{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":385313902,"defaultBranch":"master","name":"ovn-kubernetes","ownerLogin":"SalDaniele","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2021-07-12T16:33:32.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/78045843?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1689889046.0","currentOid":""},"activityList":{"items":[{"before":"c2d6de7095769c439612e48928436e0a6577fdeb","after":"099f80327093ed5b002513929a7333b70d173cbb","ref":"refs/heads/main","pushedAt":"2023-09-11T19:22:46.000Z","pushType":"push","commitsCount":257,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Add e2e for mimicing pod2external case via GR\n\nSigned-off-by: Surya Seetharaman ","shortMessageHtmlLink":"Add e2e for mimicing pod2external case via GR"}},{"before":"3251a8f342cf8b96e64fda767aa699c42b84a097","after":"389c771c1af34be554a165103e7a9d50423cce7a","ref":"refs/heads/update_dpu_host_addresses","pushedAt":"2023-07-24T18:14:38.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jcaamano","name":"Jaime Caamaño Ruiz","path":"/jcaamano","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/32458427?s=80&v=4"},"commit":{"message":"Use NodePrimaryIp to set host-addresses in DPU mode\n\nWhen running a 2-cluster design in DPU mode, currently host-addresses\nis set to the IP address of the Infrastructure node (i.e. Bluefield\nnode) running of behalf of the tenant cluster. This will cause all\nNodePort service to fail in DPU mode.\n\nInstead, we should calculate the host-addresses using the NodePrimaryIP\nwhich will set the host-address to the Node IP of the tenant node.\n\nSigned-off-by: Salvatore Daniele ","shortMessageHtmlLink":"Use NodePrimaryIp to set host-addresses in DPU mode"}},{"before":"8b723d53551275342abe54ac90545be95eccb0fb","after":"3251a8f342cf8b96e64fda767aa699c42b84a097","ref":"refs/heads/update_dpu_host_addresses","pushedAt":"2023-07-24T08:12:47.000Z","pushType":"push","commitsCount":214,"pusher":{"login":"jcaamano","name":"Jaime Caamaño Ruiz","path":"/jcaamano","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/32458427?s=80&v=4"},"commit":{"message":"Merge branch 'master' into update_dpu_host_addresses","shortMessageHtmlLink":"Merge branch 'master' into update_dpu_host_addresses"}},{"before":"8db2bc0b0cabe68057b33877e09f7dca349a3be1","after":"8b723d53551275342abe54ac90545be95eccb0fb","ref":"refs/heads/update_dpu_host_addresses","pushedAt":"2023-07-21T22:58:37.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Use NodePrimaryIp to set host-addresses in DPU mode\n\nWhen running a 2-cluster design in DPU mode, currently host-addresses\nis set to the IP address of the Infrastructure node (i.e. Bluefield\nnode) running of behalf of the tenant cluster. This will cause all\nNodePort service to fail in DPU mode.\n\nInstead, we should calculate the host-addresses using the NodePrimaryIP\nwhich will set the host-address to the Node IP of the tenant node.\n\nSigned-off-by: Salvatore Daniele ","shortMessageHtmlLink":"Use NodePrimaryIp to set host-addresses in DPU mode"}},{"before":"246acd6a3383e0910dce07a65c44dbc4da1d30f3","after":"c2d6de7095769c439612e48928436e0a6577fdeb","ref":"refs/heads/main","pushedAt":"2023-07-21T13:57:41.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"multi-homing, e2e tests: wait for NAD creation to settle\n\nWhen there are multiple NADs for the same network, we need to wait a bit\nmore before starting to create pods and multi-net policies, since the\nNAD controller is not level driven - thus, on pod / policy creation we\nmay reach a situation where the target NAD was not yet added to the\nnetwork controller.\n\nSigned-off-by: Miguel Duarte Barroso ","shortMessageHtmlLink":"multi-homing, e2e tests: wait for NAD creation to settle"}},{"before":"c2d6de7095769c439612e48928436e0a6577fdeb","after":"246acd6a3383e0910dce07a65c44dbc4da1d30f3","ref":"refs/heads/main","pushedAt":"2023-07-20T22:08:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Ensure minimum disruption when upgrading to ic\n\nWhile upgrading to IC clusters from nonIC, since\nthe OVN DBs are moving from master to node, there is\nbasically an overhead time for the new ovnkube-controller\nto create the DBs and flow from stratch that leads to\nnetwork disruptions between 1-2mins. This PR hopes to\nreduce that disruption time by keeping the ovnkube-node\nand ovn-controller connected to legacy SBDB up until the\nnew SBDB is all ready with the populated elements.\n\nSigned-off-by: Surya Seetharaman ","shortMessageHtmlLink":"Ensure minimum disruption when upgrading to ic"}},{"before":null,"after":"c2d6de7095769c439612e48928436e0a6577fdeb","ref":"refs/heads/main","pushedAt":"2023-07-20T21:37:26.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"multi-homing, e2e tests: wait for NAD creation to settle\n\nWhen there are multiple NADs for the same network, we need to wait a bit\nmore before starting to create pods and multi-net policies, since the\nNAD controller is not level driven - thus, on pod / policy creation we\nmay reach a situation where the target NAD was not yet added to the\nnetwork controller.\n\nSigned-off-by: Miguel Duarte Barroso ","shortMessageHtmlLink":"multi-homing, e2e tests: wait for NAD creation to settle"}},{"before":null,"after":"8db2bc0b0cabe68057b33877e09f7dca349a3be1","ref":"refs/heads/update_dpu_host_addresses","pushedAt":"2023-06-26T21:19:17.985Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Use NodePrimaryIp to set host-addresses in DPU mode\n\nWhen running a 2-cluster design in DPU mode, currently host-addresses\nis set to the IP address of the Infrastructure node (i.e. Bluefield\nnode) running of behalf of the tenant cluster. This will cause all\nNodePort service to fail in DPU mode.\n\nInstead, we should calculate the host-addresses using the NodePrimaryIP\nwhich will set the host-address to the Node IP of the tenant node.\n\nSigned-off-by: Salvatore Daniele ","shortMessageHtmlLink":"Use NodePrimaryIp to set host-addresses in DPU mode"}},{"before":"61541cfcaa5f5a111b93f0234440d18e778969df","after":"c2bcf8baa7163db65bfa9719e121b975f2ba497e","ref":"refs/heads/master","pushedAt":"2023-05-25T17:03:00.173Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Merge pull request #3530 from bn222/NHE-241\n\nExpose PfID and VfID in DPU host mode and read that in DPU mode","shortMessageHtmlLink":"Merge pull request ovn-org#3530 from bn222/NHE-241"}},{"before":"cfae31d09ad36f613c4c07492e2969aa97e28074","after":"61541cfcaa5f5a111b93f0234440d18e778969df","ref":"refs/heads/master","pushedAt":"2023-05-25T16:32:09.558Z","pushType":"push","commitsCount":333,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"initial pods are not wired for hybrid overlay\n\nWhen toggling hybrid overlay the linux nodes can miss wiring the pods on\nstart-up if allocating the hybrid overlay DRIP and DRMAC takes too long.\n\nFix this issue by passing a podLister to the node controller and running\nthe AddPods() command on all local pods once on startup after DRIP and\nDRMAC are set.\n\nSigned-off-by: Jacob Tanenbaum ","shortMessageHtmlLink":"initial pods are not wired for hybrid overlay"}},{"before":null,"after":"2f2dd1b0abfe0059fdd5afdd90a7f95c5f8bf9b8","ref":"refs/heads/zenguis_change_us","pushedAt":"2023-05-04T22:05:33.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"SalDaniele","name":"Salvatore Daniele","path":"/SalDaniele","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78045843?s=80&v=4"},"commit":{"message":"Fix handling of DPU on node address change/setting\n\nThe DPU needs to use the DPU host's IP address (the tenant cluster's\nhost internal IP address) instead of the DPU's external bridge IP\naddress for the node primary address annotations and the L3 gateway\nconfiguration.\n\nAn incorrect node primary address annotation and L3 gateway configuration\nresults in broken flows when using NodePort or ClusterIP service.\n\nSigned-off-by: William Zhao ","shortMessageHtmlLink":"Fix handling of DPU on node address change/setting"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"Y3Vyc29yOnYyOpK7MjAyMy0wOS0xMVQxOToyMjo0Ni4wMDAwMDBazwAAAAN-KSxr","startCursor":"Y3Vyc29yOnYyOpK7MjAyMy0wOS0xMVQxOToyMjo0Ni4wMDAwMDBazwAAAAN-KSxr","endCursor":"Y3Vyc29yOnYyOpK7MjAyMy0wNS0wNFQyMjowNTozMy4wMDAwMDBazwAAAAMmYOqi"}},"title":"Activity · SalDaniele/ovn-kubernetes"}