Vulnerabilidades

*** Pendiente de traducción *** pgAdmin 4&amp;#39;s Backup tool appended the client-supplied &amp;#39;database&amp;#39; field from the /backup/job//object request to the pg_dump argument vector as a bare trailing positional argument, without validation. Because pg_dump parses its options with getopt_long, which permutes arguments, a value beginning with a dash was interpreted as an option rather than as a database name. A value such as --file=/absolute/path therefore overrode the storage-confined --file that pgAdmin had constructed earlier, causing pg_dump to write its output anywhere the pgAdmin process could write, outside the user&amp;#39;s File Manager storage directory. This yields arbitrary file creation and overwrite as the operating-system account running pgAdmin, which can destroy pgAdmin&amp;#39;s own configuration database and, depending on the target chosen, be escalated further.<br /> <br /> The same field additionally permitted connection-string injection. libpq expands a database name containing an equals sign into a full connection string, and keywords embedded there override the --host and --port that pgAdmin passes, so a value such as &amp;#39;host=attacker.example port=5432 dbname=x&amp;#39; redirected pg_dump to a server of the attacker&amp;#39;s choosing. Because pgAdmin exports the decrypted stored database password in the PGPASSWORD environment variable before executing the utility, the redirected connection carries that credential to the attacker-nominated endpoint. Both behaviours are reachable by any authenticated user holding the tools_backup permission, which is granted to the default User role.<br /> <br /> The fix stops passing the database name through the argument vector altogether and supplies it in the PGDATABASE environment variable, which libpq treats as a literal database name and never expands as a connection string. This matches the approach already used by the Import/Export tool. Regression tests assert that the database name is absent from the constructed argument vector and that PGDATABASE carries the exact requested value.<br /> <br /> This issue affects pgAdmin 4: from the introduction of the trailing positional database argument in the Backup tool before 9.18.

*** Pendiente de traducción *** pgAdmin 4&amp;#39;s File Manager save_file endpoint, which backs saving from the Query Tool and ERD, validated the requested path with Filemanager.check_access_permission() and then opened the file for writing with a plain open() call. CVE-2026-7819 had previously hardened the separate file upload path by opening its target with O_NOFOLLOW, so that the kernel refuses to follow a symbolic link at the final path component, but save_file was left on an unprotected open(). A symbolic link already present when the check ran was rejected by the check itself; the remaining exposure was a link substituted at the final path component in the interval between the check and the write, at which point the write followed the link and landed outside the user&amp;#39;s storage directory, creating or overwriting an arbitrary file as the operating-system account running pgAdmin.<br /> <br /> Exploitation requires the ability to create or replace a symbolic link inside the requesting user&amp;#39;s storage directory. pgAdmin exposes no interface that creates symbolic links, so this requires filesystem access to the pgAdmin host or to a shared or network-mounted storage backend, together with winning the timing window. The reporter did not achieve a write against a released version, having attempted approximately 63,000 racing requests without success; the reported defect is the uncovered write path rather than a demonstrated sandbox escape.<br /> <br /> The fix routes save_file&amp;#39;s write through the same O_NOFOLLOW-protected helper already used by the upload path, so both File Manager write sinks carry identical kernel-enforced protection, and reports a symbolic-link refusal explicitly instead of surfacing the underlying system error text. Note that files created by save_file are now given mode 0600, inheriting the behaviour introduced for uploads. The protection covers the final path component; substitution of an intermediate directory component is not addressed by this change.<br /> <br /> This issue affects pgAdmin 4: from the introduction of the containment check in the File Manager save path before 9.18.

*** Pendiente de traducción *** pgAdmin 4&amp;#39;s Restore and Maintenance tools passed the client-supplied &amp;#39;database&amp;#39; field directly as the value of the --dbname option given to pg_restore and psql. libpq expands a database name containing an equals sign into a full connection string, and connection keywords embedded in that value take precedence over the --host and --port arguments that pgAdmin supplies. A value such as &amp;#39;host=attacker.example port=5432 dbname=x&amp;#39; therefore redirected the utility to a server chosen by the requesting user rather than the server the operation was invoked against. Because pgAdmin exports the decrypted stored database password in the PGPASSWORD environment variable before executing the utility, the redirected connection presents that credential to the attacker-nominated endpoint, which may capture it. The redirection additionally permits outbound connections from the pgAdmin host to arbitrary network addresses, including hosts not otherwise reachable by the requesting user.<br /> <br /> The behaviour is reachable by any authenticated user holding the tools_restore or tools_maintenance permission, both of which the default User role grants. The Maintenance tool was not affected in the earliest releases, where the value was wrapped by a quoting helper that incidentally prevented expansion; it became affected when that wrapper was removed.<br /> <br /> The fix supplies the target database in the PGDATABASE environment variable, which libpq treats as a literal database name and never expands as a connection string. Where pg_restore requires a --dbname argument to be present, an empty value is passed, which contains no equals sign and is therefore not expanded, while the real name is taken from the environment.<br /> <br /> This issue affects pgAdmin 4: from the introduction of the --dbname argument in the Restore and Maintenance tools before 9.18.

*** Pendiente de traducción *** pgAdmin 4&amp;#39;s Webserver authentication source is intended to accept an identity asserted by the web server or reverse proxy in front of pgAdmin, delivered through the WSGI/CGI environment. WebserverAuthentication.get_user() read config.WEBSERVER_REMOTE_USER from request.environ and, when that returned nothing, fell back to reading the same name directly from the inbound HTTP request headers via request.headers.get(). An inbound HTTP header is written by whoever sends the request, so any client able to reach pgAdmin could supply that header itself and be authenticated as any username it named, including an existing Administrator, without presenting a password or any other credential. The environment lookup could also be satisfied by a client-supplied header whenever WEBSERVER_REMOTE_USER was configured to an HTTP_-prefixed or hyphenated name such as HTTP_X_FORWARDED_USER or X-Forwarded-User, since WSGI servers place inbound headers into the environment under exactly those names. Deployments are affected only when &amp;#39;webserver&amp;#39; is enabled in AUTHENTICATION_SOURCES.<br /> <br /> The fix distinguishes a genuine CGI/WSGI variable from a header-derived one and implicitly trusts only the former. A header-asserted identity is now accepted only when the operator explicitly opts in via WEBSERVER_REMOTE_USER_FROM_HEADER, the request arrives from a peer listed in WEBSERVER_TRUSTED_PROXIES, and, when configured, a shared secret supplied in WEBSERVER_SHARED_SECRET_HEADER matches WEBSERVER_SHARED_SECRET under a constant-time comparison. The trusted-peer check deliberately reads the real socket peer address rather than request.remote_addr, because ProxyFix rewrites the latter from the client-controlled X-Forwarded-For header and would otherwise allow an attacker to claim to be the trusted proxy. As defence in depth, login() now refuses any account whose auth_source is not &amp;#39;webserver&amp;#39;, so a misconfigured trust gate cannot be used to assume an internal or LDAP account.<br /> <br /> This issue affects pgAdmin 4: from 6.2 before 9.18.

*** Pendiente de traducción *** libp2p is a JavaScript implementation of the libp2p networking stack. From 8.0.0 until 12.0.24, @libp2p/peer-store in packages/peer-store/src/index.ts uses consumePeerRecord to verify a RecordEnvelope signature but does not require PeerRecord.peerId in the signed payload to equal the signer peer ID derived by RecordEnvelope.openAndCertify. The expectedPeer option checks only the envelope signer, and the gossipsub Peer Exchange path can provide the attacker&amp;#39;s own peer ID as expectedPeer. An attacker can therefore sign a record with the attacker&amp;#39;s key, place a victim peer ID and attacker-controlled multiaddrs in the payload, and have certified addresses stored for the victim. The poisoned addresses can cause address-book corruption, dial redirection or failure, routing manipulation, and reachability disruption, although the connection upgrade still verifies remote peer identity and prevents a complete identity takeover. The issue is fixed in version 12.0.24.

*** Pendiente de traducción *** libp2p is a JavaScript implementation of the libp2p networking stack. Prior to 11.0.26, @libp2p/floodsub accepts unauthenticated RPC frames on /floodsub/1.0.0 through PeerStreams.attachInboundStream in packages/floodsub/src/peer-streams.ts without protobuf element limits, then processRpc and processRpcSubOpt in packages/floodsub/src/floodsub.ts synchronously process the subscriptions array without a per-frame cap. A single bounded-size frame can decode into millions of empty subscription entries that block the event loop, while hundreds of thousands of unique-topic SUBSCRIBE entries allocate PeerSet objects in this.topics that are not removed after peer removal or stop. Empty entries cause CPU exhaustion but do not grow this.topics; persistent memory growth requires unique topics. The subscription path bypasses message signature validation and the message-only processing queue, allowing a remote peer to cause sustained CPU denial of service, memory exhaustion, out-of-memory termination, and node unavailability. The issue is fixed in version 11.0.26.

*** Pendiente de traducción *** The AsyncHttpClient (AHC) library allows Java applications to easily execute HTTP requests and asynchronously process HTTP responses. From 2.1.0 until 2.16.1 and 3.0.12, requests using an authenticated SOCKS proxy can expose the proxy&amp;#39;s credentials to the origin because NettyRequestFactory and NettyRequestSender attach Proxy-Authorization without confirming that the request is being sent to an HTTP proxy. With preemptive proxy authentication, the header is attached to a plaintext HTTP request, exposing credentials such as directly reversible Basic credentials to the origin. With the default non-preemptive flow, a hostile origin can return a 407 response and ProxyUnauthorized407Interceptor sends the proxy credentials through the existing SOCKS tunnel, including NTLM, Kerberos, and SPNEGO credentials. Releases before 2.1.0 lack SOCKS proxy support. This issue is fixed in versions 2.16.1 and 3.0.12.

*** Pendiente de traducción *** Soup Sieve is a CSS selector library designed to be used with Beautiful Soup 4. Prior to 2.9, selector_iter in src/soupsieve/css_parser.py trims the raw selector with RE_WS_END, an end-anchored WSC whitespace-and-comment expression used with search(), so the regular expression engine retries a greedy scan at every starting offset. An attacker-controlled valid selector containing a long internal whitespace run, or a selector containing a long CSS comment run followed by another token, causes quadratic CPU work before tokenization. User-controlled selectors can reach the path through soupsieve.compile() and BeautifulSoup.select(), while applications using only hard-coded selectors are unaffected. This root cause is separate from the IDENTIFIER and VALUE backtracking vulnerability because the cost occurs in RE_WS_END.search during trimming rather than token matching. The resulting CPU consumption can hold the Python GIL, exhaust workers, and stall a service without causing memory corruption or code execution. The issue is fixed in version 2.9.

*** Pendiente de traducción *** The AsyncHttpClient (AHC) library allows Java applications to easily execute HTTP requests and asynchronously process HTTP responses. From 3.0.8 until 3.0.12, a client with maxConnections or maxConnectionsPerHost set above zero leaks one connection permit whenever TLS connection establishment fails before the handshake completes. NettyConnectListener removes the partitionKeyLock permit from NettyResponseFuture before every failure path is bound to the channel closeFuture, so an abort can leave the permit unreleased. Repeated failures can permanently lock out one host under a per-host limit or drain the shared pool under a global limit, blocking later requests even when no connection remains open. The default unlimited connection setting is not affected. This issue is fixed in version 3.0.12.

*** Pendiente de traducción *** The AsyncHttpClient (AHC) library allows Java applications to easily execute HTTP requests and asynchronously process HTTP responses. From 2.0.0 until 2.16.1 and 3.0.12, automatic response decompression on the HTTP/1.1 path uses ChannelManager.newHttpContentDecompressor() to install Http1ContentDecompressor without a cumulative output-size limit. A hostile or compromised server, or an attacker who can alter a response in transit, can send a small gzip, deflate, or snappy response that expands across chunks until the client exhausts its heap and raises OutOfMemoryError; brotli and zstd are also affected when their optional codecs are present. In versions 3.0.8 through 3.0.10, the HTTP/2 decompressor is also unbounded, so switching protocols does not mitigate the issue on those releases. A limit applied to each decode call is insufficient because the response can be delivered as many small chunks, so the fixed implementation tracks total decompressed bytes for the whole response. This issue is fixed in versions 2.16.1 and 3.0.12.

*** Pendiente de traducción *** Soup Sieve is a CSS selector library designed to be used with Beautiful Soup 4. Prior to 2.9, the selector parser in src/soupsieve/css_parser.py defines IDENTIFIER with adjacent quantified groups over overlapping character classes, and VALUE embeds IDENTIFIER for attribute selectors. When an attacker-controlled selector contains a long identifier or unquoted attribute-value run followed by input that makes the overall match fail, the regular expression engine explores quadratically many splits between the overlapping groups. User-controlled selectors can reach this path through soupsieve.compile(), soupsieve.select(), or BeautifulSoup.select(), while applications using only hard-coded selectors are unaffected. The resulting CPU consumption can hold the Python GIL, exhaust application workers, and stall a service; successful plain identifier matches are linear, and the issue does not cause memory corruption or code execution. The issue is fixed in version 2.9.

*** Pendiente de traducción *** libp2p is a JavaScript implementation of the libp2p networking stack. From 15.0.0 until 16.0.5, @libp2p/gossipsub uses the default StrictSign policy in packages/gossipsub/src/utils/buildRawMessage.ts, where validateToRawMessage verifies a signature with attacker-controlled msg.key but skips binding that key to msg.from when the claimed author is an RSA peer ID that does not inline a public key. An unauthenticated attacker can place a victim RSA peer ID in msg.from, sign the message with the attacker&amp;#39;s private key, and supply the attacker&amp;#39;s public key in msg.key, causing the message to be accepted and propagated as authored by the victim. Applications that trust message.from for validators, authorization, accounting, moderation, reputation, or audit logging can process attacker-controlled data under false origin attribution. The issue is fixed in version 16.0.5.