bind9 bind 9.20.24

[Bruce Dubbs]

New point version.

[Bruce Dubbs]

BIND 9.20.24

Security Fixes

• Fix DNS64 owner case after DNAME restart.

When BIND 9 is configured to use DNS64 and encounters a DNAME redirect, it could end up using freed memory for the DNS response owner name. This caused the response to contain corrupted data. This fix ensures the correct owner name is used when constructing the synthesized response after a DNAME redirect.

New Features

• Enable PR-Agent reviews on merge requests.

Adds a CI job that runs PR-Agent against each merge request opened from the canonical repository, posting an automated review and code-improvement suggestions as MR comments. The job is gated to same-project source branches so the OpenAI key and personal access token are not exposed to fork pipelines.

Removed Features

• Remove ineffective TCP fallback after repeated UDP timeouts.

When an authoritative server failed to respond to two consecutive UDP queries, named marked the next retry as TCP but still sent it over UDP, producing misleading dnstap records. The ineffective retry path has been removed; a corrected TCP fallback will be restored in future BIND 9 versions.

• Remove useless PR-Agent jobs.

The experiment was a failure, the PR-Agent doesn't send a full context to the AI Agents and the results are abysmal because of that.

Feature Changes

• Fall back to TCP on a UDP response with a mismatched query id.

BIND used to wait silently for the correct DNS message id on a UDP fetch even after receiving a response from the expected server with the wrong id, leaving room for off-path spoofing attempts to keep guessing within that window. The resolver now retries the fetch over TCP on the first such response, and a new MismatchTCP statistics counter tracks how often the fallback fires.

• Limit the number of glue records cached from a referral.

When a delegation response contained many glue addresses per listed nameserver, all of them were cached without a per-nameserver bound, inflating resolver cache memory beyond what resolution could ever use. The cache now keeps at most 20 IPv4 and 20 IPv6 glue addresses per nameserver from a delegation.

• Fix a resolver stall on a CNAME response to a DS query.

A validating resolver could stall for about twelve seconds and then return SERVFAIL when an authoritative server answered a DS query with a CNAME. Such responses are now rejected promptly, so the query fails fast instead of hanging.

• Named could crash on concurrent TKEY DELETE for the same key.

On a server configured with tkey-gssapi-keytab (or tkey-gssapi-credential), an authenticated peer could crash named by sending two TKEY DELETE requests for the same dynamic key in rapid succession. This has been fixed.

Bug Fixes

• The resolver now removes other RRsets at the same name when caching a CNAME.

When an RRset is in stale cache, and the authoritative server changes the record type to CNAME, the resolver fails to refresh the stale cache. This has been fixed.

• Fix nxdomain-redirect combined with dns64.

When a resolver was configured with both nxdomain-redirect and dns64 in the same view, an AAAA query for a nonexistent name could abort named. The combination failed whenever the redirect zone held A records but no AAAA records. The server now serves the empty AAAA response from the redirect zone as-is, instead of attempting DNS64 synthesis on top of it.

• Clear REDIRECT flag when it isn't needed.

When nxdomain-redirect is in use, and a recursive query is used to get the redirected answer, a flag is set to distinguish it from a normal recursive response. Previously, that flag was left set afterward, which could trigger an assertion if a normal recursive query was sent later on behalf of the same client: for example, because the filter-aaaa plugin was in use. This has been fixed.

• Fix data race during rndc dumpdb or zone load.

'rndc dumpdb' against a server with zones, and async zone load, had a timing window where the operation's completion could fire before the server had finished registering the operation, occasionally leading to a possible crash. The completion is now delivered after the registration is in place.

• Bound memory use during incoming zone transfers.

During an incoming zone transfer, an optimization could let the batch of pending records grow without bound for a large zone, raising memory usage. It gave no measurable performance benefit, so it has been removed.

• Disable output escaping in bind9.xsl.

The statistics charts where not displaying on some browsers. This has been fixed.

• Fix crash on badly configured secondary signer.

A badly configured secondary signer that was missing the 'file' entry caused the server to crash, rather than to reject the configuration. This has been fixed.

• Avoid named assertion failure during parent-NS lookups when none exist.

Configuring the root zone as a signed primary with parental agents (or with notify-on-cds-changes) caused named to exit on an internal assertion as soon as the DS-publication machinery tried to look up the parent NS RRset — the root has no parent. The lookup is now short-circuited cleanly.

Similar, a zone with no NS records in the parent caused named to exit in the same way.

• Reject RRSIG records covering meta-types.

A recursive resolver could accept and cache an RRSIG record whose Type-Covered field names a meta-type (ANY, AXFR, IXFR, MAILA, MAILB), even though no real RRset of those types ever exists. Such records are now rejected by the DNS message parser.

• Validate nsec3hash arguments instead of relying on atoi()

The nsec3hash tool parsed its algorithm, flags, and iterations arguments with atoi(), then range-checked the result. For values that overflow int during digit-by-digit accumulation, atoi() is undefined; in practice on musl libc the modular wrap leaves n == 0, which silently passes the "iterations > 0xffffU" check. On Alpine Linux this made nsec3hash succeed with iterations treated as 0 for inputs like 4294967296 (2^32).

The latent bug only surfaced when the recent image rebuild pulled in Hypothesis 6.152.9 (2026-05-19), which unified the distribution used for bounded and unbounded integers() strategies. The new smoother distribution explores the 2^{32 boundary on unbounded ranges like integers(min_value=65536); earlier versions did not reach there, so test_nsec3hash_too_many_iterations only started failing on Alpine after the image refresh.}

Replace the three atoi() calls with isc_parse_uint8 / isc_parse_uint16, which uniformly reject overflow, trailing garbage, leading sign, and non-numeric input across libc implementations. As a side effect, error messages now include the offending argument and a specific reason ("out of range" vs "not a valid number").

• Fix wrong variable in named_server_sync() log message.

named_server_sync() logged isc_result_totext(result) but returns tresult. The loop accumulates errors into tresult, so result only holds the last iteration's value. If the last view succeeded but an earlier one failed, the log would incorrectly say "success".

• Refine resolver fetch loop detection.

The resolver's fetch loop detection now triggers only when a new fetch would join an already in-flight fetch that is also one of its own ancestors, which is the actual loop condition. Previously the check ran against the original request before the fetch was set up.

[Bruce Dubbs]

Fixed at commits:

c7f024b4a1 Update to bind/bind-utils-9.20.24.
fb73e794ff Update to samba-4.24.4.
efc9e618d9 Update to webkitgtk-2.52.5.