Bump to coredns/coredns v1.3.1 #6
Conversation
Signed-off-by: Miek Gieben <miek@miek.nl>
Cache IP's and ports as well. Signed-off-by: Miek Gieben <miek@miek.nl>
This revert 17d807f and re-adds the metadata plugin as a plugin that just sets a label to a value function. Add package documentation on how to use the metadata package. Make it clear that any caching is up to the Func implemented. There are now - no in tree users. We could add the request metadata by default under names that copy request.Request, i.e request/ip - remote IP request/port - remote port Variables.go has been deleted. Signed-off-by: Miek Gieben <miek@miek.nl>
k8s' client-go has been updated to v8.0.0 (1.11). This fix updates client-go dependency so that it is in sync. Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
This was done anyway, but only deep in the functions, just do this everywhere; allows for shorter code and request.Request allows for caching as well. Cleanups, make it more Go like. * remove unneeded switches * remove testdir (why was this there??) * simplify the logic * remove unneeded variables * put short functions on a single line * fix documentation. * spin off wire funcs in wire.go, make them functions. Signed-off-by: Miek Gieben <miek@miek.nl>
* request: add LocalIP Fix TODO that was added: Add LocalIP and test the Clear() method. Signed-off-by: Miek Gieben <miek@miek.nl> * Move to Errorf PR feedback and move to Errorf instead Fatalf. Signed-off-by: Miek Gieben <miek@miek.nl>
…tion (coredns#1901) (coredns#1910) Exporting Zone.File to avoid getters and setters Updating getter and setter for Zone.File to be less racy Renaming GetFile to File in zone plugin
* plugin/cache: add extra test for FORMERR Add extra test that test for not caching a formerr. Signed-off-by: Miek Gieben <miek@miek.nl> * govet Signed-off-by: Miek Gieben <miek@miek.nl>
Allow plugins to dump messages in text pcap to the log. The forward plugin does this when a reply does not much the query. If the debug plugin isn't loaded Hexdump and Hexdumpf are noop. Signed-off-by: Miek Gieben <miek@miek.nl>
Automatically submitted.
* plugin/metadata: adjust doc to latest code. Signed-off-by: Miek Gieben <miek@miek.nl>
* Doc updates Make the name section fit on one line. Signed-off-by: Miek Gieben <miek@miek.nl> * Regen docs Signed-off-by: Miek Gieben <miek@miek.nl>
Automatically submitted.
* plugin/forward: add prefer_udp option * updated according to code review - fixed linter warning - removed metric parameter in Proxy.Connect()
* DoH: put in pkg/doh Factor out the DoH stuff into its own package, add function to request a DoH response. This can be used by forward (and maybe proxy) to implement DoH client support. Signed-off-by: Miek Gieben <miek@miek.nl> * lint Signed-off-by: Miek Gieben <miek@miek.nl> * ... and make it compile Signed-off-by: Miek Gieben <miek@miek.nl>
Fix documentation and touch up plugin/forward/README.md Signed-off-by: Miek Gieben <miek@miek.nl>
Fix documentation and remove the unused From method. Signed-off-by: Miek Gieben <miek@miek.nl>
…oredns#1928) * - add support of metadata values for edns0 local variables * - comments from review. * - simplify label check. Add UT * - enhance check for Labels, add UT - remove IsMetadataSet * - edns0 variable - if variable is not found just ignore the rewrite.
…re rewrite - like if metadata does not exist. (coredns#1953)
* plugin/metadata: finish documentation Finish the README.md, add corner case in the IsLabel test and reword some code comments slightly. Generate the man-pages and add man/coredns-metadata.7 as well. Signed-off-by: Miek Gieben <miek@miek.nl> * Fix test Signed-off-by: Miek Gieben <miek@miek.nl>
Prevent future; "remove trailing whitespace" PR, but adding a simple presubmit that checks for this. This presubmit flagged quite some offenders, remove all trailing whitespace from. Apart from that there aren't any other changes. Signed-off-by: Miek Gieben <miek@miek.nl>
* plugin/forward: add HealthChecker interface Make the HealthChecker interface and morph the current DNS health checker into that interface. Remove all whole bunch of method on Forward that didn't make sense. This is done in preparation of adding a DoH client to forward - which requires a completely different healthcheck implementation (and more, but lets start here) Signed-off-by: Miek Gieben <miek@miek.nl> * Use protocol Signed-off-by: Miek Gieben <miek@miek.nl> * Dial doesnt need to be method an Forward either Signed-off-by: Miek Gieben <miek@miek.nl> * Address comments Address various comments on the PR. Signed-off-by: Miek Gieben <miek@miek.nl>
* release: automate the release This PR aims to various pieces into place so we can automate the coredns release. It needs the script from coredns/release to be installed. Dreck is to be setup as described in coredns/release/README.md The release-coredns script can be tested and allows for other branches than master to be test. This PR also features some cleanup in the Makefile.release so we don't call the godep target for each linux release - this speeds it up for some bit. Manually running ./release-coredns -t auto-release builds the artifects for this release, but (of course) doesn't upload anything yet. Add /release to the OWNERS and allow command to be executed (this still needs to be tested). Signed-off-by: Miek Gieben <miek@miek.nl> * that makefile target doesnt exist anymore Signed-off-by: Miek Gieben <miek@miek.nl> * test release for now Signed-off-by: Miek Gieben <miek@miek.nl> * Slightly better output Signed-off-by: Miek Gieben <miek@miek.nl> * remove again Signed-off-by: Miek Gieben <miek@miek.nl>
Automatically submitted.
Releasing 1.2.0 uncovered some rough edges that need to be documented. Also fix github-push target and docker login. Signed-off-by: Miek Gieben <miek@miek.nl>
Automatically submitted.
Log and returns an error when the name rewrite creates a name that is illegal. Add test in name_test.go to see if an error is returned. Possible followup could be the only check this if a name-rewrite is done. Fixes: coredns#1638 Signed-off-by: Miek Gieben <miek@miek.nl>
Signed-off-by: Miek Gieben <miek@miek.nl>
* Remove version pinning of thrift, ugoriji/go, and etcd For incompatibility reasons at one point, we were forced to pining the version of thrift, ugoriji/go, and etcd to very specific versions (some are not versioned commits) to get around the build issues. It looks like those incompatibility issues are gone so we could remove the pinning of thrift, ugoriji/go, and etcd. Signed-off-by: Yong Tang <yong.tang.github@outlook.com> * Update Gopkg.lock Signed-off-by: Yong Tang <yong.tang.github@outlook.com> * Update vendor library Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Tag the 1.3.1 version. Signed-off-by: Miek Gieben <miek@miek.nl>
This reverts commit c463189.
Previous coredns versions had a `make test` target and our ci-operator currently uses this target to kick off the tests. So, at least to get tests running against this PR without requiring changes to the openshift/release ci-operator setup, I'm adding that target back. This just runs the full litany of tests that Travis would run except for the coverage targets and the tests that require a running etcd server.
At least in the case of mholt/caddy, compilation fails if not pinned to the exact same version that upstream uses (ie v0.11.3 fails, v0.11.1 does not).
This pins the Kubernetes client dependencies to the same version used in previous releases of our fork. And updates the mismatched comment vs reality in Gopkg.toml that we had previously to avoid future confusion.
openshift-ci-robot
added
the
size/XXL
|
I did not squash any of my commits that dealt with restoring our The coredns tests pass for me locally. I swapped out the openshift-dns DaemonSet in my OpenShift 4 cluster to use the updated CoreDNS image: So far everything seems to be working as expected with some basic manual testing of DNS in the cluster. |
|
/retest |
1 similar comment
|
/retest |
|
Awesome work, thanks. One last thought: what about introducing a bespoke Makefile containing the test target we want rather than patching the upstream file? The CI job could be configured to include an execution of make against the file. Worth it? /cc @openshift/sig-network-edge |
|
A new Makefile to avoid future merge conflicts may not be a bad idea. But merges would still need to pull in any relevant changes as far as test suites, where tests live, how they're run, etc into that new Makefile. Would it be better to get alerted to that fact via a merge conflict or to manually keep an eye on it? And, a new Makefile will require adjustment to ci-operator in the openshift/release repo for this component, right? That's the main reason I didn't do anything in this PR as that would have prevented Prow from being able to run the tests here. |
Good points. Let's keep it simple with what you already have. /lgtm |
|
@ironcladlou: changing LGTM is restricted to assignees, and only openshift/coredns repo collaborators may be assigned issues. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
@smarterclayton oh yeah, we don't patch OWNERS 😁 How do you want to manage merging? |
|
Replace owners.
Delete existing Owners in one commit and then create another in the next
file.
On Feb 7, 2019, at 11:19 AM, Dan Mace <notifications@github.com> wrote:
@smarterclayton <https://github.com/smarterclayton> oh yeah, we don't patch
OWNERS 😁
How do you want to manage merging?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#6 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ABG_p0vY_RbGFPIXgcQlo_lzfJe9eeaSks5vLFIagaJpZM4akcOz>
.
|
|
/lgtm |
|
Several e2e tests failed on that run, but from initial digging it looks like a couple of known flaking tests and some known manifest digest issues that sometimes crop up. I'll kick off the e2e tests again and see if the results repeat. /test e2e-aws |
|
/retest |
This bumps our fork to coredns/coredns v1.3.1 while maintaining our patches as well.
Specifically:
This also adds a new patch to our that adds a
make testtarget back to theMakefile. Upstream got rid of thetesttarget. We could choose to drop this patch, but there's not another target that easily exposes running tests that don't depend on a running etcd server.