Varnish Cache before 7.6.2 and Varnish Enterprise before 6.0.13r10 allow client-side desync via HTTP/1 requests.

HTTP/2 Rapid reset attack The HTTP/2 protocol allows clients to indicate to the server that a previous stream should be canceled by sending a RSTSTREAM frame. The protocol does not require the client and server to coordinate the cancellation in any way, the client may do it unilaterally. The client may also assume that the cancellation will take effect immediately when the server receives the RSTSTREAM frame, before any other data from that TCP connection is processed.

Abuse of this feature is called a Rapid Reset attack because it relies on the ability for an endpoint to send a RSTSTREAM frame immediately after sending a request frame, which makes the other endpoint start working and then rapidly resets the request. The request is canceled, but leaves the HTTP/2 connection open.

The HTTP/2 Rapid Reset attack built on this capability is simple: The client opens a large number of streams at once as in the standard HTTP/2 attack, but rather than waiting for a response to each request stream from the server or proxy, the client cancels each request immediately.

The ability to reset streams immediately allows each connection to have an indefinite number of requests in flight. By explicitly canceling the requests, the attacker never exceeds the limit on the number of concurrent open streams. The number of in-flight requests is no longer dependent on the round-trip time (RTT), but only on the available network bandwidth.

In a typical HTTP/2 server implementation, the server will still have to do significant amounts of work for canceled requests, such as allocating new stream data structures, parsing the query and doing header decompression, and mapping the URL to a resource. For reverse proxy implementations, the request may be proxied to the backend server before the RSTSTREAM frame is processed. The client on the other hand paid almost no costs for sending the requests. This creates an exploitable cost asymmetry between the server and the client.

Multiple software artifacts implementing HTTP/2 are affected. This advisory was originally ingested from the swift-nio-http2 repo advisory and their original conent follows.

swift-nio-http2 specific advisory swift-nio-http2 is vulnerable to a denial-of-service vulnerability in which a malicious client can create and then reset a large number of HTTP/2 streams in a short period of time. This causes swift-nio-http2 to commit to a large amount of expensive work which it then throws away, including creating entirely new Channels to serve the traffic. This can easily overwhelm an EventLoop and prevent it from making forward progress.

swift-nio-http2 1.28 contains a remediation for this issue that applies reset counter using a sliding window. This constrains the number of stream resets that may occur in a given window of time. Clients violating this limit will have their connections torn down. This allows clients to continue to cancel streams for legitimate reasons, while constraining malicious actors.

An HTTP Request Forgery issue was discovered in Varnish Cache 5.x and 6.x before 6.0.11, 7.x before 7.1.2, and 7.2.x before 7.2.1. An attacker may introduce characters through HTTP/2 pseudo-headers that are invalid in the context of an HTTP/1 request line, causing the Varnish server to produce invalid HTTP/1 requests to the backend. This could, in turn, be used to exploit vulnerabilities in a server behind the Varnish server. Note: the 6.0.x LTS series (before 6.0.11) is affected.

An issue was discovered in Varnish Cache 7.x before 7.1.2 and 7.2.x before 7.2.1. A request smuggling attack can be performed on Varnish Cache servers by requesting that certain headers are made hop-by-hop, preventing the Varnish Cache servers from forwarding critical headers to the backend.

In Varnish Cache 7.0.0, 7.0.1, 7.0.2, and 7.1.0, it is possible to cause the Varnish Server to assert and automatically restart through forged HTTP/1 backend responses. An attack uses a crafted reason phrase of the backend response status line. This is fixed in 7.0.3 and 7.1.1.

In Varnish Cache before 6.6.2 and 7.x before 7.0.2, Varnish Cache 6.0 LTS before 6.0.10, and and Varnish Enterprise (Cache Plus) 4.1.x before 4.1.11r6 and 6.0.x before 6.0.9r4, request smuggling can occur for HTTP/1 connections.

Varnish Cache, with HTTP/2 enabled, allows request smuggling and VCL authorization bypass via a large Content-Length header for a POST request. This affects Varnish Enterprise 6.0.x before 6.0.8r3, and Varnish Cache 5.x and 6.x before 6.5.2, 6.6.x before 6.6.1, and 6.0 LTS before 6.0.8.

Varnish HTTP cache before 3.0.4: ACL bug

An issue was discovered in Varnish Cache before 6.0.4 LTS, and 6.1.x and 6.2.x before 6.2.1. An HTTP/1 parsing failure allows a remote attacker to trigger an assert by sending crafted HTTP/1 requests. The assert will cause an automatic restart with a clean cache, which makes it a Denial of Service attack.

vbfstperror in bin/varnishd/cache/cachefetch.c in Varnish HTTP Cache 4.1.x before 4.1.9 and 5.x before 5.2.1 allows remote attackers to obtain sensitive information from process memory because a VFPGetStorage buffer is larger than intended in certain circumstances involving -sfile Stevedore transient objects.

An issue was discovered in Varnish HTTP Cache 4.0.1 through 4.0.4, 4.1.0 through 4.1.7, 5.0.0, and 5.1.0 through 5.1.2. A wrong if statement in the varnishd source code means that particular invalid requests from the client can trigger an assert, related to an Integer Overflow. This causes the varnishd worker process to abort and restart, losing the cached contents in the process. An attacker can therefore crash the varnishd worker process on demand and effectively keep it from serving content - a Denial-of-Service attack. The specific source-code filename containing the incorrect statement varies across releases.

Varnish 3.x before 3.0.7, when used in certain stacked installations, allows remote attackers to inject arbitrary HTTP headers and conduct HTTP response splitting attacks via a header line terminated by a \r (carriage return) character in conjunction with multiple Content-Length headers in an HTTP request.

varnish 3.0.3 uses world-readable permissions for the /var/log/varnish/ directory and the log files in the directory, which allows local users to obtain sensitive information by reading the files. NOTE: some of these details are obtained from third party information.

Varnish before 3.0.5 allows remote attackers to cause a denial of service (child-process crash and temporary caching outage) via a GET request with trailing whitespace characters and no URI.