oversized or highly compressed response body. hackney:request/5 and hackney:body/1 return the whole response body, and nothing bounds how much the connection process buffers. Any server the client requests can trigger it.

\n

hackney_conn:read_full_body/2 and the HTTP/2 and HTTP/3 accumulators append every chunk until the body ends. With {auto_decompress, true} the buffered body is then inflated in one step with zlib:gunzip/1 or zlib:inflate/2, which return the whole output at once.

\n

This issue affects hackney: from 2.0.0-beta.1 before 4.9.0.

"},{"base64":false,"type":"text/markdown","value":"Allocation of Resources Without Limits or Throttling vulnerability in `hackney_conn` in benoitc hackney allows a malicious HTTP server to exhaust client memory via an oversized or highly compressed response body. `hackney:request/5` and `hackney:body/1` return the whole response body, and nothing bounds how much the connection process buffers. Any server the client requests can trigger it.\n\n`hackney_conn:read_full_body/2` and the HTTP/2 and HTTP/3 accumulators append every chunk until the body ends. With `{auto_decompress, true}` the buffered body is then inflated in one step with `zlib:gunzip/1` or `zlib:inflate/2`, which return the whole output at once.\n\nThis issue affects hackney: from 2.0.0-beta.1 before 4.9.0."}],"value":"Allocation of Resources Without Limits or Throttling vulnerability in hackney_conn in benoitc hackney allows a malicious HTTP server to exhaust client memory via an oversized or highly compressed response body. hackney:request/5 and hackney:body/1 return the whole response body, and nothing bounds how much the connection process buffers. Any server the client requests can trigger it.\n\nhackney_conn:read_full_body/2 and the HTTP/2 and HTTP/3 accumulators append every chunk until the body ends. With {auto_decompress, true} the buffered body is then inflated in one step with zlib:gunzip/1 or zlib:inflate/2, which return the whole output at once.\n\nThis issue affects hackney: from 2.0.0-beta.1 before 4.9.0."}],"impacts":[{"capecId":"CAPEC-130","descriptions":[{"lang":"en","supportingMedia":[{"base64":false,"type":"text/html","value":"

A server the client requests can make the client allocate memory proportional to the body it sends, or to its decompressed size, until the node runs out of memory.

"},{"base64":false,"type":"text/markdown","value":"A server the client requests can make the client allocate memory proportional to the body it sends, or to its decompressed size, until the node runs out of memory."}],"value":"A server the client requests can make the client allocate memory proportional to the body it sends, or to its decompressed size, until the node runs out of memory."}]}],"metrics":[{"cvssV4_0":{"Automatable":"NOT_DEFINED","Recovery":"NOT_DEFINED","Safety":"NOT_DEFINED","attackComplexity":"LOW","attackRequirements":"PRESENT","attackVector":"NETWORK","baseScore":8.2,"baseSeverity":"HIGH","privilegesRequired":"NONE","providerUrgency":"NOT_DEFINED","subAvailabilityImpact":"NONE","subConfidentialityImpact":"NONE","subIntegrityImpact":"NONE","userInteraction":"NONE","valueDensity":"NOT_DEFINED","vectorString":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","version":"4.0","vulnAvailabilityImpact":"HIGH","vulnConfidentialityImpact":"NONE","vulnIntegrityImpact":"NONE","vulnerabilityResponseEffort":"NOT_DEFINED"},"format":"CVSS","scenarios":[{"lang":"en","value":"GENERAL"}]}],"problemTypes":[{"descriptions":[{"cweId":"CWE-409","description":"CWE-409 Improper Handling of Highly Compressed Data (Data Amplification)","lang":"en","type":"CWE"}]},{"descriptions":[{"cweId":"CWE-770","description":"CWE-770 Allocation of Resources Without Limits or Throttling","lang":"en","type":"CWE"}]}],"providerMetadata":{"dateUpdated":"2026-10-26T13:49:40.392Z","orgId":"6b3ad84c-e1a6-4bf7-a703-f496b71e49db","shortName":"EEF"},"references":[{"name":"hackney 4.9.0 release notes","tags":["vendor-advisory","related"],"url":"https://github.com/benoitc/hackney/releases/tag/4.9.0"},{"name":"EEF CNA record for CVE-2026-97200","tags":["related"],"url":"https://cna.erlef.org/cves/CVE-2026-97200.html"},{"name":"OSV record EEF-CVE-2026-97200","tags":["related"],"url":"https://osv.dev/vulnerability/EEF-CVE-2026-97200"},{"name":"Cap buffered response bodies with max_body_size and bound decompression (benoitc/hackney#980)","tags":["related"],"url":"https://github.com/benoitc/hackney/pull/980"},{"name":"Introducing commit dc41b79 in benoitc/hackney","tags":["related"],"url":"https://github.com/benoitc/hackney/commit/dc41b79339ca0c41e16d3f2dd17172bc2de855d6"},{"name":"Fix commit 493e8bf in benoitc/hackney","tags":["patch"],"url":"https://github.com/benoitc/hackney/commit/493e8bf83a391548e5101792276ab3fc428fa76c"}],"source":{"discovery":"EXTERNAL"},"title":"Unbounded response body buffering and decompression in hackney","workarounds":[{"lang":"en","supportingMedia":[{"base64":false,"type":"text/html","value":"

Leave auto_decompress disabled (the default). Without it hackney_conn:maybe_decompress_body/2 never inflates the body, which closes the amplification half. The uncapped raw body read has no workaround.

"},{"base64":false,"type":"text/markdown","value":"Leave `auto_decompress` disabled (the default). Without it `hackney_conn:maybe_decompress_body/2` never inflates the body, which closes the amplification half. The uncapped raw body read has no workaround."}],"value":"Leave auto_decompress disabled (the default). Without it hackney_conn:maybe_decompress_body/2 never inflates the body, which closes the amplification half. The uncapped raw body read has no workaround."}],"x_proofOfConcept":[{"lang":"en","supportingMedia":[{"base64":false,"type":"text/html","value":"
    \n
  1. Build a gzip stream whose inflated size is 1 GB from a reused 64 KB chunk of zero bytes (deflate with window bits 31, best compression). The stream is about 1 MB on the wire.
    2. \n
  2. Start a TCP listener on loopback that accepts one connection, reads the request, and answers HTTP/1.1 200 OK with Content-Encoding: gzip, a Content-Length matching the stream, Connection: close, and the stream as body.
    3. \n
  3. From a node running a vulnerable hackney, call hackney:request(get, Url, [], <<>>, [{auto_decompress, true}, with_body]) against that listener.
    4. \n
  4. Observe the returned body of 1,048,576,000 bytes and the VM memory growing by about 1.8 GB for a 1,019,197 byte response. A larger stream, or an endless body without auto_decompress, drives the node out of memory.
    5. \n
"},{"base64":false,"type":"text/markdown","value":"1. Build a gzip stream whose inflated size is 1 GB from a reused 64 KB chunk of zero bytes (deflate with window bits 31, best compression). The stream is about 1 MB on the wire.\n2. Start a TCP listener on loopback that accepts one connection, reads the request, and answers `HTTP/1.1 200 OK` with `Content-Encoding: gzip`, a `Content-Length` matching the stream, `Connection: close`, and the stream as body.\n3. From a node running a vulnerable hackney, call `hackney:request(get, Url, [], <<>>, [{auto_decompress, true}, with_body])` against that listener.\n4. Observe the returned body of 1,048,576,000 bytes and the VM memory growing by about 1.8 GB for a 1,019,197 byte response. A larger stream, or an endless body without `auto_decompress`, drives the node out of memory."}],"value":"* Build a gzip stream whose inflated size is 1 GB from a reused 64 KB chunk of zero bytes (deflate with window bits 31, best compression). The stream is about 1 MB on the wire.\n* Start a TCP listener on loopback that accepts one connection, reads the request, and answers HTTP/1.1 200 OK with Content-Encoding: gzip, a Content-Length matching the stream, Connection: close, and the stream as body.\n* From a node running a vulnerable hackney, call hackney:request(get, Url, [], <<>>, [{auto_decompress, true}, with_body]) against that listener.\n* Observe the returned body of 1,048,576,000 bytes and the VM memory growing by about 1.8 GB for a 1,019,197 byte response. A larger stream, or an endless body without auto_decompress, drives the node out of memory."}],"x_technicalAnalysis":[{"lang":"en","supportingMedia":[{"base64":false,"type":"text/html","value":"

1. Unbounded accumulation. hackney:request/5 with a body, and hackney:body/1 after start_response/1, read the response in src/hackney_conn.erl through read_full_body/2, which appends each chunk to a binary accumulator until the body ends. The HTTP/2 and HTTP/3 synchronous paths accumulate stream data the same way. No option caps the size.

\n

2. One-step decompression. With {auto_decompress, true} hackney sends Accept-Encoding: gzip, deflate. Once the body is complete, maybe_decompress_body/2 dispatches on Content-Encoding to decompress_gzip/1 (zlib:gunzip/1) or decompress_deflate/1 (zlib:inflate/2). Both return the whole output at once, so the inflated size is bounded only by the compression ratio. The report measured a 1 MB gzip body inflating to 1 GB on hackney 4.8.4.

"},{"base64":false,"type":"text/markdown","value":"**1. Unbounded accumulation.** `hackney:request/5` with a body, and `hackney:body/1` after `start_response/1`, read the response in `src/hackney_conn.erl` through `read_full_body/2`, which appends each chunk to a binary accumulator until the body ends. The HTTP/2 and HTTP/3 synchronous paths accumulate stream data the same way. No option caps the size.\n\n**2. One-step decompression.** With `{auto_decompress, true}` hackney sends `Accept-Encoding: gzip, deflate`. Once the body is complete, `maybe_decompress_body/2` dispatches on `Content-Encoding` to `decompress_gzip/1` (`zlib:gunzip/1`) or `decompress_deflate/1` (`zlib:inflate/2`). Both return the whole output at once, so the inflated size is bounded only by the compression ratio. The report measured a 1 MB gzip body inflating to 1 GB on hackney 4.8.4."}],"value":"1. Unbounded accumulation. hackney:request/5 with a body, and hackney:body/1 after start_response/1, read the response in src/hackney_conn.erl through read_full_body/2, which appends each chunk to a binary accumulator until the body ends. The HTTP/2 and HTTP/3 synchronous paths accumulate stream data the same way. No option caps the size.\n\n2. One-step decompression. With {auto_decompress, true} hackney sends Accept-Encoding: gzip, deflate. Once the body is complete, maybe_decompress_body/2 dispatches on Content-Encoding to decompress_gzip/1 (zlib:gunzip/1) or decompress_deflate/1 (zlib:inflate/2). Both return the whole output at once, so the inflated size is bounded only by the compression ratio. The report measured a 1 MB gzip body inflating to 1 GB on hackney 4.8.4."}]}},"cveMetadata":{"assignerOrgId":"6b3ad84c-e1a6-4bf7-a703-f496b71e49db","assignerShortName":"EEF","cveId":"CVE-2026-97200","datePublished":"2026-10-26T13:49:40.392Z","dateReserved":"2026-10-26T05:15:01.465Z","dateUpdated":"2026-10-26T13:49:40.392Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"}