Debian Patches
Status for netty/1:4.1.48-12
| Patch | Description | Author | Forwarded | Bugs | Origin | Last update |
|---|---|---|---|---|---|---|
| 01-ignore-npn.patch | Disable the NPN/ALPN support since it relies on jetty npn-api and alpn-api which aren't available in Debian | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 03-ignore-jboss-marshalling.patch | Disable the JBoss Marshalling support (not in Debian) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 04-netty-all-light.patch | Turn the netty-all jar into an empty jar to optimize the size of the package but keep the dependencies on the other artifacts | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 05-reproducible-versions-properties.patch | Make the versions.properties files reproducible | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 06-remove-tcnative-classifier.patch | Removes the empty classifier for the tcnative dependency since it breaks the Gradle dependencies resolution (seen with the projectreactor package) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 07-netty-all-epoll-dependency.patch | Moves the netty-transport-native-epoll out of the profiles since it doesn't work with Gradle (required for the projectreactor package) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 08-codegen-without-groovy-plugin.patch | Adapts codegen.groovy to run without the groovy-maven-plugin (not in Debian yet) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 09-ignore-lz4.patch | Disables lz4 support (missing dependency) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 10-ignore-lzma.patch | Disables lzma support (missing dependency) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 11-ignore-protobuf-nano.patch | Disables protobuf nano support (missing dependency) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 13-ignore-conscrypt.patch | Disables Conscrypt support (missing dependency) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 15-disable-Werror.patch | Build native parts without Werror as that can cause build failures | Sjoerd Simons <sjoerd.simons@collabora.co.uk> | no | debian | ||
| 16-disable-substratevm-support.patch | Disables SubstrateVM support (missing dependencies) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| 17-disable-blockhound-integration.patch | Disables BlocHound integration (missing dependency) | Emmanuel Bourg <ebourg@apache.org> | not-needed | |||
| CVE-2021-21290.patch | CVE-2021-21290 | Markus Koschany <apo@debian.org> | no | https://github.com/netty/netty/commit/c735357bf29d07856ad171c6611a2e1a0e0000ec | 2021-02-12 | |
| CVE-2021-21295.patch | CVE-2021-21295 | Markus Koschany <apo@debian.org> | no | https://github.com/netty/netty/commit/89c241e3b1795ff257af4ad6eadc616cb2fb3dc4 | 2021-03-26 | |
| CVE-2021-21409.patch | CVE-2021-21409 (was: [PATCH] Merge pull request from GHSA-f256-j965-7f32) Motivation: We also need to ensure that all the header validation is done when a single header with the endStream flag is received Modifications: - Adjust code to always enforce the validation - Add more unit tests Result: Always correctly validate |
Norman Maurer <norman_maurer@apple.com> | no | https://github.com/netty/netty/commit/b0fa4d5aab4215f3c22ce6123dd8dd5f38dc0432 | 2021-03-30 | |
| 21-java-17.patch | compile with JDK 17 | no | debian | |||
| CVE-2021-37136.patch | CVE-2021-37136 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/41d3d61a61608f2223bb364955ab2045dd5e4020 | 2023-01-01 |
| CVE-2021-37137.patch | CVE-2021-37137 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/6da4956b31023ae967451e1d94ff51a746a9194f | 2023-01-01 |
| CVE-2021-43797.patch | CVE-2021-43797 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/07aa6b5938a8b6ed7a6586e066400e2643897323 | 2023-01-01 |
| CVE-2022-41881.patch | CVE-2022-41881 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/cd91cf3c99123bd1e53fd6a1de0e3d1922f05bb2 | 2023-01-01 |
| CVE-2022-41915.patch | CVE-2022-41915 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/fe18adff1c2b333acb135ab779a3b9ba3295a1c4 | 2023-01-01 |
| CVE-2023-34462.patch | CVE-2023-34462 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/535da17e45201ae4278c0479e6162bb4127d4c32 | 2023-11-05 |
| CVE-2023-44487.patch | CVE-2023-44487 | Markus Koschany <apo@debian.org> | no | debian | https://github.com/netty/netty/commit/58f75f665aa81a8cbcf6ffa74820042a285c5e61 | 2023-11-05 |
| 22-java-21.patch | Java 21 compatibility patch Use reflection to access sun.x509 API due to the breaking changes between Java 17 and Java 21. Cherry-pick self-signed certificate generator changes from netty-4.1.91.Final. |
Norman Maurer <norman_maurer@apple.com> | yes | debian upstream | 2023-11-28 | |
| CVE-2024-29025.patch | CVE-2024-29025 | Markus Koschany <apo@debian.org> | no | https://github.com/netty/netty/commit/0d0c6ed782d13d423586ad0c71737b2c7d02058c | 2024-05-12 | |
| CVE-2025-59419.patch | CVE-2025-59419: Merge commit from fork * Patch 1 of 3 * Patch 2 of 3 * Patch 3 of 3 * Fix indentation style * Update 2025 * Optimize allocations * Update codec-smtp/src/main/java/io/netty/handler/codec/smtp/SmtpUtils.java |
DepthFirst Disclosures <disclosures@depthfirst.com> | no | 2025-10-14 | ||
| CVE-2025-55163_before-1.patch | commit 9b80d081ff3478c46152b012ae0e21f939467ac3 Only enable the RST limit for servers by default (#13671) Motivation: We dont need to limit the number of RST frames per connection when we are bulding a codec for the client side. Modifications: Dont limit the numbers of RST frames per connection when building a codec for the client side. Result: Only add limit where needed |
Norman Maurer <norman_maurer@apple.com> | no | backport, https://github.com/netty/netty/commit/9b80d081ff3478c46152b012ae0e21f939467ac3 | 2023-10-28 | |
| CVE-2025-55163_1.patch | =?utf-8?q?HTTP2=3A_Http2ConnectionHandler_should_always_use_Http2C?= =?utf-8?q?onnectionEncode=E2=80=A6_=28=2315518=29?= …r (#15516) Motivation: We sometimes directly used the Http2FrameWriter which is not correct as someone might have supplied a custom Http2ConnectionEncoder Modifications: Use Http2ConnectionEncoder when writing RST frames Result: Don't by-pass Http2ConnectionEncoder |
Norman Maurer <norman_maurer@apple.com> | yes | upstream | backport, https://github.com/netty/netty/commit/be53dc3c9acd9af2e20d0c3c07cd77115a594cf1 | 2025-07-28 |
| CVE-2025-55163_2.patch | [PATCH] Merge commit from fork * Enforce the maximum number of RST frames that can be sent in window of time Motivation: A remote peer might be able to trigger an instance to generate and send RST frames by sending invalid frames on an existing stream. This can cause high resource usage and so might be abused by a remote peer. Modifications: Limit the number of RSTs that we allow to be generated and so send in a specific time window. If this limit is reached a GO_AWAY frame is send and the connection be closed. Result: Fix high resource usage that can be caused by a remote peer by trigger RST frames * Adjust testing * Address comments |
Norman Maurer <norman_maurer@apple.com> | yes | upstream | backport, https://github.com/netty/netty/commit/009bd17b38a39fb1eecf9d22ea8ae8108afaac59 | 2025-08-13 |
| CVE-2025-58057.patch | [PATCH] Merge commit from fork (#15612) Motivation: We should ensure our decompressing decoders will fire their buffers through the pipeliner as fast as possible and so allow the user to take ownership of these as fast as possible. This is needed to reduce the risk of OOME as otherwise a small input might produce a large amount of data that can't be processed until all the data was decompressed in a loop. Beside this we also should ensure that other handlers that uses these decompressors will not buffer all of the produced data before processing it, which was true for HTTP and HTTP2. Modifications: - Adjust affected decoders (Brotli, Zstd and ZLib) to fire buffers through the pipeline as soon as possible - Adjust HTTP / HTTP2 decompressors to do the same - Add testcase. Result: Less risk of OOME when doing decompressing |
Norman Maurer <norman_maurer@apple.com> | no | backport, https://github.com/netty/netty/commit/34894ac73b02efefeacd9c0972780b32dc3de04f | 2025-09-03 |
All known versions for source package 'netty'
- 1:4.1.48-13 (sid)
- 1:4.1.48-12 (forky)
- 1:4.1.48-10 (trixie)
- 1:4.1.48-7+deb12u1 (bookworm, bookworm-security)
