This release include 11 security fixes:
- cmd/go: malicious module proxy can bypass checksum database
A malicious module proxy could exploit a flaw in the go command's
validation of module checksums to bypass checksum database validation.
This vulnerability affects any user using an untrusted module proxy
(GOMODPROXY) or checksum database (GOSUMDB).
A malicious module proxy can serve altered versions of the Go toolchain.
When selecting a different version of the Go toolchain than the
currently installed toolchain (due to the GOTOOLCHAIN environment variable,
or a go.work or go.mod with a toolchain line), the go command will download
and execute a toolchain provided by the module proxy. A malicious module
proxy can bypass checksum database validation for this downloaded
toolchain.
Since this vulnerability affects the security of toolchain downloads,
setting GOTOOLCHAIN to a fixed version is not sufficient. You must upgrade
your base Go toolchain.
The go tool always validates the hash of a toolchain before executing it,
so fixed versions will refuse to execute any cached, altered versions of the
toolchain.
The go tool trusts go.sum files to contain accurate hashes of the current
module's dependencies. A malicious proxy exploiting this vulnerability to
serve an altered module will have caused an incorrect hash to be recorded
in the go.sum. Users who have configured a non-trusted GOPROXY can determine
if they have been affected by running "rm go.sum ; go mod tidy ; go mod verify",
which will revalidate all dependencies of the current module.
The specific flaw in more detail:
The go command consults the checksum database to validate downloaded modules,
when a module is not listed in the go.sum file. It verifies that the module hash
reported by the checksum database matches the hash of the downloaded module.
If, however, the checksum database returns a successful response that contains
no entry for the module, the go command incorrectly permitted validation to succeed.
A module proxy may mirror or proxy the checksum database, in which case the go
command will not connect to the checksum database directly. Checksums reported
by the checksum database are cryptographically signed, so a malicious proxy
cannot alter the reported checksum for a module. However, a proxy which returns
an empty checksum response, or a checksum response for an unrelated module,
could cause the go command to proceed as if a downloaded module has been validated.
The go command now properly checks checksum database responses to ensure
that the expected module signature is present, not just that if a signature is
present it matches the expectation.
Thanks to Mundur (https://github.com/M0nd0R) for reporting this issue.
This is CVE-2026-42501 and Go issue https://go.dev/issue/79070.
- net/http/httputil: ReverseProxy forwards queries with more than urlmaxqueryparams parameters
When used with a Rewrite function, or a Director function which parses query parameters,
ReverseProxy sanitizes the forwarded request to remove query parameters which are not
parsed by url.ParseQuery. ReverseProxy did not take ParseQuery's limit on the total number
of query parameters (controlled by GODEBUG=urlmaxqueryparams=N) into account.
This could permit ReverseProxy to forward a request containing a query parameter
that was not visible to the Rewrite function.
For example, the query "a1=x&a2=x&...&a10000=x&hidden=y" could forward the parameter
"hidden=y" while hiding it from the proxy's Rewrite function.
ReverseProxy now avoids forwarding parameters that exceed the ParseQuery limit.
This is CVE-2026-39825 and Go issue https://go.dev/issue/78948.
- net: panic in Dial and LookupPort when handling NUL byte on Windows
The Dial and LookupPort functions would panic on Windows when provided
with an input containing a NUL (0). These functions now return an error
rather than panicking.
This is CVE-2026-39836 and Go issue https://go.dev/issue/79006.
- net/mail: quadratic string concatenation in consumePhrase
Pathological inputs could cause DoS through consumePhrase
when parsing an email address according to RFC 5322.
This is CVE-2026-42499 and Go issue https://go.dev/issue/78987.
- net/mail: quadratic string concatentation in consumeComment
Well-crafted inputs reaching ParseAddress, ParseAddressList,
and ParseDate were able to trigger excessive CPU exhaustion
and memory allocations.
This is CVE-2026-39820 and Go issue https://go.dev/issue/78566.
- cmd/go: "go bug" follows symlinks in predictable temporary filenames
The "go bug" command wrote to two files with predictable names in
the system temporary directory (for example, "/tmp").
An attacker with access to the temporary directory could create a
symlink in one of these names, causing "go bug" to overwrite the
target of the symlink.
The "go bug" command now uses os.MkdirTemp to create a safe
working directory.
Thanks to Harshit Gupta (Mr HAX) for reporting this issue.
This is CVE-2026-39819 and Go issue https://go.dev/issue/78584.
- cmd/go: "go tool pack" does not sanitize output paths
The "go tool pack" subcommand is a minimal version of the Unix ar utility.
It is used by the compiler as an internal tool with known-good inputs.
The "pack" subcommand did not sanitize output filenames.
When invoked to extract a malicious archive file, it could write
files to arbitrary locations on the filesystem.
The "pack" subcommand now refuses to extract files with names
containing any directory components.
Thanks to Harshit Gupta (Mr HAX) for reporting this issue.
This is CVE-2026-39817 and Go issue https://go.dev/issue/78778.
- net/http: infinite loop in HTTP/2 transport when given bad SETTINGS_MAX_FRAME_SIZE
When processing HTTP/2 SETTINGS frames, transport will enter an infinite loop of
writing CONTINUATION frames if it receives a SETTINGS_MAX_FRAME_SIZE with a
value of 0.
This allows potential DoS against a client by a malicious server. HTTP/2
transport now properly checks that the received SETTINGS_MAX_FRAME_SIZE is
valid.
Thanks to Marwan Atia (marwansamir688@gmail.com) for reporting this issue.
This is CVE-2026-33814 and Go issue https://go.dev/issue/78476.
- html/template: escaper bypass leads to XSS
If a trusted template author were to write a
tag containing an empty type attribute or a type
attribute with an ASCII whitespace, the execution of
the template would incorrectly escape any data passed
into the block.
Thanks to Mundur (https://github.com/M0nd0R) for reporting this issue.
This is CVE-2026-39826 and Go issue https://go.dev/issue/78981.
- net: crash when handling long CNAME response
When using LookupCNAME with the cgo DNS resolver,
a very long CNAME response could trigger a double-free of C memory
and a crash. The double-free has been fixed.
Thanks to hamayanhamayan for reporting this issue.
This is CVE-2026-33811 and Go issue https://go.dev/issue/78803.
- html/template: bypass of meta content URL escaping causes XSS
CVE-2026-27142 fixed a vulnerability in which URLs were not
correctly escaped inside of a tag's attribute.
If the URL content were to insert ASCII whitespaces around the
= rune inside of the attribute, the escaper would
fail to similarly escape it, leading to XSS.
Dynamic inputs to a tag's attribute are now
whitespace sanitized prior to escaping.
Thanks to Samy Ghannad for reporting this issue.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Several Labels() methods iterated over maps without sorting, producing
non-deterministic output.
In early versions of Go, map iteration order happened to be stable in
practice, so the original code appeared to work correctly.
Since Go 1.12, the runtime intentionally randomizes map iteration order,
making the output unpredictable between runs.
The API response produces sorted labels and the container formatter
already sorted its labels (changed in 5ee17eef).
Apply the same fix to the volume, network, config, and secret
formatters, and update tests to assert exact ordering instead of using
order-independent comparison.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Ensure PRs have a milestone set and that it matches the next release
version declared in the root VERSION file.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Add support for the `bind-create-src` option in bind mounts, which
instructs the daemon to create the source mountpoint on the host if it
doesn't exist.
This allows to replace the legacy `-v /src/dir:/dst` with the `--mount`.
Usage:
--mount type=bind,src=/host/path,dst=/container/path,bind-create-src
--mount type=bind,src=/host/path,dst=/container/path,bind-create-src=true
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This releases includes 6 security fixes following the security policy:
- archive/zip: denial of service when parsing arbitrary ZIP archives
archive/zip used a super-linear file name indexing algorithm that is invoked the first time a file in an archive is opened. This can lead to a denial of service when consuming a maliciously constructed ZIP archive.
Thanks to Thanks to Jakub Ciolek for reporting this issue.
This is CVE-2025-61728 and Go issue https://go.dev/issue/77102.
- net/http: memory exhaustion in Request.ParseForm
When parsing a URL-encoded form net/http may allocate an unexpected amount of
memory when provided a large number of key-value pairs. This can result in a
denial of service due to memory exhaustion.
Thanks to jub0bs for reporting this issue.
This is CVE-2025-61726 and Go issue https://go.dev/issue/77101.
- crypto/tls: Config.Clone copies automatically generated session ticket keys, session resumption does not account for the expiration of full certificate chain
The Config.Clone methods allows cloning a Config which has already been passed
to a TLS function, allowing it to be mutated and reused.
If Config.SessionTicketKey has not been set, and Config.SetSessionTicketKeys has
not been called, crypto/tls will generate random session ticket keys and
automatically rotate them. Config.Clone would copy these automatically generated
keys into the returned Config, meaning that the two Configs would share session
ticket keys, allowing sessions created using one Config could be used to resume
sessions with the other Config. This can allow clients to resume sessions even
though the Config may be configured such that they should not be able to do so.
Config.Clone no longer copies the automatically generated session ticket keys.
Config.Clone still copies keys which are explicitly provided, either by setting
Config.SessionTicketKey or by calling Config.SetSessionTicketKeys.
This issue was discoverd by the Go Security team while investigating another
issue reported by Coia Prant (github.com/rbqvq).
Additionally, on the server side only the expiration of the leaf certificate, if
one was provided during the initial handshake, was checked when considering if a
session could be resumed. This allowed sessions to be resumed if an intermediate
or root certificate in the chain had expired.
Session resumption now takes into account of the full chain when determining if
the session can be resumed.
Thanks to Coia Prant (github.com/rbqvq) for reporting this issue.
This is CVE-2025-68121 and Go issue https://go.dev/issue/77113.
- cmd/go: bypass of flag sanitization can lead to arbitrary code execution
Usage of 'CgoPkgConfig' allowed execution of the pkg-config
binary with flags that are not explicitly safe-listed.
To prevent this behavior, compiler flags resulting from usage
of 'CgoPkgConfig' are sanitized prior to invoking pkg-config.
Thank you to RyotaK (https://ryotak.net) of GMO Flatt Security Inc.
for reporting this issue.
This is CVE-2025-61731 and go.dev/issue/77100.
- cmd/go: unexpected code execution when invoking toolchain
The Go toolchain supports multiple VCS which are used retrieving modules and
embedding build information into binaries.
On systems with Mercurial installed (hg) downloading modules (e.g. via go get or
go mod download) from non-standard sources (e.g. custom domains) can cause
unexpected code execution due to how external VCS commands are constructed.
On systems with Git installed, downloading and building modules with malicious
version strings could allow an attacker to write to arbitrary files on the
system the user has access to. This can only be triggered by explicitly
providing the malicious version strings to the toolchain, and does not affect
usage of @latest or bare module paths.
The toolchain now uses safer VCS options to prevent misinterpretation of
untrusted inputs. In addition, the toolchain now disallows module version
strings prefixed with a "-" or "/" character.
Thanks to splitline (@splitline) from DEVCORE Research Team for reporting this
issue.
This is CVE-2025-68119 and Go issue https://go.dev/issue/77099.
- crypto/tls: handshake messages may be processed at the incorrect encryption level
During the TLS 1.3 handshake if multiple messages are sent in records that span
encryption level boundaries (for instance the Client Hello and Encrypted
Extensions messages), the subsequent messages may be processed before the
encryption level changes. This can cause some minor information disclosure if a
network-local attacker can inject messages during the handshake.
Thanks to Coia Prant (github.com/rbqvq) for reporting this issue.
This is CVE-2025-61730 and Go issue https://go.dev/issue/76443
View the release notes for more information:
https://go.dev/doc/devel/release#go1.25.6
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The --all flag description was misleading by only mentioning
intermediate images, when it actually also controls the visibility of
dangling (untagged) images.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This reverts part of the logic introduced in 207bf52c27 which
incorrectly gated untagged images behind the --all flag in non-expanded
view.
The original fix was addressing the wrong layer of the problem.
The actual issue was that dangling images were being incorrectly passed
to the tree code in the first place.
This was properly fixed in 67f5e3413 which corrected the dangling image
detection logic to properly filter them out before reaching the tree
display code.
Now that dangling images are correctly filtered upstream, untagged
images that reach the tree view should be displayed regardless of the
--all flag setting.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
When calculating column widths for the tree view, untagged images
weren't being properly accounted for in the width calculation.
This caused layout issues when there were tagged images were shorter
than the `<untagged>` string.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The isDangling function was incorrectly identifying images as dangling
when they had no RepoTags but had valid RepoDigests.
This can occur when the graphdrivers are used instead of the containerd
image store.
An image should only be considered dangling if it has no RepoTags,
regardless of whether it has RepoDigests.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Paweł GronowskiandGitHub
