Skip to content

build(deps): update module github.com/labstack/echo/v5 to v5.4.1 [security] - #1171

Merged
mergify[bot] merged 1 commit into
masterfrom
renovate/go-github.com-labstack-echo-v5-vulnerability
Oct 24, 2026
Merged

mergify[bot] merged 1 commit into
masterfrom
renovate/go-github.com-labstack-echo-v5-vulnerability

Conversation

@renovate

@renovate renovate Bot commented Oct 24, 2026

Copy link
Copy Markdown
Contributor

This PR contains the following updates:

Package Change Age Confidence
github.com/labstack/echo/v5 v5.4.0 → v5.4.1 age confidence

Echo: WebSocket proxy can retain connections and handlers after client disconnect

GHSA-9969-6676-49h5

More information

Details

Impact

Echo's Proxy middleware could retain a handler, copy goroutines and network connections after a WebSocket client disconnected if the upstream kept its connection open. The raw proxy waited for both copy directions to finish without reliably closing the other connection when one direction ended. Rejected upgrade responses with HTTP keep-alive were one way to trigger this behavior.

Repeated connections can accumulate resources and cause denial of service. The impact depends on the proxy route's accessibility, upstream behavior and configured connection deadlines. This is a medium-severity resource-retention issue; no confidentiality or integrity impact is claimed.

Affected versions and patch

Affected versions are v4.15.0 through v4.16.0 and v5.0.0 through v5.4.0. The regression was introduced by commit e2133320c7, merged in #​2852.

Upgrade to v4.16.1 or v5.4.1. The patch uses Go's HTTP reverse proxy, preserves ordinary HTTP handling for rejected upgrades, and closes both sides of a WebSocket tunnel when either direction ends. It also covers an upstream that receives client EOF but deliberately keeps its side open.

Fix commits: v4, v5.

Workarounds

Disable WebSocket proxy routes until upgrading, or apply bounded connection lifetimes and resource limits at the frontend and upstream. Server idle timeouts alone may not cover hijacked WebSocket connections.

Credits

Reported by @​irisitymichaelgrundberg.

Severity

  • CVSS Score: 8.2 / 10 (High)
  • Vector String: 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

References

This data is provided by OSV and the GitHub Advisory Database (CC-BY 4.0).


Echo: Rejected WebSocket upgrades can bypass Echo route-level access controls

GHSA-cxf3-9h37-7rr8

More information

Details

Impact

Echo's Proxy middleware selected a raw tunnel when the client requested a WebSocket upgrade, before validating whether the configured upstream accepted it. If the upstream returned an ordinary HTTP response and kept the connection alive, subsequent HTTP requests on that connection were forwarded directly to the same upstream without passing through Echo's router or route-specific middleware.

This can bypass authentication or authorization enforced only at Echo's routes. Exploitation requires an accessible proxy route, an upstream that rejects the upgrade without closing the connection, and a target resource on that upstream without independent authorization. The issue does not provide arbitrary-host proxying or bypass the upstream's own authentication.

Affected versions and patch

The raw proxy implementation is present throughout the v4 and v5 release lines up to v4.16.0 and v5.4.0. Runtime reproductions confirmed v4.15.0, v4.16.0, v5.0.0 and v5.4.0; source history also shows the missing upgrade-response validation in v4.0.0.

Upgrade to v4.16.1 or v5.4.1. The patch routes WebSocket upgrades through Go's HTTP reverse proxy and enters a tunnel only after a valid upstream upgrade response. Rejected upgrades remain ordinary HTTP requests.

Fix commits: v4, v5.

Workarounds

Reject WebSocket upgrade requests before they reach affected Echo proxy routes, or enforce authorization independently at every upstream resource. Merely returning a non-101 response from the upstream does not prevent the vulnerable raw tunnel.

Credits

Reported by @​Duodum.

Severity

  • CVSS Score: 9.1 / 10 (Critical)
  • Vector String: CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N

References

This data is provided by OSV and the GitHub Advisory Database (CC-BY 4.0).


Echo: X-Real-IP extractor trusts the header value instead of the sending peer

GHSA-h53f-rm2w-w7rg

More information

Details

Impact

In Echo v5, ExtractIPFromRealIPHeader checked whether the address supplied in X-Real-IP belonged to a trusted range instead of checking the address of the direct network peer. An untrusted client could therefore supply a loopback or private address and have it returned by c.RealIP() with the extractor's default trust options.

Applications using this extractor for IP allowlists, IP-based rate limiting, audit logs or forwarding headers could accept a spoofed client identity. Exploitation requires that the attacker can reach Echo directly or that an intervening proxy passes through the attacker's X-Real-IP value. Applications using the default direct IP extractor are unaffected. Echo v4 is unaffected by this v5 regression.

Affected versions and patch

All v5 releases from v5.0.0 through v5.4.0 are affected. Upgrade to v5.4.1.

The patch validates the direct peer before accepting a syntactically valid X-Real-IP value, restoring the v4 trust model. Fix commit.

Workarounds

Use ExtractIPDirect until upgrading, or ensure that only trusted proxies can reach Echo and that they replace client-supplied forwarding headers. Configure trusted proxy ranges narrowly for the deployment.

Credits

Reported and patched by @​rm-Umar. @​Duodum independently reported the same issue in GHSA-c3r7-8j36-7mx8, which is tracked as a duplicate of this advisory.

Severity

  • CVSS Score: 5.3 / 10 (Medium)
  • Vector String: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N

References

This data is provided by OSV and the GitHub Advisory Database (CC-BY 4.0).


Echo: BodyDump can discard BodyLimit overflow errors for unknown-length requests

GHSA-x9ff-22gv-65f4

More information

Details

Impact

Echo v5's BodyDump middleware captured a bounded prefix, drained the remaining request body while ignoring read errors, and then replaced the body with the captured prefix. With an outer BodyLimit, an unknown-length request such as an HTTP/1.1 chunked request, and a BodyDump capture limit below the BodyLimit threshold, an overflow detected during that drain could be discarded.

The handler could consequently process a truncated request and return success even though the original body exceeded the configured limit. This is a conditional size-policy bypass; no general authentication bypass, production data modification or denial of service is claimed. Known-length requests rejected before BodyDump runs are unaffected by this path.

Affected versions and patch

All v5 releases from v5.0.0 through v5.4.0 contain the affected drain-and-replace logic. Runtime reproduction confirmed v5.4.0; source comparison confirms the same request handling in v5.0.0.

Upgrade to v5.4.1. BodyDump now captures only the configured prefix for logging and reconstructs the handler's reader from that prefix plus the unread original stream. It preserves the original closer and read errors instead of silently draining the rest. Fix commit.

Streaming BodyLimit enforcement still requires the application to consume the request body and handle read errors. The BodyDump capture cap is a logging cap, not an application body limit.

Workarounds

Disable BodyDump on affected routes, or ensure the BodyLimit overflow is detected during the initial capture by setting the capture limit above the enforced body limit. Upgrading is recommended because it also preserves the complete body for handlers.

Credits

Reported by @​Duodum. The fix incorporates @​jakezwang's request-body preservation work in #​3105.

Severity

  • CVSS Score: 6.3 / 10 (Medium)
  • Vector String: CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N

References

This data is provided by OSV and the GitHub Advisory Database (CC-BY 4.0).


Release Notes

labstack/echo (github.com/labstack/echo/v5)

v5.4.1

Compare Source

Security fixes

Upgrade with go get github.com/labstack/echo/v5@v5.4.1.

  • Validate the sending peer before accepting X-Real-IP in ExtractIPFromRealIPHeader, preventing client IP spoofing when that extractor is configured. GHSA-h53f-rm2w-w7rg
  • Validate upstream WebSocket upgrade responses through Go's HTTP reverse proxy. Rejected upgrades no longer leave a raw tunnel that bypasses Echo routing and route middleware. GHSA-cxf3-9h37-7rr8
  • Close both WebSocket connections when either side disconnects, including when an upstream ignores client EOF. GHSA-9969-6676-49h5
  • Preserve the complete request body beyond the BodyDump capture limit and propagate read errors, including BodyLimit overflows. GHSA-x9ff-22gv-65f4, #​3105
  • Restore the original response writer before returning BodyDump buffers to the pool, preventing writes into a reused buffer during error handling. #​3174
  • Accept decompressed bodies exactly at the configured limit; keep overflow errors sticky. #​3136
  • Include the BodyLimit overflow-check correction and static-path hardening already merged since v5.4.0.

Behavior to review when upgrading

The proxy removes incoming Forwarded headers as defensive hardening. Backends relying on that header should use forwarding headers generated from the configured trusted IP extractor and scheme. WebSocket requests now honor the configured proxy transport, response modifier, target path and query in the same way as HTTP requests; ws and wss targets remain supported.

BodyDump capture limits bound logging only: handlers receive the full request, and clients receive the full response. Applications must still consume request bodies and handle read errors to enforce streaming body limits.

Go 1.25 remains the module minimum. Build deployments with a supported, patched Go toolchain.

Other changes

This release also includes ValidateCtx, router fixes for escaped colons, custom methods and parameter storage, binding support for OpenAPI date formats, and documentation updates.

Thanks to @​rm-Umar, @​Duodum, @​irisitymichaelgrundberg, @​Hama1cco, @​jakezwang, @​Shubham-Padkonde and @​rajpreetcodes for reports and contributions, and to the contributors credited in the individual commits.

Validation: upstream lint checks and race tests passed on Linux, macOS and Windows with Go 1.25, 1.26 and 1.27. Local tests on patched toolchains, vet, staticcheck and govulncheck also passed; govulncheck reported no reachable vulnerabilities.

Full changelog: labstack/echo@v5.4.0...v5.4.1


Configuration

📅 Schedule: (in timezone Asia/Shanghai)

  • Branch creation
    • At any time (no schedule defined)
  • Automerge
    • On day 4, 5, 6, 7, and 8 of the month (* * 4,5,6,7,8 * *)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.


  • If you want to rebase/retry this PR, check this box

This PR was generated by Mend Renovate. View the repository job log.

@renovate renovate Bot added ci:auto-merge dependencies Pull requests that update a dependency file labels Oct 24, 2026
@mergify

mergify Bot commented Oct 24, 2026 •

Copy link
Copy Markdown
Contributor

Queued — the merge queue status continues in this comment ↓.

@mergify mergify Bot added the ci:queued label Oct 24, 2026
@mergify

mergify Bot commented Oct 24, 2026 •

Copy link
Copy Markdown
Contributor

Merge Queue Status

  • ✅ Entered queue — 2026-10-24 03:01 UTC · Rule: renovate · triggered by rule Automatic merge on approval
  • 🚫 Left the queue — 2026-10-24 03:02 UTC · at 2a6ccdfbbd287e0682b509a41042708d8476364f

This pull request spent 12 seconds in the queue, with no time running CI.

Reason

The pull request can't be updated

merge conflict between base and head

Hint

You should update or rebase your pull request manually. If you do, this pull request will automatically be requeued once the queue conditions match again.
If you think this was a flaky issue, you can requeue the pull request, without updating it, by posting a @mergifyio queue comment.

Requeued — the merge queue status continues in this comment ↓.

@renovate
renovate Bot force-pushed the renovate/go-github.com-labstack-echo-v5-vulnerability branch from 2a6ccdf to a862540 Compare October 24, 2026 03:02
@mergify
mergify Bot merged commit 1997713 into master Oct 24, 2026
7 checks passed
@mergify

mergify Bot commented Oct 24, 2026

Copy link
Copy Markdown
Contributor

Merge Queue Status

  • ✅ Entered queue — 2026-10-24 03:06 UTC · Rule: renovate · triggered by rule Automatic merge on approval
  • ✅ Checks skipped · PR is already up-to-date
  • ✅ Merged — 2026-10-24 03:06 UTC · at 1997713eafc51487cbea705b37011442f35539fa

This pull request spent 7 seconds in the queue, with no time running CI.

Required conditions to merge

@mergify
mergify Bot deleted the renovate/go-github.com-labstack-echo-v5-vulnerability branch October 24, 2026 03:06
@mergify mergify Bot removed the ci:queued label Oct 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ci:auto-merge dependencies Pull requests that update a dependency file size/XS

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant