{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":534609533,"defaultBranch":"master","name":"cinder-operator","ownerLogin":"Akrog","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2022-09-09T11:10:18.000Z","ownerAvatar":"https://github.com/avatars/u/5928725?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1727270252.0","currentOid":""},"activityList":{"items":[{"before":null,"after":"82a8c7e31703caad81348624272b96d3a23b644e","ref":"refs/heads/ocp-fix-nvme","pushedAt":"2024-09-25T13:17:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"NVMe sample: Fix nvme ids\n\nOpenShift image has the NVMe `hostid` and `hostnqn` hardcoded, so it\nends up being the same in all the control plane nodes, which is\nproblematic.\n\nThis `MachineConfig` fixes this issue by recreating both files when the\n`hostid` doesn't match the system-uuid of the machine it is running on.\n\nBugs:\n- https://issues.redhat.com/browse/OCPBUGS-34629\n- https://github.com/openshift/os/issues/1519\n- https://github.com/openshift/os/pull/1520","shortMessageHtmlLink":"NVMe sample: Fix nvme ids"}},{"before":null,"after":"e5d776dc67dea613bf278693dac29d636ad3a92b","ref":"refs/heads/general-docs","pushedAt":"2024-09-19T16:16:53.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Docs: Add generic guide\n\nThis patch adds documentation to explain concepts needed to understand\nthe configuration of the Block Storage (cinder) service on OpenShift.\n\nThere is general information meant for users explaining extra volumes,\nplacement restrictions, snippets, etc., as well as cinder specific\ntopics.","shortMessageHtmlLink":"Docs: Add generic guide"}},{"before":"ff7f871352c6bba5a3663713a92914f8a5feb544","after":"4eeef99b7eb39a2d5281b3e947bb1effb53f918b","ref":"refs/heads/containerImage","pushedAt":"2024-09-19T12:08:58.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Update backend samples using custom images\n\nThe way custom images are specified has changed, and now it no longer\nuses the `containerImage` file in the cinder sections but instead needs\nto be specified in the `OpenStackVersion`.\n\nThis patch updates the examples to match the latest code.","shortMessageHtmlLink":"Update backend samples using custom images"}},{"before":null,"after":"ff7f871352c6bba5a3663713a92914f8a5feb544","ref":"refs/heads/containerImage","pushedAt":"2024-07-10T15:56:41.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Update backend samples using custom images\n\nTe way custom images are specified has changed, and now it no longer\nuses the `containerImage` file in the cinder sections but instead needs\nto be specified in the `OpenStackVersion`.\n\nThis patch updates the examples to match the latest code.","shortMessageHtmlLink":"Update backend samples using custom images"}},{"before":null,"after":"97b7ab02f89dbd75f6a63b644a821937b3be0d70","ref":"refs/heads/remove-icon","pushedAt":"2024-07-08T15:37:08.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Remove the cinder icon field from the CSV\n\nThere are legal reasons why the cinder project mascot cannot be used as\nan icon in the operator, so this patch removes it.\n\nThe icon is not really important, since it's the openstack-operator the\none that should be important in the catalog.\n\nJira: OSPRH-8413","shortMessageHtmlLink":"Remove the cinder icon field from the CSV"}},{"before":"99be1db480cfc5c3e9e50a273b297eefee106712","after":null,"ref":"refs/heads/crypsetup-nsenter","pushedAt":"2024-06-28T15:31:59.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"}},{"before":null,"after":"99be1db480cfc5c3e9e50a273b297eefee106712","ref":"refs/heads/crypsetup-nsenter","pushedAt":"2024-06-28T12:47:26.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix attaching encrypted volume on controller nodes\n\nJust like with iscsiadm and other command tools there is a need to run\nthe command on the host's namespace.\n\nIf we don't run it like that the command will get stuck in:\n\n  cryptsetup luksOpen --key-file=- /dev/dm-8 crypt-os-brick+dev+dm-8\n\nAnd if we manually run the command with debug option (`--debug\n--verbose`) we'll see that it gets stuck in:\n\n  # Udev cookie 0xd4d73ad (semid 3) waiting for zero\n\nAnd is unable to continue with the rest:\n\n  # Udev cookie 0xd4d3938 (semid 3) destroyed\n  # Releasing crypt device /dev/mapper/3600140584b9fc02da024f9c8130ce253 context.\n  # Releasing device-mapper backend.\n  # Closing read only fd for /dev/mapper/3600140584b9fc02da024f9c8130ce253.\n  Command successful.\n\nThis patch ensure we use `nsenter` to run `cryptsetup`.","shortMessageHtmlLink":"Fix attaching encrypted volume on controller nodes"}},{"before":"07b9e7dc39860b025a1d3b84d84ce4b0bcccf281","after":null,"ref":"refs/heads/fix-api-logfile","pushedAt":"2024-06-26T14:44:31.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"}},{"before":"9c249ae096c20b18f77b3c832002dbb9c6752e31","after":"07b9e7dc39860b025a1d3b84d84ce4b0bcccf281","ref":"refs/heads/fix-api-logfile","pushedAt":"2024-06-26T12:42:45.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix api log file increases indefinitely\n\nThe cinder-api pod uses a file between the cinder-api service and a\nsidecar to pass the cinder logs from the service to the stdout.\n\nWith the current code it will grow forever, which makes no sense, since\nthe stdout from the pod itself is already being stored, under\n/var/log/pods and/or in a centralized logging service.\n\nThis patch makes the oslo.log service rotate the log file once it\nreaches 20MB in size.\n\nOther projects like nova and glance set the limit to 50MB and they keep\n5 rotated files, but I think that keeping rotated files is useless here,\nand 50MB seems a bit much.\n\nUnfortunately we need to at least keep 1 copy, or oslo.log will not do\nthe right thing if set to 0.\n\nWe also need to redirect errors in the sidecar container that tails the\nlog to /dev/null to prevent tail error messages whenever the file gets\nrotated.","shortMessageHtmlLink":"Fix api log file increases indefinitely"}},{"before":null,"after":"9c249ae096c20b18f77b3c832002dbb9c6752e31","ref":"refs/heads/fix-api-logfile","pushedAt":"2024-06-20T18:21:17.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix api log file increases indefinitely\n\nThe cinder-api pod uses a file between the cinder-api service and a\nsidecar to pass the cinder logs from the service to the stdout.\n\nWith the current code it will grow forever, which makes no sense, since\nthe stdout from the pod itself is already being stored, under\n/var/log/pods and/or in a centralized logging service.\n\nThis patch makes the oslo.log service rotate the log file once it\nreaches 20MB in size.\n\nOther projects like nova and glance set the limit to 50MB and they keep\n5 rotated files, but I think that keeping rotated files is useless here,\nand 50MB seems a bit much.\n\nUnfortunately we need to at least keep 1 copy, or oslo.log will not do\nthe right thing if set to 0.\n\nWe also need to redirect errors in the sidecar container that tails the\nlog to /dev/null to prevent tail error messages whenever the file gets\nrotated.","shortMessageHtmlLink":"Fix api log file increases indefinitely"}},{"before":"1d098ffa071f0d91475e2d7e5d7fb1550b8c0074","after":"86479bbe8bedf418e2b631e8ac6114ef648eb8da","ref":"refs/heads/fix-backend-name-custom-cr-name","pushedAt":"2024-06-19T15:28:51.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Correctly remove DB finalizers\n\nOn commit a8892c746cdb7dad60700cad867fd7171637b55e we forgot to update\nthe name of the MariaDBAccount when removing the finalizer.\n\nThis patch uses the \"cinder\" MariaDBAccount name since that's the name\nwe always use instead of it being based on the `Cinder` instance name.","shortMessageHtmlLink":"Correctly remove DB finalizers"}},{"before":"846fc93332dea91e864db1575115625c8572f60c","after":"1d098ffa071f0d91475e2d7e5d7fb1550b8c0074","ref":"refs/heads/fix-backend-name-custom-cr-name","pushedAt":"2024-06-19T15:21:03.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Correctly remove DB finalizers\n\nOn commit a8892c746cdb7dad60700cad867fd7171637b55e we forgot to update\nthe name of the MariaDBAccount when removing the finalizer.\n\nThis patch uses the \"cinder\" MariaDBAccount name since that's the name\nwe always use instead of it being based on the `Cinder` instance name.","shortMessageHtmlLink":"Correctly remove DB finalizers"}},{"before":"8574921624dc7211e40ea94c2d4c7744426bb681","after":"846fc93332dea91e864db1575115625c8572f60c","ref":"refs/heads/fix-backend-name-custom-cr-name","pushedAt":"2024-06-18T15:38:44.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix backend label when using non cinder name\n\nWhen the Cinder CR is created using a name that is not \"cinder\" the\n`backend` label in the StatefulSet and the final cinder volume Pod will\nhave the wrong value.\n\nThis is because the code that determines the backend name in the cinder\nvolume controller doesn't have access to the actual name, so it tries to\nfigure it out from the name of the CinderVolume CR, but since the name\nno longer starts with `cinder-volume`, then the value is wrong.\n\nThis patch changes the way we do this, and now the cinder-operator adds\na label to the CinderVolume CR with the name of the backend, then the\ncinder volume controller uses it to pass it to the statefulset .\n\nJira: OSPRH-7396","shortMessageHtmlLink":"Fix backend label when using non cinder name"}},{"before":"f1f4e289dc2280adc240e866346ca1317ce08cbe","after":null,"ref":"refs/heads/iscsid-login-timeout","pushedAt":"2024-06-12T18:55:05.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"}},{"before":null,"after":"8574921624dc7211e40ea94c2d4c7744426bb681","ref":"refs/heads/fix-backend-name-custom-cr-name","pushedAt":"2024-06-12T18:01:42.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix backend label when using non cinder name\n\nWhen the Cinder CR is created using a name that is not \"cinder\" the\n`backend` label in the StatefulSet and the final cinder volume Pod will\nhave the wrong value.\n\nThis is because the code that determines the backend name in the cinder\nvolume controller doesn't have access to the actual name, so it tries to\nfigure it out from the name of the CinderVolume CR, but since the name\nno longer starts with `cinder-volume`, then the value is wrong.\n\nThis patch changes the way we do this, and now the cinder-operator adds\na label to the CinderVolume CR with the name of the backend, then the\ncinder volume controller uses it to pass it to the statefulset .\n\nJira: OSPRH-7396","shortMessageHtmlLink":"Fix backend label when using non cinder name"}},{"before":null,"after":"a8892c746cdb7dad60700cad867fd7171637b55e","ref":"refs/heads/fix-cr-name","pushedAt":"2024-06-12T14:48:17.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Always name database \"cinder\"\n\nCurrent code names the database for cinder based on the Cinder CR name,\nand this is ok if the instance is called cinder, but if the name is\nsomething like \"mycloud-cinder\" then the database creation doesn't work\nas it fails with:\n\n  ERROR 1064 (42000) at line 1: You have an error in your SQL syntax;\n  check the manual that corresponds to your MariaDB server version for\n  the right syntax to use near '-cinder' at line 1\n\nWe have a constant called `DatabaseName` that contains the name that we\nshould be using for the database creation, and this patch uses it.\n\nJira: OSPRH-7396","shortMessageHtmlLink":"Always name database \"cinder\""}},{"before":null,"after":"65140f2eff475779d024142c8df8ef3f5b9bde0a","ref":"refs/heads/fix-apitimeout-default","pushedAt":"2024-06-10T08:56:34.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Set APITimeout defaults on the webhook\n\nKubebuilder doesn't always set the default of APITimeout, so with this\npatch we ensure that APITimeout field always has a valid value.","shortMessageHtmlLink":"Set APITimeout defaults on the webhook"}},{"before":"bfb7a850697803865eaee536bd80210dd49a075f","after":null,"ref":"refs/heads/apiTimeout","pushedAt":"2024-06-07T13:16:39.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"}},{"before":null,"after":"f1f4e289dc2280adc240e866346ca1317ce08cbe","ref":"refs/heads/iscsid-login-timeout","pushedAt":"2024-06-06T15:46:48.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Samples: iscsid login timeout\n\nThis patch changes the time to fail an iSCSI login request in our\nsamples.\n\nThe default is 8 retries and 15 seconds for each one, so 120 seconds in\ntotal.\n\nThis patch changes the OpenShift cluster default to 3 retries and 5\nseconds each (15 seconds in total), which is convenient for testing, as\nany healthy deployment and backend should be able to login to the\nbackend in that amount of time, and if there is a broken path it will\nnot take 2 minutes to give up.\n\nJira: https://issues.redhat.com/browse/OSPRH-7393\nJira: https://issues.redhat.com/browse/OSPRH-7415","shortMessageHtmlLink":"Samples: iscsid login timeout"}},{"before":null,"after":"bfb7a850697803865eaee536bd80210dd49a075f","ref":"refs/heads/apiTimeout","pushedAt":"2024-06-06T14:11:38.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Add configurable API Timeouts\n\nMultiple limitations exist with existing code:\n\n- Human operator cannot configure Apache Timeout, so it's fixed at the\n  defaults of 60 seconds.\n\n  This can be problematic for cinder backends that under load can take\n  longer on the export and mapping.\n\n- Default HAProxy timeout is 30 seconds, which is different than the\n  default in OSP17 (60 seconds)\n\n- Human operators would need to sync values between 3 places: HAProxy,\n  Apache, oslo.messaging sync RPC (call).\n\nThis patch adds an `apiTimeout` field to the `CinderSpecBase` to allow\nhuman operators to simultaneously configure the timeouts for HAProxy,\nApache, and the `rpc_response_timeout`.\n\nThe `apiTimeout` defaults to 60 seconds, to mimic the behavior present\nin OSP17.\n\nHaving different timeouts for HAProxy, Apache, and\n`rpc_response_timeout` is possible setting the HAProxy timeout in the\n`apiOverride`, the Apache timeout with `apiTimeout`, and setting\n`rpc_response_timeout` in the top level Cinder `customServiceConfig`.\n\nTo be able to change the HAProxy value based on the `apiTimeout` with\nany update (and not just the first time) the code adds a custom\nannotation \"api.cinder.openstack.org/timeout\" with the value that was\ninitially set, this way flags it as being set by the cinder-operator.\n\nJira: https://issues.redhat.com/browse/OSPRH-7393","shortMessageHtmlLink":"Add configurable API Timeouts"}},{"before":"803ef0777a8e05c9aeeacdd32cefd390eb1cb236","after":"b0747944394807862e4cdcfa7052f1f8d1febf94","ref":"refs/heads/change-db-wait","pushedAt":"2024-05-29T14:32:38.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Wait for DB writes only on reads (causality checks)\n\nOn PR 392 we changed to waiting on writes for updates and inserts, but\nupon further analysis we have decided that just doing it for reads\nshould be enough, as the order of the updates will be reconciled\nautomatically by the DB engine.\n\nIn this patch we change the value of `mysql_wsrep_sync_wait` to 1 to\nonly wait on reads.","shortMessageHtmlLink":"Wait for DB writes only on reads (causality checks)"}},{"before":null,"after":"803ef0777a8e05c9aeeacdd32cefd390eb1cb236","ref":"refs/heads/change-db-wait","pushedAt":"2024-05-29T14:31:29.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Wait for DB writes only on reads (causality checks)\n\nOn PR 392 we changed to waiting on writes for updates and inserts, but\nupon further analysis we have decided that just doing it for reads\nshould be enough, as the order of the updates will be reconciled\nautomatically by the DB engine.\n\nIn this patch we change the value of `mysql_wsrep_sync_wait` to 1 to\nonly wait on reads.","shortMessageHtmlLink":"Wait for DB writes only on reads (causality checks)"}},{"before":"67b2877d6edeb6ed633c529afdc40f954714654b","after":"dd24e2d31d9ac2541e724ebf35299010df169f74","ref":"refs/heads/fix-workers-issues","pushedAt":"2024-05-29T14:16:26.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Wait for DB writes to propagate (causality checks)\n\nBecause we deploy the database in multi-master mode we can have cases\nwhere a service writes something in the database and when another one\nreads from the DB that data is not yet there.\n\nThis is very problematic, because all the cinder code assumes that that\ncan never happen.\n\nFor example we've seen in CI jobs the following behavior:\n\n- Cinder api creates a worker registry in the DB when deleting a volume\n  on DB node 1.\n\n- Cinder api makes an RPC call to cinder-volume to delete the volume.\n\n- Cinder volume tries to read the worker registry from DB node 2 but\n  the data is not there yet, so it misbehaves.\n\nIn this patch we change the default value on the DB engine of not\nwaiting for writes to wait on read or on update\n\nThis will have a performance impact, but the alternative is for cinder\nto misbehave.\n\nWe use `mysql_wsrep_sync_wait` from oslo.db [1] setting it to 3 as per\nthe documented values of this parameter in the DBMS [2] [3].\n\nWe don't expect inserts to need this.\n\n[1]: https://opendev.org/openstack/oslo.db/commit/009d23df45969036c70e4cf59eb4019aaace9a55\n[2]: https://mariadb.com/docs/server/ref/mdb/system-variables/wsrep_sync_wait/\n[3]: https://galeracluster.com/library/documentation/mysql-wsrep-options.html","shortMessageHtmlLink":"Wait for DB writes to propagate (causality checks)"}},{"before":null,"after":"67b2877d6edeb6ed633c529afdc40f954714654b","ref":"refs/heads/fix-workers-issues","pushedAt":"2024-05-29T10:14:25.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Wait for DB writes to propagate (causality checks)\n\nBecause we deploy the database in multi-master mode we can have cases\nwhere a service writes something in the database and when another one\nreads from the DB that data is not yet there.\n\nThis is very problematic, because all the cinder code assumes that that\ncan never happen.\n\nFor example we've seen in CI jobs the following behavior:\n\n- Cinder api creates a worker registry in the DB when deleting a volume\n  on DB node #1.\n\n- Cinder api makes an RPC call to cinder-volume to delete the volume.\n\n- Cinder volume tries to read the worker registry from DB node #2 but\n  the data is not there yet, so it misbehaves.\n\nIn this patch we change the default value on the DB engine of not\nwaiting for writes to wait on read, update, and insert.\n\nThis will have a performance impact, but the alternative is for cinder\nto misbehave.\n\nWe use `mysql_wsrep_sync_wait` from oslo.db [1] setting it to 7 as per\nthe documented values of this parameter in the DBMS [2][3].\n\n[1]: https://opendev.org/openstack/oslo.db/commit/009d23df45969036c70e4cf59eb4019aaace9a55\n[2]: https://mariadb.com/docs/server/ref/mdb/system-variables/wsrep_sync_wait/\n[3]: https://galeracluster.com/library/documentation/mysql-wsrep-options.html","shortMessageHtmlLink":"Wait for DB writes to propagate (causality checks)"}},{"before":"901e9660552e4b1885faf5704e56885b8fc50cce","after":"5d2ab3b0c8b4e7b806c7f2337c59517d771a94d1","ref":"refs/heads/update-quickstart","pushedAt":"2024-05-14T09:42:02.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"README cleanup\n\nOur current README.md file does not provide a quick guide to start\npaying with OpenStack and a Cinder backend.\n\nIn this patch we update the README.md to present a quick starting guide\nthat deploys OpenStack with LVM as a block device backend.\n\nIt also moves a number of sections to their own independent files as a\ncleanup of the top level README.","shortMessageHtmlLink":"README cleanup"}},{"before":null,"after":"901e9660552e4b1885faf5704e56885b8fc50cce","ref":"refs/heads/update-quickstart","pushedAt":"2024-05-09T16:17:13.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"README cleanup\n\nOur current README.md file does not provide a quick guide to start\npaying with OpenStack and a Cinder backend.\n\nIn this patch we update the README.md to present a quick starting guide\nthat deploys OpenStack with LVM as a block device backend.\n\nIt also moves a number of sections to their own independent files as a\ncleanup of the top level README.","shortMessageHtmlLink":"README cleanup"}},{"before":"dda6c1c72e5b558bcf5a39b0a40e39d831b133e0","after":"6ed45d8b64cf9cc4caacc07a8551e977cfac3860","ref":"refs/heads/improvelogs","pushedAt":"2024-04-30T16:35:25.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Remote useless reconciliation methods\n\nThere are a number of placeholder reconciliation method that do nothing\nexcept logging messages of how the method has been entered and exited.\n\nThis is useless code as well as clutter in the logs, so this patch\nremoves them.\n\nAffected methods are:\n\n- reconcileInit\n- reconcileUpdate\n- reconcileUpgrade\n\nThere are instances where reconcileInit is actually doing things, and in\nthose cases methods were left untouched.","shortMessageHtmlLink":"Remote useless reconciliation methods"}},{"before":"60d271e4fb9f3c34b5661d18da65ee8699b1178c","after":"dda6c1c72e5b558bcf5a39b0a40e39d831b133e0","ref":"refs/heads/improvelogs","pushedAt":"2024-04-30T16:09:35.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Remote useless reconciliation methods\n\nThere are a number of placeholder reconciliation method that do nothing\nexcept logging messages of how the method has been entered and exited.\n\nThis is useless code as well as clutter in the logs, so this patch\nremoves them.\n\nAffected methods are:\n\n- reconcileInit\n- reconcileUpdate\n- reconcileUpgrade\n\nThere are instances where reconcileInit is actually doing things, and in\nthose cases methods were left untouched.","shortMessageHtmlLink":"Remote useless reconciliation methods"}},{"before":"3b05b9d64689f7df2a1c04ee5e333684ba3ba34d","after":"7aba2b7de5cbbfbd665367b5948790aa6143296a","ref":"refs/heads/more-fixing-conditions","pushedAt":"2024-04-29T16:48:38.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix status conditions\n\nThis patch leverages the ObservedGeneration to fix the status\nconditions.\n\nTop level cinder CRD cannot use the information of the sub CRDs content\nunless the Generation and ObservedGeneration match, and it will only be\nready if it's generation is right and all the subCRDs are ready.\n\nFor the sub CRDs to be ready they need to:\n\n- Have observed and reacted to the latest Generation\n- Have the right number of replicas\n- Be deployment ready or have requested 0 replicas and not be deployment\n  ready","shortMessageHtmlLink":"Fix status conditions"}},{"before":"01192fd4e6988b8f51ff31eb89e80d35b3d7d48d","after":"3b05b9d64689f7df2a1c04ee5e333684ba3ba34d","ref":"refs/heads/more-fixing-conditions","pushedAt":"2024-04-24T10:43:37.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Akrog","name":"Gorka Eguileor","path":"/Akrog","primaryAvatarUrl":"https://github.com/avatars/u/5928725?s=80&v=4"},"commit":{"message":"Fix status conditions\n\nThis patch leverages the ObservedGeneration to fix the status\nconditions.\n\nTop level cinder CRD cannot use the information of the sub CRDs content\nunless the Generation and ObservedGeneration match, and it will only be\nready if it's generation is right and all the subCRDs are ready.\n\nFor the sub CRDs to be ready they need to:\n\n- Have observed and reacted to the latest Generation\n- Have the right number of replicas\n- Be deployment ready or have requested 0 replicas and not be deployment\n  ready","shortMessageHtmlLink":"Fix status conditions"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"startCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0yNVQxMzoxNzozMi4wMDAwMDBazwAAAATAWfAI","endCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNC0yNFQxMDo0MzozNy4wMDAwMDBazwAAAAQ5A6AZ"}},"title":"Activity · Akrog/cinder-operator"}