Update the release checklist to incorporate some minor tweaks that we
have been applying manually for the past few months as a result of
release process evolution.
Rework the Security Incident Handling Checklist so that it does not only
contain the SWENG-side steps for handling a security incident, but also
all the other steps required by ISC procedures.
Artifacts of the "stress" jobs in GitLab CI are used for generating the
QA summary for a given set of releases. It happened in the past that
these artifacts were purged before the QA summary was prepared,
unnecessarily prolonging the release process. A complete set of
artifacts from all such jobs for a single pipeline should be less than 1
GB in size, so extend their lifetime from one day to one week as a more
reasonable compromise between availability and disk space usage. (Note:
these jobs are also run in scheduled pipelines, but that is expected to
be acceptable in the context of artifact lifetime.)
Artifacts of the "release" jobs in GitLab CI are what we eventually
publish on the FTP server. It happened in the past that these artifacts
were purged before we copied them to their destination, unnecessarily
prolonging the release process, even though we usually press the "Keep"
button for the "release" jobs manually to keep them around indefinitely
(as they are the source of what we publish in other places). Since
there is only one "release" job per tag pipeline and its artifacts only
take up tens of megabytes of space, keep the artifacts of the "release"
jobs around indefinitely by default.
commit --fixup=amend:<hash> produces a subject starting with amend!
by default. Have danger look for this to ensure that it is squashed
before merging.
The util/release-tarball-comparison.sh script compares a release-ready
BIND 9 tarball to a temporary BIND 9 tarball created from the same
signed Git tag to ensure that their content does not differ
(significantly).
Add check for extracting the public 'n' component on OpenSSL 3.0
path. This is mandatory component, and it's presence is checked
already on the other code path.
Also document the reason why private key component getting errors
are ignored.
This restores the Malloced memory counter and it's now always equal to
InUse counter. This is only for backwards compatibility reason and
there is no separate counter.
The commit also cleanups little things like structure with a single
item (summary.inuse), and shuts up a wrong cppcheck warning (the
notorious NULL check after assignment).
Instead of enforcing stronger synchronization between threads, make all
the atomic operations relaxed. We are not really interested in exact
numbers at all times - the single place where we need the exact number
is when the memory context is being destroyed. Even when there's a
overmem counter, we don't care about exact ordering or exact number.
The Lost memory counter would count the memory "lost" by external
libraries. There's really no such thing as `named` require the memory
contexts to be clean on destroy.
The stats buckets were again more useful for internal allocator, because
we would see the individual "block" caches where the allocations would
fall into. Remove the stats buckets, and if needed, we can pull more
detailed statistics out of the jemalloc.
The total memory counter had again little or no meaning when we removed
the internal memory allocator. It was just a monotonic counter that
would count add the allocation sizes but never subtracted anything, so
it would be just a "big number".
The maxinuse memory counter indicated the highest amount of
memory allocated in the past. Checking and updating this high-
water mark value every time memory was allocated had an impact
on server performance, so it has been removed. Memory size can
be monitored more efficiently via an external tool logging RSS.
The malloced and maxmalloced memory counters were mostly useless since
we removed the internal allocator blocks - it would only differ from
inuse by the memory context size itself.
Both receive_secure_serial() and setnsec3param() run on the same zone
loop, therefore they are serialized. Remove the mechanism to enqueue
the nsec3param and secure serial updates in case one of them is
running (as they can not) and replace it with sanity check.
Previously, the zone loading and dumping was effectively serialized by
the dns_io_t mechanism. In theory, more IO operations could be run in
parallel, but the zone manager .iolimit was set to 1 and never increased
as dns_zonemgr_setiolimit() was never ever called.
As the dns_master asynchronous load and dump was already offloaded to
non-worker threads with isc_work mechanism, drop the whole dns_io_t
and just rely on the isc_work to do the load and dump scheduling.
The signal handler in the isc_loop would wait for all the work to finish
before interrupting the signing. Add teardown handlers via
isc_loopmgr_teardown() to signal the assignwork() it should stop signing
and bail-out early.
NOTE: The dnssec-signzone binary still can't be interrupted during zone
loading, zone cleaning, nsec(3) chain generation or zone writing. This
might get addressed in the future if it becomes a problem.
Instead of dumping the signed zone contents node by node during the
signing, dump the entire zone at the end. This was already done for the
raw zone format, but it shows that the IO is better utilized when the
zone dump is done in one single write rather than in small chunks.
A side effect of dumping node by node was that all names were printed
relative to the zone origin rather than being grouped under different
$ORIGINs as would normally be the case when dumping a zone. Also, state
was not maintained from one node to the next regarding whether the CLASS
has already been printed, so it was always included with the first
record of each node.
Since dnssec-signzone uses the dns_master_style_explicittl text format
style, and is the only application that does so, we can revise that
style and add a new DNS_STYLEFLAG_CLASS_PERNAME flag to get the output
back to what it was before this change.
Use isc_job_run() instead of isc_task_send() for dnssec-signzone
worker threads.
Also fix the issue where the additional assignwork() would be run only
from the main thread effectively serializing all the signing.
The validity default days value of 1 was used for debugging and
left as such accidentally.
Use 10950 days, as used elsewhere (for example, in doth test CA).
This does not affect anything, the value will be effective when
generating new test certificates in the future.
Change the 'forward' system test to enable DoT on ns2 server,
and test that forwarding from ns4 to the DoT-enabled ns2 works.
In order to test different scenarios, create a test CA (based on
similar CAs for 'doth' and 'nsupdate' system tests), and test
both insecure (no certificate validation) and secure (also with
mutual TLS) TLS configurations, as well as a configuration with an
expired certificate.
Add support for loading and validating the 'tls' parameter from
the forwarders' configuration.
This prepares ground for adding support to forward queries to
DoT-enabled upstream servers.
A 'tls' statement can be specified both for individual addresses
and for the whole list (as a default value when an individual
address doesn't have its own 'tls' set), just as it was done
before for the 'port' value.
Create a new function 'print_rawqstring()' to print a string residing
in a 'isc_textregion_t' type parameter.
Create a new function 'copy_string()' to copy a string from a
'cfg_obj_t' object into a 'isc_textregion_t'.
