{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":120642119,"defaultBranch":"master","name":"kubernetes","ownerLogin":"pohly","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2018-02-07T16:39:26.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/22076?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1716558180.0","currentOid":""},"activityList":{"items":[{"before":"f04c3aba716c3744427b01d7da466f885ad0faa5","after":"1412d24a2e0fd2371d6e1f7a1c653be270e4fb6b","ref":"refs/heads/dra-one-of-source","pushedAt":"2024-05-24T15:24:14.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: remove \"source\" indirection from v1 Pod API\n\nThis makes the API nicer:\n\n    resourceClaims:\n    - name: with-template\n      resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      resourceClaimName: test-shared-claim\n\nPreviously, this was:\n\n    resourceClaims:\n    - name: with-template\n      source:\n        resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      source:\n        resourceClaimName: test-shared-claim\n\nA more long-term benefit is that other, future alternatives\nmight not make sense under the \"source\" umbrella.\n\nThis is a breaking change. It's justified because DRA is still\nalpha and will have several other API breaks in 1.31.","shortMessageHtmlLink":"DRA: remove \"source\" indirection from v1 Pod API"}},{"before":"12da58e63cb9dc2e25c014698c6d21a9002bb87a","after":"f04c3aba716c3744427b01d7da466f885ad0faa5","ref":"refs/heads/dra-one-of-source","pushedAt":"2024-05-24T14:05:11.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: remove \"source\" indirection from v1 Pod API\n\nThis makes the API nicer:\n\n    resourceClaims:\n    - name: with-template\n      resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      resourceClaimName: test-shared-claim\n\nPreviously, this was:\n\n    resourceClaims:\n    - name: with-template\n      source:\n        resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      source:\n        resourceClaimName: test-shared-claim\n\nA more long-term benefit is that other, future alternatives\nmight not make sense under the \"source\" umbrella.\n\nThis is a breaking change. It's justified because DRA is still\nalpha and will have several other API breaks in 1.31.","shortMessageHtmlLink":"DRA: remove \"source\" indirection from v1 Pod API"}},{"before":"a892c3bd4baf2a6e4c581340b4aee899e2280edc","after":"3ed1de6ece05f75bfb196e67304681f0c271c86e","ref":"refs/heads/ginkgo-gomega-update","pushedAt":"2024-05-24T13:56:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"dependencies: ginkgo v1.19.0\n\nThis adds support for --label-filter with labels that represent sets (like our Feature:<Foo>).","shortMessageHtmlLink":"dependencies: ginkgo v1.19.0"}},{"before":null,"after":"12da58e63cb9dc2e25c014698c6d21a9002bb87a","ref":"refs/heads/dra-one-of-source","pushedAt":"2024-05-24T13:43:00.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: remove \"source\" indirection from v1 Pod API\n\nThis makes the API nicer:\n\n    resourceClaims:\n    - name: with-template\n      resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      resourceClaimName: test-shared-claim\n\nPreviously, this was:\n\n    resourceClaims:\n    - name: with-template\n      source:\n        resourceClaimTemplateName: test-inline-claim-template\n    - name: with-claim\n      source:\n        resourceClaimName: test-shared-claim\n\nA more long-term benefit is that other, future alternatives\nmight not make sense under the \"source\" umbrella.\n\nThis is a breaking change. It's justified because DRA is still\nalpha and will have several other API breaks in 1.31.","shortMessageHtmlLink":"DRA: remove \"source\" indirection from v1 Pod API"}},{"before":null,"after":"6d2f07549c400f68afe47886da9d09df7a1c9445","ref":"refs/heads/ginkgo-label-filter","pushedAt":"2024-05-23T16:20:45.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"e2e framework: don't add skip strings when user used --label-filter\n\nWhen the user manually sets either focus, skip or (new!) label filter, then the\nframework should not interfere and add the default skip strings.","shortMessageHtmlLink":"e2e framework: don't add skip strings when user used --label-filter"}},{"before":"0506086d3f7a5856672c0c233707456310490a57","after":"a892c3bd4baf2a6e4c581340b4aee899e2280edc","ref":"refs/heads/ginkgo-gomega-update","pushedAt":"2024-05-22T14:58:21.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI\n\nWe typically have plenty of skipped tests in an E2E run, which causes Ginkgo to\nprint many S characters (one for each skipped test). This easily fills up an entire\nconsole window. Now the default is to silence that\noutput. GINKGO_SILENCE_SKIPS=n reverts to the previous behavior.\n\nBy default, Ginkgo prints all progress characters (S and o) in a single\nline. Buffering in Prow causes that output to occur only after the run is over,\nwhich defeats the purpose of having those characters. Now a newline is added\nafter each character, so there is visible progress in Prow while Ginkgo runs.","shortMessageHtmlLink":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI"}},{"before":"a5736180018dfad51cb6d64ed2d7f283b860042c","after":"0506086d3f7a5856672c0c233707456310490a57","ref":"refs/heads/ginkgo-gomega-update","pushedAt":"2024-05-22T10:41:40.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI\n\nWe typically have plenty of skipped tests in an E2E run, which causes Ginkgo to\nprint many S characters (one for each skipped test). This easily fills up an entire\nconsole window. Now the default is to silence that\noutput. GINKGO_SILENCE_SKIPS=n reverts to the previous behavior.\n\nBy default, Ginkgo prints all progress characters (S and o) in a single\nline. Buffering in Prow causes that output to occur only after the run is over,\nwhich defeats the purpose of having those characters. Now a newline is added\nafter each character, so there is visible progress in Prow while Ginkgo runs.","shortMessageHtmlLink":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI"}},{"before":"44be1b567f945ce2be3d588523842f80cceb0ff6","after":"a5736180018dfad51cb6d64ed2d7f283b860042c","ref":"refs/heads/ginkgo-gomega-update","pushedAt":"2024-05-22T09:05:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI\n\nWe typically have plenty of skipped tests in an E2E run, which causes Ginkgo to\nprint many S characters (one for each skipped test). This easily fills up an entire\nconsole window. Now the default is to silence that\noutput. GINKGO_SILENCE_SKIPS=n reverts to the previous behavior.\n\nBy default, Ginkgo prints all progress characters (S and o) in a single\nline. Buffering in Prow causes that output to occur only after the run is over,\nwhich defeats the purpose of having those characters. Now a newline is added\nafter each character, so there is visible progress in Prow while Ginkgo runs.","shortMessageHtmlLink":"hack/ginkgo-e2e.sh: skip S character by default, add newlines in CI"}},{"before":"18f0af1f000f95749ca1ea075d62ca89e86bb7da","after":"44be1b567f945ce2be3d588523842f80cceb0ff6","ref":"refs/heads/ginkgo-gomega-update","pushedAt":"2024-05-22T08:30:58.000Z","pushType":"push","commitsCount":2217,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"dependencies: ginkgo v2.18.0, gomega v1.33.1\n\nThe most interesting aspect of the new Ginkgo is tweaking the output so that\nit's easier to follow while a job runs in\nProw (https://github.com/onsi/ginkgo/issues/1347). Using this in\nhack/ginkgo-e2e.sh will follow in a separate commit.\n\nGomega gets bumped to the latest release to keep it up-to-date.","shortMessageHtmlLink":"dependencies: ginkgo v2.18.0, gomega v1.33.1"}},{"before":null,"after":"939f3551d2ceb35e8aca869a0b3ae5d4d4fcf3a6","ref":"refs/heads/dra-scheduler-prebind-fix","pushedAt":"2024-05-17T14:43:21.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: fix scheduler/resource claim controller race\n\nThere was a race caused by having to update claim finalizer and status in two\ndifferent operations:\n- Resource claim controller removes allocation, does not yet\n  get to remove the finalizer.\n- Scheduler prepares an allocation, without adding the finalizer\n  because it's there.\n- Controller removes finalizer.\n- Scheduler adds allocation.\n\nThis is an invalid state. Automatic checking found this during the execution of\nthe \"with translated parameters on single node.*supports sharing a claim\nsequentially\" E2E test, but only when run stand-alone. When running in\nparallel (as in the CI), the bad outcome of the race did not occur.\n\nThe fix is to check that the finalizer is still set when adding the\nallocation. The apiserver doesn't check that because it doesn't know which\nfinalizer goes with the allocation result. It could check for \"some finalizer\",\nbut that is not guaranteed to be correct (could be some unrelated one).\n\nChecking the finalizer can only be done with a JSON patch. Despite the\ncomplications, having the ability to add multiple pods concurrently to\nReservedFor seems worth it (avoids expensive rescheduling or a local retry\nloop).\n\nThe resource claim controller doesn't need this, it can do a normal update\nwhich implicitly checks ResourceVersion.","shortMessageHtmlLink":"DRA: fix scheduler/resource claim controller race"}},{"before":"e0ede9437768e99408e2f1965421157bdc3f17ef","after":"e8323872c5cf72530777b9551f8c7f0f2881f243","ref":"refs/heads/dra-kubelet-api-version-independent","pushedAt":"2024-05-03T15:10:55.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: move ResourceSlice publishing into DRA drivers\n\nThis is a first step towards making kubelet independent of the resource.k8s.io\nAPI versioning because it now doesn't need to copy structs defined by that API\nfrom the driver to the API server. The next step is removing the other\ndirection (reading ResourceClaim status and passing the resource handle to\ndrivers).\n\nThe drivers are still using the kubelet to access ResourceSlices, so the node\nauthorizer still applies and limits access to those objects that kubelet has\naccess to. Drivers don't need separate credentials. This works with the help of\na REST proxy that is provided to drivers by kubelet. That proxy checks that\ndrivers really only use it to access ResourceSlices.\n\nAs before, the kubelet removes all ResourceSlices for its node at startup, then\nDRA drivers recreate them if (and only if) they start up again. This ensures\nthat there are no orphaned ResourceSlices when a driver gets removed while the\nkubelet was down.\n\nWhile at it, logging gets cleaned up and updated to use structured, contextual\nlogging as much as possible. gRPC requests and streams now use a shared,\nper-process request ID and streams also get logged.","shortMessageHtmlLink":"DRA: move ResourceSlice publishing into DRA drivers"}},{"before":"762aa39bac3eff7b04fa2827a33e8a3f2b9659b8","after":"e0ede9437768e99408e2f1965421157bdc3f17ef","ref":"refs/heads/dra-kubelet-api-version-independent","pushedAt":"2024-05-02T17:37:30.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: move ResourceSlice publishing into DRA drivers\n\nThis is a first step towards making kubelet independent of the resource.k8s.io\nAPI versioning because it now doesn't need to copy structs defined by that API\nfrom the driver to the API server. The next step is removing the other\ndirection (reading ResourceClaim status and passing the resource handle to\ndrivers).\n\nThe drivers are still using the kubelet to access ResourceSlices, so the node\nauthorizer still applies and limits access to those objects that kubelet has\naccess to. Drivers don't need separate credentials. This works with the help of\na REST proxy that is provided to drivers by kubelet. That proxy checks that\ndrivers really only use it to access ResourceSlices.\n\nAs before, the kubelet removes all ResourceSlices for its node at startup, then\nDRA drivers recreate them if (and only if) they start up again. This ensures\nthat there are no orphaned ResourceSlices when a driver gets removed while the\nkubelet was down.\n\nWhile at it, logging gets cleaned up and updated to use structured, contextual\nlogging as much as possible.","shortMessageHtmlLink":"DRA: move ResourceSlice publishing into DRA drivers"}},{"before":null,"after":"77a7680db3ce73fdcd00c08ae01c9e49cbd9209a","ref":"refs/heads/PR144-e2e_node-DRA-test-plugin-failures","pushedAt":"2024-04-30T16:54:13.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"DRA: implement error injection and blocking inside test-specific interceptor\n\nPreviously, blocking and error injection where implemented inside the kubelet\nplugin code of the driver. This was a bit hacky and not thread-safe (no\nlocking).\n\nFor upcoming tests we need more flexibility. Now a test can inject its own gRPC\ninterceptor into the kubelet plugin code. A pre-built one supports blocking and\nerror injection, but completely different interceptors are also possible if\nneeded.","shortMessageHtmlLink":"DRA: implement error injection and blocking inside test-specific inte…"}},{"before":null,"after":"ff779f1cb56cf896405e52f7923188b99b88bb00","ref":"refs/heads/event-broadcaster-shutdown-fix","pushedAt":"2024-04-30T10:22:51.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"client-go/tools/record: fix and test Broadcaster shutdown + logging\n\nConstructing a Broadcaster already starts a watch which runs in the\nbackground. Shutdown must be called to avoid leaking the goroutine.  Providing\na context was supposed to remove the need to call Shutdown, but that did not\nactually work because the logic for \"must check for cancellation\" was\naccidentally inverted.\n\nWhile at it, structured log output also gets tested together with checking for\ngoroutine leaks.","shortMessageHtmlLink":"client-go/tools/record: fix and test Broadcaster shutdown + logging"}},{"before":"7d7ffeebdc71ee4f15ddfa08e4c21438e6b2b62e","after":"8d6a5be8854d12d2ae37756f03324f70141171e0","ref":"refs/heads/log-client-go-tools-cache","pushedAt":"2024-04-30T10:21:30.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"convert more event broadcasters to structured logging","shortMessageHtmlLink":"convert more event broadcasters to structured logging"}},{"before":"78581784f168c2064d521d2ce337e27f3e6467d0","after":"b92273a760503cc57aba37c4d3a28554f7fec7f8","ref":"refs/heads/controllers-context-support","pushedAt":"2024-04-29T19:01:46.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"apiserver + controllers: enhance context support\n\n27a68aee3a4834 introduced context support for events. Creating an event\nbroadcaster with context makes tests more resilient against leaking goroutines\nwhen that context gets canceled at the end of a test and enables per-test\noutput via ktesting.\n\nThe context could get passed to the constructor. A cleaner solution is to\nenhance context support for the apiserver and then pass the context into the\ncontroller's run method. This ripples up the call stack to all places which\nstart an apiserver.","shortMessageHtmlLink":"apiserver + controllers: enhance context support"}},{"before":"cd457e1861fd393a65b568c78d051f7d0ee71cb7","after":"78581784f168c2064d521d2ce337e27f3e6467d0","ref":"refs/heads/controllers-context-support","pushedAt":"2024-04-29T14:01:31.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"apiserver + controllers: enhance context support\n\n27a68aee3a4834 introduced context support for events. Creating an event\nbroadcaster with context makes tests more resilient against leaking goroutines\nwhen that context gets canceled at the end of a test and enables per-test\noutput via ktesting.\n\nThe context could get passed to the constructor. A cleaner solution is to\nenhance context support for the apiserver and then pass the context into the\ncontroller's run method. This ripples up the call stack to all places which\nstart an apiserver.","shortMessageHtmlLink":"apiserver + controllers: enhance context support"}},{"before":"2d66ba2f85a71d65291ec78c6d8e8ee6f9742683","after":"0b0e8e3882933e2c4642febac0442716060e107b","ref":"refs/heads/dra-scheduler-assume-cache-eventhandlers","pushedAt":"2024-04-29T12:43:39.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"scheduler: central ResourceClaim assume cache\n\nThis enables connecting the event handler for ResourceClaim to the assume\ncache, which addresses a theoretic race condition.\n\nIt may also be useful for implementing the autoscaler support, because now\nthe autoscaler can modify the content of the cache.","shortMessageHtmlLink":"scheduler: central ResourceClaim assume cache"}},{"before":"7d9abd57c6ca3c979e289c96f000ddfbaefbcc56","after":"2d66ba2f85a71d65291ec78c6d8e8ee6f9742683","ref":"refs/heads/dra-scheduler-assume-cache-eventhandlers","pushedAt":"2024-04-29T08:59:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"scheduler: central ResourceClaim assume cache\n\nThis enables connecting the event handler for ResourceClaim to the assume\ncache, which addresses a theoretic race condition.\n\nIt may also be useful for implementing the autoscaler support, because now\nthe autoscaler can modify the content of the cache.","shortMessageHtmlLink":"scheduler: central ResourceClaim assume cache"}},{"before":"3cc6fe20590f590934ab8ef00720768480c35e7d","after":"7d9abd57c6ca3c979e289c96f000ddfbaefbcc56","ref":"refs/heads/dra-scheduler-assume-cache-eventhandlers","pushedAt":"2024-04-29T06:35:27.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"fixup! scheduler: AddEventHandler for assume cache","shortMessageHtmlLink":"fixup! scheduler: AddEventHandler for assume cache"}},{"before":"24557fd928c3f20e518b1c07079968a9c5437b43","after":"3cc6fe20590f590934ab8ef00720768480c35e7d","ref":"refs/heads/dra-scheduler-assume-cache-eventhandlers","pushedAt":"2024-04-28T12:40:21.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"fixup! scheduler: AddEventHandler for assume cache","shortMessageHtmlLink":"fixup! scheduler: AddEventHandler for assume cache"}},{"before":"414b9f639eff2595854d774712432de91e5644df","after":"6eba52e79541796a96941a95710726c16eaf4fea","ref":"refs/heads/apidiff","pushedAt":"2024-04-26T09:00:08.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/apidiff.sh: compare between two revisions, usability enhancements\n\nIn a Prow job, the current work tree is the result of merging a PR into the\ntarget. We want apidiff.sh from there, but then need to invoke it for two\nspecific revisions and compare.\n\nWhile at it, output and usability get enhanced. The directory parameter(s) may\nbe absolute paths or lack the leading . that is required by apidiff.","shortMessageHtmlLink":"hack/apidiff.sh: compare between two revisions, usability enhancements"}},{"before":"b120eca5662435ee222a940956d783ae8382bd17","after":"414b9f639eff2595854d774712432de91e5644df","ref":"refs/heads/apidiff","pushedAt":"2024-04-26T08:55:21.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/apidiff.sh: compare between two revisions, usability enhancements\n\nIn a Prow job, the current work tree is the result of merging a PR into the\ntarget. We want apidiff.sh from there, but then need to invoke it for two\nspecific revisions and compare.\n\nWhile at it, output and usability get enhanced. The directory parameter(s) may\nbe absolute paths or lack the leading . that is required by apidiff.","shortMessageHtmlLink":"hack/apidiff.sh: compare between two revisions, usability enhancements"}},{"before":"679195d8cc56e7f6ed4f75cde13ce484f452c996","after":"b120eca5662435ee222a940956d783ae8382bd17","ref":"refs/heads/apidiff","pushedAt":"2024-04-26T08:48:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/apidiff.sh: compare between two revisions, usability enhancements\n\nIn a Prow job, the current work tree is the result of merging a PR into the\ntarget. We want apidiff.sh from there, but then need to invoke it for two\nspecific revisions and compare.\n\nWhile at it, output and usability get enhanced. The directory parameter(s) may\nbe absolute paths or lack the leading . that is required by apidiff.","shortMessageHtmlLink":"hack/apidiff.sh: compare between two revisions, usability enhancements"}},{"before":"97040efd0b2b82d881329dd3f4abc6192629b183","after":"4306a8b1927ed734f82ae81cb149270c93e56a4c","ref":"refs/heads/dra-scheduler-perf-structured-parameters","pushedAt":"2024-04-26T06:39:35.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"scheduler_perf: add DRA structured parameters test with shared claims\n\nSeveral pods sharing the same claim is not common, but can be useful and thus\nshould get tested.\n\nBefore, createPods and createAny operations were not able to do this because\neach generated object was the same. What we need are different, predictable\nnames of the claims (from createAny) and different references to those in the\npods (from createPods). Now text/template processing with the index number of\nthe pod respectively claim as input is used to inject these varying fields. A\n\"div\" function is needed to use the same claim in several different pods.\n\nWhile at it, some existing test cases get cleaned up a bit (removal of\nincorrect comments, adding comments for testing with queuing hints).","shortMessageHtmlLink":"scheduler_perf: add DRA structured parameters test with shared claims"}},{"before":null,"after":"97040efd0b2b82d881329dd3f4abc6192629b183","ref":"refs/heads/dra-scheduler-perf-structured-parameters","pushedAt":"2024-04-26T06:37:07.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"scheduler_perf: add DRA structured parameters test with shared claims\n\nSeveral pods sharing the same claim is not common, but can be useful and thus\nshould get tested.\n\nBefore, createPods and createAny operations were not able to do this because\neach generated object was the same. What we need are different, predictable\nnames of the claims (from createAny) and different references to those in the\npods (from createPods). Now text/template processing with the index number of\nthe pod respectively claim as input is used to inject these varying fields.\n\nWhile at it, some existing test cases get cleaned up a bit (removal of\nincorrect comments, adding comments for testing with queuing hints).","shortMessageHtmlLink":"scheduler_perf: add DRA structured parameters test with shared claims"}},{"before":"06281bb8fb9f243cdeb48801cc5784f03911f9a1","after":"50c12437604b0cd5a73514389409fc2fde8b91bd","ref":"refs/heads/cloud-providers-context-support","pushedAt":"2024-04-25T09:57:25.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"cloud providers: enhance context support\n\n27a68aee3a4834 introduced context support for events. Creating an event\nbroadcaster with context makes tests more resilient against leaking goroutines\nwhen that context gets canceled at the end of a test and enables per-test\noutput via ktesting.\n\nWhile at it, all context.TODO calls get removed in files that were touched.","shortMessageHtmlLink":"cloud providers: enhance context support"}},{"before":"44df029ebdb6738bc724ba0d4d3e540c1c819405","after":"7f54c5dfec3e95a2751156a3a1ae4407be68297a","ref":"refs/heads/dra-scheduler-assume-cache","pushedAt":"2024-04-25T09:47:28.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"scheduler: remove AssumeCache interface\n\nThere's no reason for having the interface because there is only one\nimplementation. Makes the implementation of the test functions a bit\nsimpler (no casting). They are still stand-alone functions instead of methods\nbecause they should not be considered part of the \"normal\" API.","shortMessageHtmlLink":"scheduler: remove AssumeCache interface"}},{"before":"73eb79512f2524bf5c9b981b8c2f16ea8f9e61fc","after":"679195d8cc56e7f6ed4f75cde13ce484f452c996","ref":"refs/heads/apidiff","pushedAt":"2024-04-25T08:39:26.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"hack/apidiff.sh: handle lack of origin\n\nThe script failed in the Prow job with:\n\n   fatal: Not a valid object name origin/master\n\nThat came from \"git merge-base origin/master HEAD\", which isn't needed because\nthe Prow job intentionally sets -r explicitly.\n\nNow the fallback is only tried if needed. While at it, the informational\nmessages get enhanced.","shortMessageHtmlLink":"hack/apidiff.sh: handle lack of origin"}},{"before":"36aa8b1d528eb928accf39c6e2f244051848dd58","after":"06281bb8fb9f243cdeb48801cc5784f03911f9a1","ref":"refs/heads/cloud-providers-context-support","pushedAt":"2024-04-25T08:28:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"pohly","name":"Patrick Ohly","path":"/pohly","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/22076?s=80&v=4"},"commit":{"message":"cloud providers: enhance context support\n\n27a68aee3a4834 introduced context support for events. Creating an event\nbroadcaster with context makes tests more resilient against leaking goroutines\nwhen that context gets canceled at the end of a test and enables per-test\noutput via ktesting.\n\nWhile at it, all context.TODO calls get removed in files that were touched.","shortMessageHtmlLink":"cloud providers: enhance context support"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEUy4MDwA","startCursor":null,"endCursor":null}},"title":"Activity · pohly/kubernetes"}