CVE-2026-23394

{"dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": {"cveId": "CVE-2026-23394", "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "state": "PUBLISHED", "assignerShortName": "Linux", "dateReserved": "2026-01-13T15:37:46.011Z", "datePublished": "2026-03-25T10:33:18.180Z", "dateUpdated": "2026-08-19T16:27:54.571Z"}, "containers": {"cna": {"providerMetadata": {"orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "shortName": "Linux", "dateUpdated": "2026-08-19T16:27:54.571Z"}, "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\naf_unix: Give up GC if MSG_PEEK intervened.\n\nIgor Ushakov reported that GC purged the receive queue of\nan alive socket due to a race with MSG_PEEK with a nice repro.\n\nThis is the exact same issue previously fixed by commit\ncbcf01128d0a (\"af_unix: fix garbage collect vs MSG_PEEK\").\n\nAfter GC was replaced with the current algorithm, the cited\ncommit removed the locking dance in unix_peek_fds() and\nreintroduced the same issue.\n\nThe problem is that MSG_PEEK bumps a file refcount without\ninteracting with GC.\n\nConsider an SCC containing sk-A and sk-B, where sk-A is\nclose()d but can be recv()ed via sk-B.\n\nThe bad thing happens if sk-A is recv()ed with MSG_PEEK from\nsk-B and sk-B is close()d while GC is checking unix_vertex_dead()\nfor sk-A and sk-B.\n\n GC thread User thread\n --------- -----------\n unix_vertex_dead(sk-A)\n -> true <------.\n \\\n `------ recv(sk-B, MSG_PEEK)\n invalidate !! -> sk-A's file refcount : 1 -> 2\n\n close(sk-B)\n -> sk-B's file refcount : 2 -> 1\n unix_vertex_dead(sk-B)\n -> true\n\nInitially, sk-A's file refcount is 1 by the inflight fd in sk-B\nrecvq. GC thinks sk-A is dead because the file refcount is the\nsame as the number of its inflight fds.\n\nHowever, sk-A's file refcount is bumped silently by MSG_PEEK,\nwhich invalidates the previous evaluation.\n\nAt this moment, sk-B's file refcount is 2; one by the open fd,\nand one by the inflight fd in sk-A. The subsequent close()\nreleases one refcount by the former.\n\nFinally, GC incorrectly concludes that both sk-A and sk-B are dead.\n\nOne option is to restore the locking dance in unix_peek_fds(),\nbut we can resolve this more elegantly thanks to the new algorithm.\n\nThe point is that the issue does not occur without the subsequent\nclose() and we actually do not need to synchronise MSG_PEEK with\nthe dead SCC detection.\n\nWhen the issue occurs, close() and GC touch the same file refcount.\nIf GC sees the refcount being decremented by close(), it can just\ngive up garbage-collecting the SCC.\n\nTherefore, we only need to signal the race during MSG_PEEK with\na proper memory barrier to make it visible to the GC.\n\nLet's use seqcount_t to notify GC when MSG_PEEK occurs and let\nit defer the SCC to the next run.\n\nThis way no locking is needed on the MSG_PEEK side, and we can\navoid imposing a penalty on every MSG_PEEK unnecessarily.\n\nNote that we can retry within unix_scc_dead() if MSG_PEEK is\ndetected, but we do not do so to avoid hung task splat from\nabusive MSG_PEEK calls."}], "affected": [{"product": "Linux", "vendor": "Linux", "defaultStatus": "unaffected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": ["net/unix/af_unix.c", "net/unix/af_unix.h", "net/unix/garbage.c"], "versions": [{"version": "61a75360dca93c945ef6bd757f8b8a96f39b77cb", "lessThan": "980999a96e1ad043f1197606e5d637219cb102b9", "status": "affected", "versionType": "git"}, {"version": "7b1ffbd3b22e755d481d49647dcb7c5cfbde5844", "lessThan": "3106f326f67c03dd9da4ca64663d11e40138cf40", "status": "affected", "versionType": "git"}, {"version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "lessThan": "e3dd56fb5683ba80bf8d7a2f9aa21cfa53f05202", "status": "affected", "versionType": "git"}, {"version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "lessThan": "72cf49ad50c16270b52bc512d9c2df5743922968", "status": "affected", "versionType": "git"}, {"version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "lessThan": "37dd7ab332396eb8dd80b2dc7ea4b61abf767436", "status": "affected", "versionType": "git"}, {"version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "lessThan": "e5b31d988a41549037b8d8721a3c3cae893d8670", "status": "affected", "versionType": "git"}, {"version": "6.1.141", "lessThan": "6.1.183", "status": "affected", "versionType": "semver"}, {"version": "6.6.93", "lessThan": "6.6.142", "status": "affected", "versionType": "semver"}]}, {"product": "Linux", "vendor": "Linux", "defaultStatus": "affected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": ["net/unix/af_unix.c", "net/unix/af_unix.h", "net/unix/garbage.c"], "versions": [{"version": "6.10", "status": "affected"}, {"version": "0", "lessThan": "6.10", "status": "unaffected", "versionType": "semver"}, {"version": "6.1.183", "lessThanOrEqual": "6.1.*", "status": "unaffected", "versionType": "semver"}, {"version": "6.6.142", "lessThanOrEqual": "6.6.*", "status": "unaffected", "versionType": "semver"}, {"version": "6.12.92", "lessThanOrEqual": "6.12.*", "status": "unaffected", "versionType": "semver"}, {"version": "6.18.23", "lessThanOrEqual": "6.18.*", "status": "unaffected", "versionType": "semver"}, {"version": "6.19.10", "lessThanOrEqual": "6.19.*", "status": "unaffected", "versionType": "semver"}, {"version": "7.0", "lessThanOrEqual": "*", "status": "unaffected", "versionType": "original_commit_for_fix"}]}], "cpeApplicability": [{"nodes": [{"operator": "OR", "negate": false, "cpeMatch": [{"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.1.141", "versionEndExcluding": "6.1.183"}, {"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.6.93", "versionEndExcluding": "6.6.142"}, {"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.10", "versionEndExcluding": "6.12.92"}, {"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.10", "versionEndExcluding": "6.18.23"}, {"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.10", "versionEndExcluding": "6.19.10"}, {"vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.10", "versionEndExcluding": "7.0"}]}]}], "references": [{"url": "https://git.kernel.org/stable/c/980999a96e1ad043f1197606e5d637219cb102b9"}, {"url": "https://git.kernel.org/stable/c/3106f326f67c03dd9da4ca64663d11e40138cf40"}, {"url": "https://git.kernel.org/stable/c/e3dd56fb5683ba80bf8d7a2f9aa21cfa53f05202"}, {"url": "https://git.kernel.org/stable/c/72cf49ad50c16270b52bc512d9c2df5743922968"}, {"url": "https://git.kernel.org/stable/c/37dd7ab332396eb8dd80b2dc7ea4b61abf767436"}, {"url": "https://git.kernel.org/stable/c/e5b31d988a41549037b8d8721a3c3cae893d8670"}], "title": "af_unix: Give up GC if MSG_PEEK intervened.", "x_generator": {"engine": "bippy-1.2.0"}}}}

{}

{"affected": [{"affectedData": [{"defaultStatus": "unaffected", "product": "Linux", "programFiles": ["net/unix/af_unix.c", "net/unix/af_unix.h", "net/unix/garbage.c"], "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "versions": [{"lessThan": "980999a96e1ad043f1197606e5d637219cb102b9", "status": "affected", "version": "61a75360dca93c945ef6bd757f8b8a96f39b77cb", "versionType": "git"}, {"lessThan": "3106f326f67c03dd9da4ca64663d11e40138cf40", "status": "affected", "version": "7b1ffbd3b22e755d481d49647dcb7c5cfbde5844", "versionType": "git"}, {"lessThan": "e3dd56fb5683ba80bf8d7a2f9aa21cfa53f05202", "status": "affected", "version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "versionType": "git"}, {"lessThan": "72cf49ad50c16270b52bc512d9c2df5743922968", "status": "affected", "version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "versionType": "git"}, {"lessThan": "37dd7ab332396eb8dd80b2dc7ea4b61abf767436", "status": "affected", "version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "versionType": "git"}, {"lessThan": "e5b31d988a41549037b8d8721a3c3cae893d8670", "status": "affected", "version": "118f457da9ed58a79e24b73c2ef0aa1987241f0e", "versionType": "git"}, {"lessThan": "6.1.183", "status": "affected", "version": "6.1.141", "versionType": "semver"}, {"lessThan": "6.6.142", "status": "affected", "version": "6.6.93", "versionType": "semver"}]}, {"defaultStatus": "affected", "product": "Linux", "programFiles": ["net/unix/af_unix.c", "net/unix/af_unix.h", "net/unix/garbage.c"], "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "versions": [{"status": "affected", "version": "6.10"}, {"lessThan": "6.10", "status": "unaffected", "version": "0", "versionType": "semver"}, {"lessThanOrEqual": "6.1.*", "status": "unaffected", "version": "6.1.183", "versionType": "semver"}, {"lessThanOrEqual": "6.6.*", "status": "unaffected", "version": "6.6.142", "versionType": "semver"}, {"lessThanOrEqual": "6.12.*", "status": "unaffected", "version": "6.12.92", "versionType": "semver"}, {"lessThanOrEqual": "6.18.*", "status": "unaffected", "version": "6.18.23", "versionType": "semver"}, {"lessThanOrEqual": "6.19.*", "status": "unaffected", "version": "6.19.10", "versionType": "semver"}, {"lessThanOrEqual": "*", "status": "unaffected", "version": "7.0", "versionType": "original_commit_for_fix"}]}], "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}], "configurations": [{"nodes": [{"cpeMatch": [{"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "matchCriteriaId": "B0A3421E-59FE-4E16-9DC6-4CAF59C361C7", "versionEndExcluding": "6.2", "versionStartIncluding": "6.1.141", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "matchCriteriaId": "4CAF81AF-E0B7-4112-B091-4DE2CB164414", "versionEndExcluding": "6.7", "versionStartIncluding": "6.6.93", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "matchCriteriaId": "7109F074-E2CB-44FB-A4DB-E45B24EF0E17", "versionEndExcluding": "6.18.23", "versionStartIncluding": "6.10.1", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "matchCriteriaId": "96D34333-38BE-4414-9E79-6EB764329581", "versionEndExcluding": "6.19.10", "versionStartIncluding": "6.19", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:6.10:-:*:*:*:*:*:*", "matchCriteriaId": "9EA80796-744E-45F5-8632-2AB4F7889FCD", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc1:*:*:*:*:*:*", "matchCriteriaId": "F253B622-8837-4245-BCE5-A7BF8FC76A16", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc2:*:*:*:*:*:*", "matchCriteriaId": "4AE85AD8-4641-4E7C-A2F4-305E2CD9EE64", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc3:*:*:*:*:*:*", "matchCriteriaId": "F666C8D8-6538-46D4-B318-87610DE64C34", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc4:*:*:*:*:*:*", "matchCriteriaId": "02259FDA-961B-47BC-AE7F-93D7EC6E90C2", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc5:*:*:*:*:*:*", "matchCriteriaId": "58A9FEFF-C040-420D-8F0A-BFDAAA1DF258", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc6:*:*:*:*:*:*", "matchCriteriaId": "1D2315C0-D46F-4F85-9754-F9E5E11374A6", "vulnerable": true}, {"criteria": "cpe:2.3:o:linux:linux_kernel:7.0:rc7:*:*:*:*:*:*", "matchCriteriaId": "512EE3A8-A590-4501-9A94-5D4B268D6138", "vulnerable": true}], "negate": false, "operator": "OR"}]}], "cveTags": [], "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\naf_unix: Give up GC if MSG_PEEK intervened.\n\nIgor Ushakov reported that GC purged the receive queue of\nan alive socket due to a race with MSG_PEEK with a nice repro.\n\nThis is the exact same issue previously fixed by commit\ncbcf01128d0a (\"af_unix: fix garbage collect vs MSG_PEEK\").\n\nAfter GC was replaced with the current algorithm, the cited\ncommit removed the locking dance in unix_peek_fds() and\nreintroduced the same issue.\n\nThe problem is that MSG_PEEK bumps a file refcount without\ninteracting with GC.\n\nConsider an SCC containing sk-A and sk-B, where sk-A is\nclose()d but can be recv()ed via sk-B.\n\nThe bad thing happens if sk-A is recv()ed with MSG_PEEK from\nsk-B and sk-B is close()d while GC is checking unix_vertex_dead()\nfor sk-A and sk-B.\n\n GC thread User thread\n --------- -----------\n unix_vertex_dead(sk-A)\n -> true <------.\n \\\n `------ recv(sk-B, MSG_PEEK)\n invalidate !! -> sk-A's file refcount : 1 -> 2\n\n close(sk-B)\n -> sk-B's file refcount : 2 -> 1\n unix_vertex_dead(sk-B)\n -> true\n\nInitially, sk-A's file refcount is 1 by the inflight fd in sk-B\nrecvq. GC thinks sk-A is dead because the file refcount is the\nsame as the number of its inflight fds.\n\nHowever, sk-A's file refcount is bumped silently by MSG_PEEK,\nwhich invalidates the previous evaluation.\n\nAt this moment, sk-B's file refcount is 2; one by the open fd,\nand one by the inflight fd in sk-A. The subsequent close()\nreleases one refcount by the former.\n\nFinally, GC incorrectly concludes that both sk-A and sk-B are dead.\n\nOne option is to restore the locking dance in unix_peek_fds(),\nbut we can resolve this more elegantly thanks to the new algorithm.\n\nThe point is that the issue does not occur without the subsequent\nclose() and we actually do not need to synchronise MSG_PEEK with\nthe dead SCC detection.\n\nWhen the issue occurs, close() and GC touch the same file refcount.\nIf GC sees the refcount being decremented by close(), it can just\ngive up garbage-collecting the SCC.\n\nTherefore, we only need to signal the race during MSG_PEEK with\na proper memory barrier to make it visible to the GC.\n\nLet's use seqcount_t to notify GC when MSG_PEEK occurs and let\nit defer the SCC to the next run.\n\nThis way no locking is needed on the MSG_PEEK side, and we can\navoid imposing a penalty on every MSG_PEEK unnecessarily.\n\nNote that we can retry within unix_scc_dead() if MSG_PEEK is\ndetected, but we do not do so to avoid hung task splat from\nabusive MSG_PEEK calls."}, {"lang": "es", "value": "En el kernel de Linux, la siguiente vulnerabilidad ha sido resuelta:\n\naf_unix: Abandonar la recolecci\u00f3n de basura (GC) si MSG_PEEK intervino.\n\nIgor Ushakov inform\u00f3 que la recolecci\u00f3n de basura (GC) purg\u00f3 la cola de recepci\u00f3n de un socket activo debido a una condici\u00f3n de carrera con MSG_PEEK con una buena reproducci\u00f3n.\n\nEste es exactamente el mismo problema previamente solucionado por el commit cbcf01128d0a ('af_unix: corregir recolecci\u00f3n de basura vs MSG_PEEK').\n\nDespu\u00e9s de que la recolecci\u00f3n de basura (GC) fue reemplazada por el algoritmo actual, el commit citado elimin\u00f3 la 'danza de bloqueo' en unix_peek_fds() y reintrodujo el mismo problema.\n\nEl problema es que MSG_PEEK incrementa un contador de referencias de archivo sin interactuar con la recolecci\u00f3n de basura (GC).\n\nConsidere un SCC que contiene sk-A y sk-B, donde sk-A est\u00e1 close()d (cerrado) pero puede ser recv()ed (recibido) a trav\u00e9s de sk-B.\n\nLo malo sucede si sk-A es recv()ed (recibido) con MSG_PEEK desde sk-B y sk-B est\u00e1 close()d (cerrado) mientras la recolecci\u00f3n de basura (GC) est\u00e1 verificando unix_vertex_dead() para sk-A y sk-B.\n\n Hilo de GC Hilo de usuario\n --------- -----------\n unix_vertex_dead(sk-A)\n -&gt; true &lt;------.\n \\\n `------ recv(sk-B, MSG_PEEK)\n \u00a1\u00a1invalidar!! -&gt; contador de referencias de archivo de sk-A : 1 -&gt; 2\n\n close(sk-B)\n -&gt; contador de referencias de archivo de sk-B : 2 -&gt; 1\n unix_vertex_dead(sk-B)\n -&gt; true\n\nInicialmente, el contador de referencias de archivo de sk-A es 1 por el descriptor de archivo en tr\u00e1nsito en la cola de recepci\u00f3n de sk-B. La recolecci\u00f3n de basura (GC) piensa que sk-A est\u00e1 muerto porque el contador de referencias de archivo es el mismo que el n\u00famero de sus descriptores de archivo en tr\u00e1nsito.\n\nSin embargo, el contador de referencias de archivo de sk-A es incrementado silenciosamente por MSG_PEEK, lo que invalida la evaluaci\u00f3n anterior.\n\nEn este momento, el contador de referencias de archivo de sk-B es 2; uno por el descriptor de archivo abierto, y uno por el descriptor de archivo en tr\u00e1nsito en sk-A. El close() (cierre) subsiguiente libera un contador de referencias por el primero.\n\nFinalmente, la recolecci\u00f3n de basura (GC) concluye incorrectamente que tanto sk-A como sk-B est\u00e1n muertos.\n\nUna opci\u00f3n es restaurar la 'danza de bloqueo' en unix_peek_fds(), pero podemos resolver esto de manera m\u00e1s elegante gracias al nuevo algoritmo.\n\nEl punto es que el problema no ocurre sin el close() (cierre) subsiguiente y en realidad no necesitamos sincronizar MSG_PEEK con la detecci\u00f3n de SCC muertos.\n\nCuando ocurre el problema, close() (el cierre) y la recolecci\u00f3n de basura (GC) tocan el mismo contador de referencias de archivo. Si la recolecci\u00f3n de basura (GC) ve que el contador de referencias es decrementado por close() (el cierre), puede simplemente abandonar la recolecci\u00f3n de basura del SCC.\n\nPor lo tanto, solo necesitamos se\u00f1alar la condici\u00f3n de carrera durante MSG_PEEK con una barrera de memoria adecuada para hacerla visible a la recolecci\u00f3n de basura (GC).\n\nUsemos seqcount_t para notificar a la recolecci\u00f3n de basura (GC) cuando ocurre MSG_PEEK y permitirle aplazar el SCC a la siguiente ejecuci\u00f3n.\n\nDe esta manera, no se necesita bloqueo en el lado de MSG_PEEK, y podemos evitar imponer una penalizaci\u00f3n a cada MSG_PEEK innecesariamente.\n\nTenga en cuenta que podemos reintentar dentro de unix_scc_dead() si se detecta MSG_PEEK, pero no lo hacemos para evitar la 'salpicadura' de tareas colgadas por llamadas abusivas a MSG_PEEK."}], "id": "CVE-2026-23394", "lastModified": "2026-08-19T17:18:45.960", "metrics": {"cvssMetricV31": [{"cvssData": {"attackComplexity": "HIGH", "attackVector": "LOCAL", "availabilityImpact": "HIGH", "baseScore": 4.7, "baseSeverity": "MEDIUM", "confidentialityImpact": "NONE", "integrityImpact": "NONE", "privilegesRequired": "LOW", "scope": "UNCHANGED", "userInteraction": "NONE", "vectorString": "CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H", "version": "3.1"}, "exploitabilityScore": 1.0, "impactScore": 3.6, "source": "nvd@nist.gov", "type": "Primary"}]}, "published": "2026-03-25T11:16:40.190", "references": [{"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "url": "https://git.kernel.org/stable/c/3106f326f67c03dd9da4ca64663d11e40138cf40"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "tags": ["Patch"], "url": "https://git.kernel.org/stable/c/37dd7ab332396eb8dd80b2dc7ea4b61abf767436"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "tags": ["Patch"], "url": "https://git.kernel.org/stable/c/72cf49ad50c16270b52bc512d9c2df5743922968"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "url": "https://git.kernel.org/stable/c/980999a96e1ad043f1197606e5d637219cb102b9"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "url": "https://git.kernel.org/stable/c/e3dd56fb5683ba80bf8d7a2f9aa21cfa53f05202"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "tags": ["Patch"], "url": "https://git.kernel.org/stable/c/e5b31d988a41549037b8d8721a3c3cae893d8670"}], "sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "vulnStatus": "Modified", "weaknesses": [{"description": [{"lang": "en", "value": "CWE-362"}], "source": "nvd@nist.gov", "type": "Primary"}]}

{"bugzilla": {"description": "kernel: af_unix: Give up GC if MSG_PEEK intervened", "id": "2451201", "url": "https://bugzilla.redhat.com/show_bug.cgi?id=2451201"}, "csaw": false, "cvss3": {"cvss3_base_score": "5.5", "cvss3_scoring_vector": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "status": "draft"}, "cwe": "CWE-367", "details": ["A flaw was found in the Linux kernel's af_unix component. A race condition exists between the MSG_PEEK operation and the garbage collection process. This can allow a local user to cause the garbage collector to incorrectly purge the receive queue of an alive socket, potentially leading to a denial of service."], "name": "CVE-2026-23394", "package_state": [{"cpe": "cpe:/o:redhat:enterprise_linux:10", "fix_state": "Fix deferred", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 10"}, {"cpe": "cpe:/o:redhat:enterprise_linux:6", "fix_state": "Out of support scope", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 6"}, {"cpe": "cpe:/o:redhat:enterprise_linux:7", "fix_state": "Not affected", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 7"}, {"cpe": "cpe:/o:redhat:enterprise_linux:7", "fix_state": "Not affected", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 7"}, {"cpe": "cpe:/o:redhat:enterprise_linux:8", "fix_state": "Not affected", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 8"}, {"cpe": "cpe:/o:redhat:enterprise_linux:8", "fix_state": "Not affected", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 8"}, {"cpe": "cpe:/o:redhat:enterprise_linux:9", "fix_state": "Not affected", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 9"}, {"cpe": "cpe:/o:redhat:enterprise_linux:9", "fix_state": "Not affected", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 9"}], "public_date": "2026-03-25T00:00:00Z", "references": ["https://www.cve.org/CVERecord?id=CVE-2026-23394\nhttps://nvd.nist.gov/vuln/detail/CVE-2026-23394\nhttps://lore.kernel.org/linux-cve-announce/2026032549-CVE-2026-23394-9205@gregkh/T"], "statement": "This flaw affects AF_UNIX socket file descriptor passing with MSG_PEEK. The race between GC's liveness check and MSG_PEEK refcount bumping can cause GC to incorrectly purge queues of live sockets. This is a regression of a previously fixed issue (commit cbcf01128d0a) that was reintroduced when the GC algorithm was replaced.", "threat_severity": "Moderate"}

{"affected": [{"configurations": [{"platform": null, "status": "affected", "versions": {"scheme": "code_commit", "value": "[118f457da9ed58a79e24b73c2ef0aa1987241f0e,37dd7ab332396eb8dd80b2dc7ea4b61abf767436)"}}, {"platform": null, "status": "affected", "versions": {"scheme": "code_commit", "value": "[118f457da9ed58a79e24b73c2ef0aa1987241f0e,e5b31d988a41549037b8d8721a3c3cae893d8670)"}}, {"platform": null, "status": "affected", "versions": {"scheme": "code_commit", "value": "61a75360dca93c945ef6bd757f8b8a96f39b77cb"}}, {"platform": null, "status": "affected", "versions": {"scheme": "code_commit", "value": "7b1ffbd3b22e755d481d49647dcb7c5cfbde5844"}}], "enrichment": {"confidence": 99.0, "confidence_source": "inferred", "scores": [{"score": 99.0, "source": "inferred"}, {"score": 100.0, "source": "matching"}]}, "original": {"product": "Linux", "source": "cna", "vendor": "Linux"}, "product": "linux_kernel", "vendor": "linux"}, {"configurations": [{"platform": null, "status": "affected", "versions": {"scheme": "semver", "value": "6.10"}}, {"platform": null, "status": "unaffected", "versions": {"scheme": "semver", "value": "[0,6.10)"}}, {"platform": null, "status": "unaffected", "versions": {"scheme": "semver", "value": "[6.19.10,6.20.0)"}}, {"platform": null, "status": "unaffected", "versions": {"scheme": "semver", "value": "[7.0-rc5,*]"}}], "enrichment": {"confidence": 99.0, "confidence_source": "inferred", "scores": [{"score": 99.0, "source": "inferred"}, {"score": 100.0, "source": "matching"}]}, "original": {"product": "Linux", "source": "cna", "vendor": "Linux"}, "product": "linux_kernel", "vendor": "linux"}], "analysis": {"en": {"generated_at": "2026-04-28T21:58:03.825618+00:00", "value": {"mitigation_remediation": ["Upgrade the Linux kernel to the latest stable release that includes the upstream commit that fixes the MSG_PEEK/GC race; the change is present in kernel 6.10 and newer releases.", "If an immediate kernel upgrade is not feasible, modify the application to avoid using MSG_PEEK on a Unix domain socket that may later be closed; refactor the code to read data directly without peeking or to recreate the socket instead of peeking.", "Apply any available vendor\u2010provided backport of the upstream fix or, if none exists, restrict local user access to the vulnerable Unix domain sockets used by the application to limit the attack surface."], "summary": {"action": "Patch", "impact": "Data Integrity / Availability Loss"}, "threat_synthesis": {"affected_systems": "All systems running the Linux kernel before the inclusion of the upstream fix for the MSG_PEEK/GC race are vulnerable. The advisory does not list specific distribution names or precise kernel versions, so affected users should verify that their kernel revision predates the commit that corrects the issue. The pressure is on any kernel that still relies on the older garbage\u2011collection algorithm for Unix sockets.", "description_and_impact": "The flaw is a race condition between the garbage\u2011collection (GC) process and the MSG_PEEK call in the Linux kernel's Unix domain socket code. When MSG_PEEK is performed on a socket whose peer is later closed, the reference counter for the socket\u2019s file is silently increased. The GC thread, running concurrently, misinterprets this state change and decides that both sockets are dead, purging the receive queue of a socket that is still in use. The result is that data lost by the affected application cannot be retrieved, effectively breaking reliable communication and potentially causing a denial of service. The underlying weakness is a concurrency bug involving race conditions (CWE\u2011362, CWE\u2011367).", "risk_and_exploitability": "The CVSS score of 4.7 indicates moderate severity, and the EPSS score of less than 1\u202f% coupled with the absence from KEV suggests a relatively low likelihood of exploitation. The race condition requires the attacker to be able to execute a MSG_PEEK on a Unix domain socket that shares a socket reference with another socket that is subsequently closed, implying a local execution context. The vulnerability does not provide remote code execution or direct system compromise; the impact is limited to loss of queued data and potential service disruption. The inferred attack vector is local user or process with access to the vulnerable Unix socket; this inference is based on the need for concurrent access to both sockets and is not explicitly stated in the description."}}}}, "created": "2026-03-25T21:31:22.129574+00:00", "updated": "2026-04-28T22:00:14.183611+00:00", "vendors": ["linux", "linux$PRODUCT$linux_kernel"]}