Since we use golang 1.12 and gomod, it is not necessary
to match the impor path to `/go/src/github.com/coredns/coredns`
in order to build coredns. For that reason the build command
inside docker could be simplified with:
```
$ docker run --rm -i -t -v $PWD:/v -w /v golang:1.12 make
```
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
This feature introduce continuous fuzzing with the following
features:
* Ruzzing: fuzz-targets are run continuously on master
( the fuzzers are updated every time new code is pushed to master)
* Regresion: In addition to unit-tests travis runs all fuzz
targets through the generated corpus to catch bugs early on
in the CI process before merge.
* plugin/clouddns: tiny fixes for the README
Did a post-merge review. Fix a few typos.
Signed-off-by: Miek Gieben <miek@miek.nl>
* Naming;
coredns -> CoreDNS
clouddns -> Cloud DNS
and italics then the plugin's name are mentioned.
Signed-off-by: Miek Gieben <miek@miek.nl>
This PR adds some content in clouddns plugin to mention about
the fact that privately hosted zone does not need to attach to
a VPC.
Also change PROJECT_NAME to PROJECTT_ID, and reformt the markdown
to replace tab with 4 whitespace.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Looks like there were some leftover in PR 3011 during
the merge conflict rebase. This PR removes the unneeded file
and fixed the plugin.cfg
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
In the setup function use plugin.Error() to wrap the errors with the
plugin name. Because there isn't a separate setup() function this is
done for all returned errors.
Remove *upstream.Upstream from the New parameters as this is always set
and adjust the tests to account for this.
Signed-off-by: Miek Gieben <miek@miek.nl>
* doc: formatting and point to plugin.md
Slight tweaks in CONTRIBUTING.md and point to plugin.md which also has a
bunch of info.
Signed-off-by: Miek Gieben <miek@miek.nl>
* Update plugin.md
Co-Authored-By: Chris O'Haver <cohaver@infoblox.com>
* Add minimize to configuration knobs
Signed-off-by: Miek Gieben <miek@miek.nl>
* more
Signed-off-by: Miek Gieben <miek@miek.nl>
* fix switch order
* remove fallthough
* add test
* fix test
* distingush nxdomain/nodata for at least first subdomain of zone
* restore fallthough; reorder switch cases
document the environment option and some cleanups. Go over the code and
fix/tweak random bits here and there.
Condense a few lines here and there.
Signed-off-by: Miek Gieben <miek@miek.nl>
* Add plugin for Azure DNS
Signed-off-by: darshanime <deathbullet@gmail.com>
* Rename AzureDNS plugin to Azure
Signed-off-by: darshanime <deathbullet@gmail.com>
* remove upstream from azure syntax
Signed-off-by: darshanime <deathbullet@gmail.com>
* Rename azure plugin block keynames
Signed-off-by: darshanime <deathbullet@gmail.com>
* Normalize zone name before lookup in zones
Signed-off-by: darshanime <deathbullet@gmail.com>
* Update import path for caddy
Signed-off-by: darshanime <deathbullet@gmail.com>
* normalize azure zone name only if required
Signed-off-by: darshanime <deathbullet@gmail.com>
* Add support for MX, SRV, TXT, records
Signed-off-by: darshanime <deathbullet@gmail.com>
* Add specs for new record types
Signed-off-by: darshanime <deathbullet@gmail.com>
* Use sequential updates for zones
Signed-off-by: darshanime <deathbullet@gmail.com>
* Add OWNERS file for azure plugin
Signed-off-by: darshanime <deathbullet@gmail.com>
* Rename imports for third party packages
Signed-off-by: darshanime <deathbullet@gmail.com>
* Capitalize values in README
Signed-off-by: darshanime <deathbullet@gmail.com>
* Shorten keys for azure plugin config
Signed-off-by: darshanime <deathbullet@gmail.com>
* Fixup readme for azure plugin
Signed-off-by: darshanime <deathbullet@gmail.com>
the current update frequency for the refresh loop in the route 53 plugin is hard-coded
to 1 minute. aws rate-limits the number of api requests so less frequent record refreshes
can help when reaching those limits depending upon your individual scenarios. this pull
adds a configuration option to the route53 plugin to adjust the refresh frequency.
thanks for getting my last pull released so quickly. this is the last local change that
i have been running and would love to get it contributed back to the project.
Signed-off-by: Matt Kulka <mkulka@parchment.com>
Various fixes to make things less flaky:
* kubernetes: put klog.SetOutput in the setup function, not in the init
function to see if that helps
* file: make z.Expired a boolean instead of a pointer to a boolean
* test: fix TestSecondaryZoneTransfer test, which wasn't actually
testing in the right way. It's more right now, but may still be racy
(race introduced because a file's lazy loading of zones)
Signed-off-by: Miek Gieben <miek@miek.nl>
* plugin/hosts: create inline map in setup
The inline map wasn't create in the setup.go fuction leading to a crash,
which is masked by a recover (but leads to a SERVFAIL, and not logging
the request).
Various other simplifications.
host plugin could use some test that actually cover these edgecases.
Signed-off-by: Miek Gieben <miek@miek.nl>
* PR review changes
Signed-off-by: Miek Gieben <miek@miek.nl>
without the paging parameter set, it will default to 100 records per request. with large enough zones and
potentially multiple coredns daemons configured to pull from route 53, this can quickly add up and reach
aws global api rate limits (5/sec per root account). increasing paging to max can help reduce the number
of requests needed to pull records for a zone without no down side that i am aware of.
this helps issue #2353, but probably is not a complete fix.
Signed-off-by: Matt Kulka <mkulka@parchment.com>
These are too hidden now. They increase the issue-load, because people
don't see them.
Add log.Errorf in the core/dnsserver recover routine.
Signed-off-by: Miek Gieben <miek@miek.nl>
Move github.com/openzipkin/zipkin-go-opentracing => github.com/openzipkin-contrib/zipkin-go-opentracing
and run `go mod tidy`
Fixes 3069
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
