2
0
mirror of https://gitlab.isc.org/isc-projects/bind9 synced 2025-08-31 14:35:26 +00:00

Reorder release notes

This commit is contained in:
Michał Kępień
2021-04-12 12:15:45 +02:00
parent 3270ac7d9b
commit 4ac14c4a83

View File

@@ -67,21 +67,6 @@ Feature Changes
Bug Fixes
~~~~~~~~~
- After ``rndc checkds -checkds`` or ``rndc dnssec -rollover`` is used,
``named`` now immediately attempts to reconfigure zone keys. This
change prevents unnecessary key rollover delays. [GL #2488]
- Zones using KASP could not be thawed after they were frozen using
``rndc freeze``. This has been fixed. [GL #2523]
- CDS/CDNSKEY DELETE records are now removed when a zone transitions
from a secure to an insecure state. ``named-checkzone`` also no longer
reports an error when such records are found in an unsigned zone.
[GL #2517]
- ``named`` crashed after skipping a primary server while transferring a
zone over TLS. This has been fixed. [GL #2562]
- TCP idle and initial timeouts were being incorrectly applied: only the
``tcp-initial-timeout`` was applied on the whole connection, even if
the connection were still active, which could prevent a large zone
@@ -102,6 +87,21 @@ Bug Fixes
been fixed. If a corrupt journal file is detected, ``named`` can now
recover from it. [GL #2600]
- CDS/CDNSKEY DELETE records are now removed when a zone transitions
from a secure to an insecure state. ``named-checkzone`` also no longer
reports an error when such records are found in an unsigned zone.
[GL #2517]
- Zones using KASP could not be thawed after they were frozen using
``rndc freeze``. This has been fixed. [GL #2523]
- After ``rndc checkds -checkds`` or ``rndc dnssec -rollover`` is used,
``named`` now immediately attempts to reconfigure zone keys. This
change prevents unnecessary key rollover delays. [GL #2488]
- ``named`` crashed after skipping a primary server while transferring a
zone over TLS. This has been fixed. [GL #2562]
- When dumping the cache to file, TTLs were being increased with
``max-stale-ttl``. Also the comment above stale RRsets could have nonsensical
values if the RRset was still marked a stale but the ``max-stale-ttl`` has