{"id":50352,"date":"2023-11-06T12:03:36","date_gmt":"2023-09-14T06:48:08","guid":{"rendered":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/"},"modified":"2024-05-04T07:48:24","modified_gmt":"2024-05-03T23:48:24","slug":"%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d","status":"publish","type":"post","link":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/","title":{"rendered":"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d"},"content":{"rendered":"<h1>\u9996\u5148<\/h1>\n<p>\u5728\u786e\u8ba4\u4e86\u4ece\u505c\u7535\u4e2d\u6062\u590d\u7684Kubernetes\u96c6\u7fa4\u72b6\u6001\u65f6\uff0c\u6211\u53d1\u73b0\u53ea\u6709\u4e00\u4e2aPod\uff08gitbucket\uff09\u4ecd\u5904\u4e8eContainerCreating\u72b6\u6001\u5e76\u957f\u65f6\u95f4\u505c\u6b62\u3002<\/p>\n<p>\u9020\u6210\u8fd9\u4e2a\u95ee\u9898\u7684\u539f\u56e0\u662fPVC\u65e0\u6cd5\u6302\u8f7d\uff0c\u7136\u800c\uff0c\u5b9e\u9645\u539f\u56e0\u5728\u4e8e\u505c\u7535\u5bfc\u81f4ext4\u7684\u65e5\u5fd7\u6570\u636e\u4ecd\u7136\u4fdd\u7559\u4e0b\u6765\u4e86\u3002<\/p>\n<p>\u6211\u5f88\u5feb\u5c31\u77e5\u9053\u53ea\u9700\u8981\u8fd0\u884cfsck.ext4\u6765\u8fdb\u884c\u4fee\u590d\uff0c\u4f46\u662f\u7531\u4e8e\u4e0d\u76f4\u89c2\uff0c\u4e0d\u592a\u6e05\u695a\u5e94\u8be5\u5bf9\u54ea\u4e2a\u5377\u6267\u884c\uff0c\u6240\u4ee5\u6211\u7559\u4e0b\u4e86\u5907\u5fd8\u5f55\u3002<\/p>\n<h2>\u73af\u5883\u3002<\/h2>\n<ul class=\"post-ul\">\n<li style=\"list-style-type: none;\">\n<ul class=\"post-ul\">Kubernetes v1.16.9 deployed by Kubespray<\/ul>\n<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<ul class=\"post-ul\">Rook\/Ceph v1.1.9 with FlexVolumes (BlockStorage using ext4)<\/ul>\n<p>\u7531\u4e8e\u5206\u533a\u7834\u574f\u6240\u5bfc\u81f4\u7684\u6570\u636e\u635f\u574f\uff0c\u5728\u4ee5\u4e0b\u7248\u672c\u4e2d\u53d1\u751f\u3002<\/p>\n<ul class=\"post-ul\">Rook\/Ceph v1.6.7 with CSI BlueStore (with HDD 4TBx2)<\/ul>\n<h2>\u60c5\u51b5<\/h2>\n<p>\u901a\u8fc7\u4f7f\u7528kubectl\u547d\u4ee4\u8fdb\u884c\u786e\u8ba4\uff0c\u6211\u4eec\u53ef\u4ee5\u770b\u5230\u5982\u4e0b\u6240\u793a\u7684\u72b6\u6001\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> gitbucket get all\r\nNAME                         READY   STATUS              RESTARTS   AGE\r\npod\/bucket-9b8d9799-khsm8    0\/1     Init:1\/2            1          27h\r\npod\/mysql-84d459c89c-w8pb6   0\/1     ContainerCreating   0          15m\r\n<\/code><\/pre>\n<p>Pod: bucket\u5728initContainers\u4e2d\u7b49\u5f85mysql\u542f\u52a8\uff0c\u6240\u4ee5\u5b83\u4fdd\u6301\u5728\u521d\u59cb\u72b6\u6001\u5e76\u505c\u6b62\u8fd0\u884c\u3002\u4e00\u65e6mysql\u5bb9\u5668\u542f\u52a8\uff0cbucket\u5bb9\u5668\u5e94\u8be5\u81ea\u52a8\u6062\u590d\uff0c\u56e0\u6b64\u6211\u4eec\u9996\u5148\u8981\u6062\u590dmysql\u5bb9\u5668\u3002<\/p>\n<h1>BlockStorage\u51fa\u73b0\u6545\u969c\u4e86\u3002<\/h1>\n<h2>\u9519\u8bef\u72b6\u51b5<\/h2>\n<p>\u4f7f\u7528kubectl describe\u547d\u4ee4\u67e5\u8be2\u542f\u52a8\u72b6\u6001\u65f6\uff0c\u7ed3\u679c\u5982\u4e0b\u6240\u793a\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> gitbucket describe pod\/mysql-84d459c89c-w8pb6\r\n....\r\nEvents:\r\n  Type     Reason       Age                  From               Message\r\n  <span class=\"nt\">----<\/span>     <span class=\"nt\">------<\/span>       <span class=\"nt\">----<\/span>                 <span class=\"nt\">----<\/span>               <span class=\"nt\">-------<\/span>\r\n  Normal   Scheduled    16m                  default-scheduler  Successfully assigned gitbucket\/mysql-84d459c89c-w8pb6 to node04\r\n  Warning  FailedMount  15m <span class=\"o\">(<\/span>x5 over 16m<span class=\"o\">)<\/span>    kubelet, node04  MountVolume.SetUp failed <span class=\"k\">for <\/span>volume <span class=\"s2\">\"pvc-8aaa318c-1212-415d-b00f-e88b9c3c3087\"<\/span> : mount <span class=\"nb\">command <\/span>failed, status: Failure, reason: Rook: Mount volume failed: failed to attach volume pvc-8aaa318c-1212-415d-b00f-e88b9c3c3087 <span class=\"k\">for <\/span>pod gitbucket\/mysql-84d459c89c-w8pb6. Volume is already attached by pod gitbucket\/mysql-84d459c89c-rllg6. Status Pending\r\n  Warning  FailedMount  2m46s <span class=\"o\">(<\/span>x3 over 11m<span class=\"o\">)<\/span>  kubelet, node04  Unable to attach or mount volumes: unmounted <span class=\"nv\">volumes<\/span><span class=\"o\">=[<\/span>mysql-persistent-storage], unattached <span class=\"nv\">volumes<\/span><span class=\"o\">=[<\/span>database-auth-conf default-token-txprc mysql-persistent-storage]: timed out waiting <span class=\"k\">for <\/span>the condition\r\n  Warning  FailedMount  92s <span class=\"o\">(<\/span>x10 over 15m<span class=\"o\">)<\/span>   kubelet, node04  MountVolume.SetUp failed <span class=\"k\">for <\/span>volume <span class=\"s2\">\"pvc-8aaa318c-1212-415d-b00f-e88b9c3c3087\"<\/span> : mount <span class=\"nb\">command <\/span>failed, status: Failure, reason: failed to mount volume \/dev\/rbd2 <span class=\"o\">[<\/span>ext4] to \/var\/lib\/kubelet\/plugins\/ceph.rook.io\/rook-ceph\/mounts\/pvc-8aaa318c-1212-415d-b00f-e88b9c3c3087, error <span class=\"s1\">'fsck'<\/span> found errors on device \/dev\/rbd2 but could not correct them: fsck from util-linux 2.31.1\r\n\/dev\/rbd2: Superblock needs_recovery flag is clear, but journal has data.\r\n\/dev\/rbd2: Run journal anyway\r\n\r\n\/dev\/rbd2: UNEXPECTED INCONSISTENCY<span class=\"p\">;<\/span> RUN fsck MANUALLY.\r\n  <span class=\"o\">(<\/span>i.e., without <span class=\"nt\">-a<\/span> or <span class=\"nt\">-p<\/span> options<span class=\"o\">)<\/span>\r\n<span class=\"nb\">.<\/span> \r\n  Warning  FailedMount  32s <span class=\"o\">(<\/span>x4 over 14m<span class=\"o\">)<\/span>  kubelet, node04  Unable to attach or mount volumes: unmounted <span class=\"nv\">volumes<\/span><span class=\"o\">=[<\/span>mysql-persistent-storage], unattached <span class=\"nv\">volumes<\/span><span class=\"o\">=[<\/span>mysql-persistent-storage database-auth-conf default-token-txprc]: timed out waiting <span class=\"k\">for <\/span>the condition\r\n<\/code><\/pre>\n<p>\u4ece\u8fd9\u4e2a\u9519\u8bef\u6d88\u606f\u4e2d\u53ef\u4ee5\u770b\u51fa\uff0c\u5728\u8282\u70b9node04\u548c**\/dev\/rbd2**\u4e0a\u51fa\u73b0\u4e86\u95ee\u9898\u3002<\/p>\n<h2>\u5728node04\u4e0a\u7684\u4fee\u6539\u3002<\/h2>\n<p>\u8d77\u521d\uff0c\u6211\u4e0d\u77e5\u9053\u8be5\u5982\u4f55\u5904\u7406\uff0c\u4f46\u662f\u5f53\u6211\u767b\u5f55\u5230node04\u5e76\u68c0\u67e5\u4e86\/dev\/\u76ee\u5f55\u540e\uff0c\u53d1\u73b0\u4e86rbd2\u8fd9\u4e2a\u5b58\u5728\uff0c\u4e8e\u662f\u6211\u51b3\u5b9a\u5c06\u5176\u4f5c\u4e3afsck.ext4\u547d\u4ee4\u7684\u53c2\u6570\u5c1d\u8bd5\u4e00\u4e0b\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>ssh node04\r\n<span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>fsck.ext4 \/dev\/rbd2\r\ne2fsck 1.44.1 <span class=\"o\">(<\/span>24-Mar-2018<span class=\"o\">)<\/span>\r\nSuperblock needs_recovery flag is clear, but journal has data.\r\nRun journal anyway&lt;y&gt;? <span class=\"nb\">yes<\/span>\r\n\/dev\/rbd2: recovering journal\r\nPass 1: Checking inodes, blocks, and sizes\r\nPass 2: Checking directory structure\r\nPass 3: Checking directory connectivity\r\nPass 4: Checking reference counts\r\nPass 5: Checking group summary information\r\nFree blocks count wrong <span class=\"o\">(<\/span>5067650, <span class=\"nv\">counted<\/span><span class=\"o\">=<\/span>5067649<span class=\"o\">)<\/span><span class=\"nb\">.<\/span>\r\nFix&lt;y&gt;? <span class=\"nb\">yes\r\n<\/span>Free inodes count wrong <span class=\"o\">(<\/span>1310519, <span class=\"nv\">counted<\/span><span class=\"o\">=<\/span>1310518<span class=\"o\">)<\/span><span class=\"nb\">.<\/span>\r\nFix&lt;y&gt;? <span class=\"nb\">yes<\/span>\r\n\r\n\/dev\/rbd2: <span class=\"k\">*****<\/span> FILE SYSTEM WAS MODIFIED <span class=\"k\">*****<\/span>\r\n\/dev\/rbd2: 202\/1310720 files <span class=\"o\">(<\/span>14.9% non-contiguous<span class=\"o\">)<\/span>, 175231\/5242880 blocks\r\n<\/code><\/pre>\n<p>Pod\u5b89\u5168\u8fd0\u884c\u8d77\u6765\u540e\uff0c\u65f6\u95f4\u5c31\u8fd9\u6837\u8fc7\u53bb\u4e86\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> gitbucket get all\r\nNAME                         READY   STATUS     RESTARTS   AGE\r\npod\/bucket-9b8d9799-khsm8    0\/1     Init:1\/2   1          27h\r\npod\/mysql-84d459c89c-w8pb6   0\/1     Running    0          27m\r\n<\/code><\/pre>\n<p>\u4f7f\u7528XFS\u6587\u4ef6\u7cfb\u7edf\u8bbe\u7f6e\u53ef\u80fd\u66f4\u597d\u4e00\u70b9\uff0c\u5c3d\u7ba1\uff08\u5373\u4f7f\u4f7f\u7528XFS\uff09\u53ef\u80fd\u4e5f\u4f1a\u51fa\u73b0\u95ee\u9898\uff0c\u4f46\u6211\u7565\u611f\u540e\u6094\u3002<\/p>\n<p>\u4ee5\u4e0a<\/p>\n<h1>\u968f\u540e\u53d1\u751f\u7684\u6545\u969c<\/h1>\n<p>\u8fd9\u4e00\u6b21\uff0c\u5728Harbor\u4e0a\u4f7f\u7528\u7684PVC\u65e0\u6cd5\u8fdb\u884c\u6302\u8f7d\uff0c\u4f46\u539f\u56e0\u5374\u662f\u53e6\u5916\u7684\u3002<\/p>\n<pre class=\"post-pre\"><code>$ kubectl -n harbor describe pod\/my-harbor-harbor-redis-0\r\n...\r\nEvents:\r\n  Type     Reason       Age                      From               Message\r\n  ----     ------       ----                     ----               -------\r\n  Warning  FailedMount  12m (x228 over 7h42m)    kubelet, node01  MountVolume.SetUp failed for volume \"pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7\" : mount command failed, status: Failure, reason: Rook: Mount volume failed: failed to attach volume pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7 for pod harbor\/my-harbor-harbor-redis-0. Volume is already attached by pod harbor\/my-harbor-harbor-redis-0. Status Pending\r\n  Warning  FailedMount  7m18s (x151 over 7h40m)  kubelet, node01  Unable to attach or mount volumes: unmounted volumes=[data], unattached volumes=[data default-token-nnqmn]: timed out waiting for the condition\r\n  Warning  FailedMount  2m47s (x51 over 7h38m)   kubelet, node01  Unable to attach or mount volumes: unmounted volumes=[data], unattached volumes=[default-token-nnqmn data]: timed out waiting for the condition\r\n<\/code><\/pre>\n<p>\u6211\u5c06\u786e\u8ba4\u6b64\u65f6 PV: pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7 \u7684\u72b6\u6001\u5982\u4f55\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> harbor get pv pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7\r\nNAME                                       CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM          \r\n                        STORAGECLASS      REASON   AGE\r\npvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7   1Gi        RWO            Delete           Bound    harbor\/data-my-harbor-harbor-redis-0   rook-ceph-block            232d\r\n<\/code><\/pre>\n<p>\u4eceRook\/Ceph\u7684\u89d2\u5ea6\u6765\u770b\uff0c\u6ca1\u6709\u51fa\u73b0\u7279\u6b8a\u95ee\u9898\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> harbor get pvc data-my-harbor-harbor-redis-0\r\nNAME                            STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS      AGE\r\ndata-my-harbor-harbor-redis-0   Bound    pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7   1Gi        RWO            rook-ceph-block   232d\r\n\r\n<span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph <span class=\"nb\">exec<\/span> <span class=\"nt\">-it<\/span> rook-ceph-tools-7764cdbbc4-wkc64 ceph status\r\n  cluster:\r\n    <span class=\"nb\">id<\/span>:     89ddf8a3-e7ce-41e6-a081-b6b5fab8b8e0\r\n    health: HEALTH_OK\r\n \r\n  services:\r\n    mon: 3 daemons, quorum v,w,y <span class=\"o\">(<\/span>age 7h<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 7h<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-a<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 4 osds: 4 up <span class=\"o\">(<\/span>since 7h<span class=\"o\">)<\/span>, 4 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 7h<span class=\"o\">)<\/span>\r\n \r\n  data:\r\n    pools:   13 pools, 104 pgs\r\n    objects: 2.64k objects, 6.1 GiB\r\n    usage:   567 GiB used, 6.6 TiB \/ 7.2 TiB avail\r\n    pgs:     104 active+clean\r\n \r\n  io:\r\n    client:   938 B\/s rd, 9.5 KiB\/s wr, 1 op\/s rd, 0 op\/s wr\r\n\r\n<span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph logs <span class=\"nt\">-l<\/span> rook-ceph-agent\r\n...\r\n2020-09-04 01:59:27.976756 I | flexdriver: calling agent to attach volume replicapool\/pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7\r\n2020-09-04 01:59:27.979645 I | flexvolume: volume attachment record rook-ceph\/pvc-d37fd3f0-22ac-42c2-a90d-7427c2fa91d7 exists <span class=\"k\">for <\/span>pod: harbor\/my-harbor-harbor-redis-0\r\n...\r\n<\/code><\/pre>\n<h2>\u5e94\u5bf9<\/h2>\n<p>\u56e0\u4e3a\u53ea\u60f3\u5230\u4e86\u4eceStatefulSet\u7684\u5b9a\u4e49\u4e2d\u5220\u9664\u5b83\u7684\u65b9\u6cd5\uff0c\u6240\u4ee5\u6211\u51b3\u5b9a\u4f7f\u7528helm\u5c06\u5176\u5f15\u5165\uff0c\u5e76\u4ecehelm\u4e2d\u5220\u9664\u5b83\uff0c\u7136\u540e\u91cd\u65b0\u5b89\u88c5\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"o\">(<\/span><span class=\"nb\">cd <\/span>harbor-helm<span class=\"p\">;<\/span> helm delete my-harbor<span class=\"o\">)<\/span>\r\n<span class=\"nv\">$ <\/span><span class=\"o\">(<\/span><span class=\"nb\">cd <\/span>harbor-helm<span class=\"p\">;<\/span> helm <span class=\"nb\">install<\/span> <span class=\"nt\">--replace<\/span> <span class=\"nt\">--name<\/span> my-harbor <span class=\"nt\">--namespace<\/span> harbor .<span class=\"o\">)<\/span>\r\n<\/code><\/pre>\n<p>\u679c\u7136\uff0c\u8fd9\u4e4b\u540e\u6ca1\u6709\u95ee\u9898\u5730\u5b89\u88c5\u4e86PVC\u3002<\/p>\n<p>\u6839\u636e\u8c03\u67e5\uff0c\u6211\u4eec\u53d1\u73b0\u6bcf\u4e2a\u8282\u70b9\u7684\u591c\u95f4\u91cd\u65b0\u542f\u52a8\u65f6\u95f4\u975e\u5e38\u63a5\u8fd1\uff0c\u56e0\u6b64\u6211\u4eec\u51b3\u5b9a\u5145\u5206\u6c9f\u901a\uff0c\u4ee5\u51cf\u5c11\u4e0d\u4e00\u81f4\uff0c\u5e76\u89c2\u5bdf\u4e00\u6bb5\u65f6\u95f4\u3002<\/p>\n<h1>\u6587\u4ef6\u7cfb\u7edf\u7684\u6545\u969c<\/h1>\n<p>\u53e6\u4e00\u4e2a Pod \u51fa\u73b0\u4e86\u65e0\u6cd5\u6302\u8f7d\u6587\u4ef6\u7cfb\u7edf\u7684\u60c5\u51b5\u3002<\/p>\n<pre class=\"post-pre\"><code>  Warning  FailedMount  13m  kubelet, node04  MountVolume.SetUp failed for volume \"rook-data\" : mount command failed, stat\r\nus: Failure, reason: failed to mount filesystem myfs to \/var\/lib\/kubelet\/pods\/297a0883-0d02-41c9-855b-b193f02210ec\/volumes\/c\r\neph.rook.io~rook\/rook-data with monitor 10.233.32.156:6789,10.233.62.103:6789,10.233.34.233:6789:\/ and options [name=admin s\r\necret=AQBuk+pctG\/7JBAATogIHHCDNN7W7u4zppFtaw== mds_namespace=myfs]: mount failed: exit status 32\r\nMounting command: systemd-run\r\nMounting arguments: --description=Kubernetes transient mount for \/var\/lib\/kubelet\/pods\/297a0883-0d02-41c9-855b-b193f02210ec\/\r\nvolumes\/ceph.rook.io~rook\/rook-data --scope -- mount -t ceph -o name=admin,secret=AQBuk+pctG\/7JBAATogIHHCDNN7W7u4zppFtaw==,m\r\nds_namespace=myfs 10.233.32.156:6789,10.233.62.103:6789,10.233.34.233:6789:\/ \/var\/lib\/kubelet\/pods\/297a0883-0d02-41c9-855b-b\r\n193f02210ec\/volumes\/ceph.rook.io~rook\/rook-data\r\nOutput: Running scope as unit: run-r35a1db0e9e564106b94b7ea5d1973fb3.scope\r\nmount: \/var\/lib\/kubelet\/pods\/297a0883-0d02-41c9-855b-b193f02210ec\/volumes\/ceph.rook.io~rook\/rook-data: mount(2) system call \r\nfailed: No route to host.\r\n  Warning  FailedMount  13m  kubelet, node04  MountVolume.SetUp failed for volume \"rook-data\" : mount command failed, stat\r\nus: Failure, reason: failed to mount filesystem myfs to \/var\/lib\/kubelet\/pods\/297a0883-0d02-41c9-855b-b193f02210ec\/volumes\/c\r\neph.rook.io~rook\/rook-data with monitor 10.233.32.156:6789,10.233.62.103:6789,10.233.34.233:6789:\/ and options [name=admin s\r\necret=AQBuk+pctG\/7JBAATogIHHCDNN7W7u4zppFtaw== mds_namespace=myfs]: mount failed: exit status 32\r\n<\/code><\/pre>\n<h2>\u5904\u7406<\/h2>\n<p>\u867d\u7136\u5728MDS\u8282\u70b9\u4e0a\u6ca1\u6709\u7279\u5b9a\u7684\u9519\u8bef\u6d88\u606f\uff0c\u4f46\u6211\u4f1a\u5c1d\u8bd5\u91cd\u65b0\u542f\u52a8\u8282\u70b9\u3002<br \/>\n\u4e3a\u4e86\u5b89\u5168\u8d77\u89c1\uff0c\u6211\u5c06\u4ece\u5907\u63f4\u8282\u70b9\u5f00\u59cb\u91cd\u65b0\u542f\u52a8\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph <span class=\"nb\">exec<\/span> <span class=\"nt\">-it<\/span> rook-ceph-tools-6dd7c595bc-xvk92 ceph status\r\n...\r\nservices:\r\n    mon: 3 daemons, quorum bm,bs,bu <span class=\"o\">(<\/span>age 15m<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 13m<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-a<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 4 osds: 4 up <span class=\"o\">(<\/span>since 60m<span class=\"o\">)<\/span>, 4 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 5w<span class=\"o\">)<\/span>\r\n...\r\n\r\n<span class=\"c\">## \"a=up:active\"\u306e\u8868\u793a\u304b\u3089\u3001b\u5074\u3092\u307e\u305a\u518d\u8d77\u52d5\u3057\u3066\u304a\u304f<\/span>\r\n<span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete pod rook-ceph-mds-myfs-b-78ccfc6754-jff2f\r\n\r\n<span class=\"c\">## kubectl -n rook-ceph get pod \u3067myfs-b\u304c\u6b63\u5e38\u306b\u52d5\u3044\u3066\u3044\u308b\u3053\u3068\u3092\u78ba\u8a8d\u3057\u3066\u304b\u3089\u3001a\u5074\u3092\u30ea\u30b9\u30bf\u30fc\u30c8\u3059\u308b<\/span>\r\n<span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete pod rook-ceph-mds-myfs-a-87448c57b-lp7hl\r\n<\/code><\/pre>\n<p>\u603b\u7684\u6765\u8bf4\uff0c\u6211\u5df2\u7ecf\u4f7f\u7528\u8fd9\u79cd\u65b9\u6cd5\u6210\u529f\u5730\u91cd\u65b0\u542f\u52a8\u4e86MDS\u8282\u70b9\uff0c\u5e76\u4e14\u53ef\u4ee5\u987a\u5229\u5730\u8bbf\u95ee\u6587\u4ef6\u7cfb\u7edf\u3002<\/p>\n<h1>\u5bf9\u4e8e HEALTH_ERR \u7684\u5904\u7406<\/h1>\n<p>\u5f53\u6211\u6ce8\u610f\u5230\u65f6\uff0c\u6574\u4e2a\u8282\u70b9\u7684\u529f\u80fd\u5df2\u7ecf\u505c\u6b62\uff0c\u5404\u79cdPod\u4ecd\u7136\u5904\u4e8e\u6302\u8d77\u6216\u7ec8\u6b62\u72b6\u6001\uff0c\u5e76\u4e14\u5728\u89e3\u51b3\u8fd9\u4e9b\u95ee\u9898\u4e4b\u540e\uff0cceph\u72b6\u6001\u51fa\u73b0\u4e86\u9519\u8bef\u3002<\/p>\n<p>\u9996\u5148\uff0c\u901a\u8fc7\u4f7f\u7528&#8221;shutdown -r now&#8221;\u547d\u4ee4\u91cd\u65b0\u542f\u52a8\u4e86\u6574\u4e2a\u7cfb\u7edf\u4ee5\u89e3\u51b3\u521d\u59cb\u7684\u8282\u70b9\u9519\u8bef\u3002\u5bf9\u4e8e\u6b64\u90e8\u5206\u7684\u6839\u672c\u539f\u56e0\uff0c\u6211\u6709\u4e00\u4e9b\u731c\u6d4b\uff0c\u4f46\u5e76\u672a\u6df1\u5165\u63a2\u7a76\u3002<\/p>\n<p>\u6211\u4f1a\u7559\u4e0b\u5173\u4e8e\u4e00\u76f4\u53d1\u751f\u7684\u9519\u8bef\u7684\u5907\u5fd8\u5f55\u3002<\/p>\n<h2>\u9519\u8bef\u7684\u60c5\u51b5<\/h2>\n<p>\u5728\u91cd\u65b0\u542f\u52a8\u8282\u70b9\u4e4b\u524d\uff0c\u5b58\u5728\u7740\u7ec8\u6b62\u7684Pod\u548c\u6302\u8d77\u7684Pod\u4e24\u8005\u540c\u65f6\u5b58\u5728\u7684\u5947\u602a\u60c5\u51b5\uff0c\u4f46\u662f\u5728\u91cd\u65b0\u542f\u52a8\u540e\uff0c\u6211\u6ce8\u610f\u5230Rook\/Ceph\u51fa\u73b0\u4e86\u5947\u602a\u7684\u7ed3\u679c\u3002<\/p>\n<pre class=\"post-pre\"><code>$ kubectl -n rook-ceph exec -it ... -- ceph status\r\n  cluster:\r\n    id:     3cedbf9c-82a4-4ce8-988b-9e64b1b3d97a\r\n    health: HEALTH_ERR\r\n            Module 'rook' has failed: HTTPSConnectionPool(host='10.233.0.1', port=443): Max retries exceeded with url: \/api\/v1\/namespaces\/rook-ceph\/pods (Caused by ProtocolError('Connection aborted.', error(99, 'Cannot assign requested address')))\r\n\r\n  services:\r\n    mon: 3 daemons, quorum bm,bs,bu (age 104s)\r\n    mgr: a(active, since 5m)\r\n    mds: myfs:1 {0=myfs-b=up:active} 1 up:standby-replay\r\n    osd: 4 osds: 4 up (since 112m), 4 in (since 8w)\r\n\r\n  data:\r\n    pools:   3 pools, 300 pgs\r\n    objects: 2.01M objects, 11 GiB\r\n    usage:   970 GiB used, 13 TiB \/ 14 TiB avail\r\n    pgs:     300 active+clean\r\n\r\n  io:\r\n    client:   1.2 KiB\/s rd, 46 KiB\/s wr, 2 op\/s rd, 2 op\/s wr\r\n<\/code><\/pre>\n<p>\u6211\u7684\u610f\u601d\u662f\uff0c\u6211\u6ce8\u610f\u5230\u5c3d\u7ba1MDS\/OSD\u6b63\u5e38\u8fd0\u884c\uff0c\u4f46\u51fa\u73b0\u4e86HEALTH_ERR\u7684\u95ee\u9898\u3002<br \/>\n\u7ecf\u8fc7\u5c1d\u8bd5\u548c\u7ea0\u9519\u540e\uff0c\u6211\u91cd\u65b0\u542f\u52a8\u4e86mgr pod\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete replicaset.apps\/rook-ceph-mgr-a-7644684465\r\nreplicaset.apps <span class=\"s2\">\"rook-ceph-mgr-a-7644684465\"<\/span> deleted\r\n<\/code><\/pre>\n<p>\u8fc7\u4e86\u4e00\u6bb5\u65f6\u95f4\u540e\uff0c\u6211\u68c0\u67e5\u4e86ceph\u7684\u72b6\u6001\uff0c\u5e76\u786e\u8ba4\u95ee\u9898\u5df2\u88ab\u4fee\u6b63\u3002<\/p>\n<pre class=\"post-pre\"><code>$ kubectl -n rook-ceph exec -it ... -- ceph status\r\n  cluster:\r\n    id:     3cedbf9c-82a4-4ce8-988b-9e64b1b3d97a\r\n    health: HEALTH_OK\r\n...\r\n<\/code><\/pre>\n<p>\u5b9e\u9645\u4e0a\uff0c\u5728mds pod\u7684\u5185\u90e8\u53d1\u751f\u4e86\u4e00\u4e9b\u9519\u8bef\u548c\u5176\u4ed6\u95ee\u9898\uff0c\u56e0\u6b64\u6211\u4eec\u4ecestandby\u4fa7\u91cd\u65b0\u542f\u52a8\u4e86\u5b83\uff0c\u7136\u540e\u5728\u786e\u8ba4\u4e86\u6240\u6709\u6b63\u5728\u8fd0\u884c\u7684Pod\u7684\u65e5\u5fd7\u540e\u8fdb\u884c\u4e86\u64cd\u4f5c\u3002<\/p>\n<p>\u7531\u4e8e\u5728\u672c\u5730\u7684\u95ee\u9898\u4e0a\u5e76\u6ca1\u6709\u63d0\u4f9b\u89e3\u51b3\u65b9\u6848\uff0c\u6240\u4ee5\u6211\u65e0\u6cd5\u786e\u5b9a\u662f\u5e94\u8be5\u7981\u7528\/\u542f\u7528\u6a21\u5757\uff0c\u8fd8\u662f\u5e94\u8be5\u91c7\u7528\u5176\u4ed6\u65b9\u6cd5\u6765\u5e94\u5bf9\u3002\u4e0d\u8fc7\uff0c\u6682\u65f6\u95ee\u9898\u5df2\u7ecf\u5173\u95ed\u4e86\uff0c\u8fd9\u662f\u4ef6\u597d\u4e8b\u3002<\/p>\n<h1>\u9047\u5230\u5168\u53f0shutdown\u5f15\u8d77\u7684race condition\u65f6\u7684\u5907\u5fd8\u5f55\u3002<\/h1>\n<blockquote><p>\u8fd9\u4e2a\u5907\u5fd8\u5f55\u5e76\u4e0d\u662f\u5173\u4e8e\u6062\u590d\u8bb0\u5f55\u7684\u3002\u5b83\u53ea\u662f\u8bb0\u5f55\u4e86\u5bf9\u6d4b\u8bd5\u95ee\u9898\u6240\u8fdb\u884c\u7684\u64cd\u4f5c\u548c\u7ed3\u679c\u3002\u5982\u679c\u9700\u8981\u4fdd\u7559\u6570\u636e\uff0c\u8bf7\u53e6\u5916\u8003\u8651\u65b9\u6cd5\u3002<\/p><\/blockquote>\n<blockquote><p>\u30102021\/08\/18\u66f4\u65b0\u3011<br \/>\n\u6b64\u73b0\u8c61\u7684\u6839\u672c\u539f\u56e0\u662f\/dev\/sdb2\u3001\/dev\/sdc2\u7b49\u53d8\u4e3abluestore\u3002\u5b98\u65b9\u6587\u6863\u5df2\u66f4\u65b0\uff0c\u8868\u793a\u8be5\u73b0\u8c61\u4f1a\u5728v1.6.0\u53ca\u66f4\u9ad8\u7248\u672c\u4e2d\u53d1\u751f\uff0c\u5e76\u63a8\u8350\u5347\u7ea7\u81f3v1.6.8\u53ca\u66f4\u9ad8\u7248\u672c\u3002\u89e3\u51b3\u65b9\u6848\u662f\u5728\u5347\u7ea7\u81f3v1.6.8\u53ca\u66f4\u9ad8\u7248\u672c\u540e\uff0c\u5220\u9664OSD\uff0c\u5bf9\u95ee\u9898\u78c1\u76d8\u8fdb\u884c\u521d\u59cb\u5316\u5e76\u91cd\u65b0\u8bc6\u522b\u4ee5\u5b9e\u73b0\u6b63\u5e38\u5316\u3002<br \/>\nhttps:\/\/rook.github.io\/docs\/rook\/v1.6\/ceph-common-issues.html#unexpected-partitions-created<\/p><\/blockquote>\n<p>2021\u5e7407\u6708\uff0c\u5728\u5173\u95ed\u6574\u4e2a\u6d4b\u8bd5\u578bk8s\u7cfb\u7edf\u8fdb\u884c\u7ef4\u62a4\u540e\u91cd\u65b0\u542f\u52a8\u7cfb\u7edf\u540e\uff0c\u51fa\u73b0\u4e86\u4ee5\u4e0b\u60c5\u51b5\u3002<\/p>\n<p>\u867d\u7136\u6709\u70b9\u5197\u957f\uff0c\u4f46\u6211\u8fd8\u662f\u5c06\u7ef4\u62a4\u7684BASH\u51fd\u6570\u4e00\u5e76\u5217\u51fa\u3002<\/p>\n<p>\u7ecf\u5e38\u6709\u62a5\u544a\u79f0\u53d1\u751f\u4e86\u7ade\u4e89\u6761\u4ef6\uff0c\u5e76\u4e14\u5728\u95ee\u9898\u4e2d\u6709\u8bb0\u5f55\u3002\u8fd8\u6709\u4e00\u4e9b\u5907\u5fd8\u5f55\u63d0\u5230\uff0c\u5982\u679c\u4f7f\u7528cri-o\u5c31\u4e0d\u4f1a\u9047\u5230\u8fd9\u4e2a\u95ee\u9898\uff0c\u4f46\u662f\u5728VMware\u4e0a\u8fdb\u884c\u7684\u6d4b\u8bd5\u8303\u56f4\u5185\uff0c\u5e76\u6ca1\u6709\u89c2\u5bdf\u5230libceph\u7684\u9519\u8bef\u53d1\u751f\u3002<\/p>\n<ul class=\"post-ul\">\n<li style=\"list-style-type: none;\">\n<ul class=\"post-ul\">https:\/\/github.com\/rook\/rook\/issues\/7734<\/ul>\n<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<ul class=\"post-ul\">https:\/\/github.com\/rook\/rook\/issues\/2517<\/ul>\n<p>\u8fd9\u6b21\u53d1\u751f\u7684\u95ee\u9898\u662f\u7531\u4e8e\u7b2c\u4e00\u4e2a\u94fe\u63a5\u4e2d\u7684\u539f\u59cb\u8bbe\u5907\u51fa\u73b0\u4e86\u6545\u969c\u3002<br \/>\n\u6574\u4e2a\u96c6\u7fa4\u4e2d\uff0c\u6bcf\u4e2a\u8282\u70b9\u4f7f\u7528\u4e24\u4e2a4TB\u7684\u786c\u76d8\uff0c\u5171\u8ba132TB\u7684\u5b58\u50a8\u5bb9\u91cf\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>lsblk <span class=\"nt\">-l<\/span>\r\nNAME  MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT\r\nloop0   7:0    0  55.4M  1 loop \/snap\/core18\/1944\r\nloop1   7:1    0  55.5M  1 loop \/snap\/core18\/2074\r\nloop2   7:2    0  67.6M  1 loop \/snap\/lxd\/20326\r\nloop3   7:3    0  32.3M  1 loop \/snap\/snapd\/12159\r\nloop5   7:5    0  32.3M  1 loop \/snap\/snapd\/12398\r\nloop6   7:6    0  70.3M  1 loop \/snap\/lxd\/21029\r\nsda     8:0    0 465.8G  0 disk \r\nsda1    8:1    0   512M  0 part \r\nsda2    8:2    0 465.3G  0 part \/\r\nsdb     8:16   0   3.7T  0 disk \r\nsdb2    8:18   0    48G  0 part \r\nsdb3    8:19   0   6.7M  0 part \r\nsdc     8:32   0   3.7T  0 disk \r\nsdc2    8:34   0    48G  0 part \r\nsdc3    8:35   0   6.2M  0 part \r\nsr0    11:0    1  1024M  0 rom  \r\n\r\n<span class=\"nv\">$ <\/span>lsblk <span class=\"nt\">-f<\/span>\r\nNAME   FSTYPE         LABEL UUID                                 FSAVAIL FSUSE% MOUNTPOINT\r\nloop0  squashfs                                                        0   100% \/snap\/core18\/1944\r\nloop1  squashfs                                                        0   100% \/snap\/core18\/2074\r\nloop2  squashfs                                                        0   100% \/snap\/lxd\/20326\r\nloop3  squashfs                                                        0   100% \/snap\/snapd\/12159\r\nloop5  squashfs                                                        0   100% \/snap\/snapd\/12398\r\nloop6  squashfs                                                        0   100% \/snap\/lxd\/21029\r\nsda                                                                             \r\n\u251c\u2500sda1 vfat                 5AD4-66D1                                           \r\n\u2514\u2500sda2 ext4                 42db9b49-9bb0-4590-a0f5-846f1a74058b  420.2G     3% \/\r\nsdb    ceph_bluestore                                                           \r\n\u251c\u2500sdb2 ceph_bluestore                                                           \r\n\u2514\u2500sdb3                                                                          \r\nsdc    ceph_bluestore                                                           \r\n\u251c\u2500sdc2 ceph_bluestore                                                           \r\n\u2514\u2500sdc3                                                                          \r\nsr0                           \r\n<\/code><\/pre>\n<p>\u5e0c\u671b\u7684\u662f\u5728\u4f7f\u7528\/dev\/sdb\u4f5c\u4e3aceph_bluestore\u7684\u60c5\u51b5\u4e0b\uff0c\/dev\/sdb{2,3}\u4f5c\u4e3a\u5206\u533a\u5e94\u8be5\u662f\u610f\u5916\u7684\u5b58\u5728\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">type <\/span>get_toolbox_podname \r\nget_toolbox_podname is a <span class=\"k\">function\r\n<\/span>get_toolbox_podname <span class=\"o\">()<\/span> \r\n<span class=\"o\">{<\/span> \r\n    <span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph get pod <span class=\"nt\">-l<\/span> <span class=\"nv\">app<\/span><span class=\"o\">=<\/span>rook-ceph-tools <span class=\"nt\">-o<\/span> <span class=\"nv\">jsonpath<\/span><span class=\"o\">=<\/span><span class=\"s1\">'{.items[*].metadata.name}'<\/span>\r\n<span class=\"o\">}<\/span>\r\n<span class=\"nv\">$ <\/span><span class=\"nb\">type <\/span>ceph_status\r\nceph_status is a <span class=\"k\">function\r\n<\/span>ceph_status <span class=\"o\">()<\/span> \r\n<span class=\"o\">{<\/span> \r\n    <span class=\"nv\">name<\/span><span class=\"o\">=<\/span><span class=\"si\">$(<\/span>get_toolbox_podname<span class=\"si\">)<\/span><span class=\"p\">;<\/span>\r\n    <span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph <span class=\"nb\">exec<\/span> <span class=\"nt\">-it<\/span> <span class=\"s2\">\"<\/span><span class=\"k\">${<\/span><span class=\"nv\">name<\/span><span class=\"k\">}<\/span><span class=\"s2\">\"<\/span> <span class=\"nt\">--<\/span> ceph status\r\n<span class=\"o\">}<\/span>\r\n\r\n<span class=\"c\">## \u3053\u3053\u304b\u3089\u304c\u30ed\u30b0<\/span>\r\n<span class=\"nv\">$ <\/span>ceph_status \r\n  cluster:\r\n    <span class=\"nb\">id<\/span>:     454d7e30-a1f5-4369-b1e5-6b9b2700054c\r\n    health: HEALTH_ERR\r\n            4\/507 objects unfound <span class=\"o\">(<\/span>0.789%<span class=\"o\">)<\/span>\r\n            1487 scrub errors\r\n            Too many repaired reads on 5 OSDs\r\n            Possible data damage: 4 pgs recovery_unfound, 26 pgs inconsistent\r\n            Degraded data redundancy: 12\/1521 objects degraded <span class=\"o\">(<\/span>0.789%<span class=\"o\">)<\/span>, 4 pgs degraded\r\n            3 pgs not deep-scrubbed <span class=\"k\">in <\/span><span class=\"nb\">time\r\n            <\/span>4 pgs not scrubbed <span class=\"k\">in <\/span><span class=\"nb\">time\r\n            <\/span>6 slow ops, oldest one blocked <span class=\"k\">for <\/span>1584994 sec, daemons <span class=\"o\">[<\/span>osd.0,osd.1,osd.2,osd.4,osd.5] have slow \r\nops.\r\n \r\n  services:\r\n    mon: 3 daemons, quorum a,b,c <span class=\"o\">(<\/span>age 2w<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 2w<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-b<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 16 osds: 16 up <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>, 16 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>\r\n    rgw: 1 daemon active <span class=\"o\">(<\/span>my.store.a<span class=\"o\">)<\/span>\r\n \r\n  task status:\r\n \r\n  data:\r\n    pools:   11 pools, 177 pgs\r\n    objects: 507 objects, 686 MiB\r\n    usage:   19 GiB used, 29 TiB \/ 29 TiB avail\r\n    pgs:     12\/1521 objects degraded <span class=\"o\">(<\/span>0.789%<span class=\"o\">)<\/span>\r\n             4\/507 objects unfound <span class=\"o\">(<\/span>0.789%<span class=\"o\">)<\/span>\r\n             148 active+clean\r\n             13  active+clean+inconsistent\r\n             12  active+clean+inconsistent+failed_repair\r\n             3   active+recovery_unfound+degraded\r\n             1   active+recovery_unfound+degraded+inconsistent+failed_repair\r\n \r\n  io:\r\n    client:   1.2 KiB\/s rd, 2 op\/s rd, 0 op\/s wr\r\n<\/code><\/pre>\n<p>\u8fd9\u4e2a\u65e5\u5fd7\u662f\u5728\u6545\u969c\u53d1\u751f\u540e\u51e0\u5929\u7684\u72b6\u6001\u4e0b\u8f93\u51fa\u7684\u3002<br \/>\n\u4e3a\u4e86\u6e05\u7406\u5de5\u4f5c\uff0c\u5f53\u5220\u9664\u6240\u6709PVC\u540e\uff0c\u5373\u4f7f\u8fc7\u4e86\u5927\u7ea68\u4e2a\u5c0f\u65f6\uff0c\u5220\u9664\u5904\u7406\u4ecd\u672a\u5b8c\u6210\uff0c\u8fd9\u4e2a\u65e5\u5fd7\u4f1a\u53d8\u6210\u4ee5\u4e0b\u8fd9\u6837\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span>ceph_status\r\n\r\n  cluster:\r\n    <span class=\"nb\">id<\/span>:     454d7e30-a1f5-4369-b1e5-6b9b2700054c\r\n    health: HEALTH_ERR\r\n            6\/507 objects unfound <span class=\"o\">(<\/span>1.183%<span class=\"o\">)<\/span>\r\n            1213 scrub errors\r\n            Too many repaired reads on 5 OSDs\r\n            Possible data damage: 6 pgs recovery_unfound, 20 pgs inconsistent\r\n            Degraded data redundancy: 18\/1521 objects degraded <span class=\"o\">(<\/span>1.183%<span class=\"o\">)<\/span>, 6 pgs degraded\r\n            3 pgs not deep-scrubbed <span class=\"k\">in <\/span><span class=\"nb\">time\r\n            <\/span>4 pgs not scrubbed <span class=\"k\">in <\/span><span class=\"nb\">time\r\n            <\/span>8 slow ops, oldest one blocked <span class=\"k\">for <\/span>1614550 sec, daemons <span class=\"o\">[<\/span>osd.0,osd.1,osd.2,osd.4,osd.5] have slow \r\nops.\r\n \r\n  services:\r\n    mon: 3 daemons, quorum a,b,c <span class=\"o\">(<\/span>age 2w<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 2w<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-b<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 16 osds: 16 up <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>, 16 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>\r\n    rgw: 1 daemon active <span class=\"o\">(<\/span>my.store.a<span class=\"o\">)<\/span>\r\n \r\n  task status:\r\n \r\n  data:\r\n    pools:   11 pools, 177 pgs\r\n    objects: 507 objects, 686 MiB\r\n    usage:   19 GiB used, 29 TiB \/ 29 TiB avail\r\n    pgs:     18\/1521 objects degraded <span class=\"o\">(<\/span>1.183%<span class=\"o\">)<\/span>\r\n             6\/507 objects unfound <span class=\"o\">(<\/span>1.183%<span class=\"o\">)<\/span>\r\n             154 active+clean\r\n             10  active+clean+inconsistent+failed_repair\r\n             7   active+clean+inconsistent\r\n             3   active+recovery_unfound+degraded\r\n             3   active+recovery_unfound+degraded+inconsistent+failed_repair\r\n \r\n  io:\r\n    client:   1.2 KiB\/s rd, 2 op\/s rd, 0 op\/s wr\r\n<\/code><\/pre>\n<p>\u8fd9\u65f6\u7684 PVC \u72b6\u6001\u5982\u4e0b\uff0c\u53ea\u5269\u4e0b\u4e86\u5076\u7136\u5728\u6700\u540e\u88ab\u5220\u9664\u7684 Rook\/Cassandra \u4f7f\u7528\u7684 PVC\u3002\u5220\u9664\u6b65\u9aa4\u672c\u8eab\u9075\u5faa\u6587\u4ef6\u4e2d\u7684\u8bf4\u660e\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl get pvc <span class=\"nt\">--all-namespaces<\/span>\r\nNAMESPACE        NAME                                                        STATUS        VOLUME             \r\n                        CAPACITY   ACCESS MODES   STORAGECLASS      AGE\r\nrook-cassandra   rook-cassandra-data-rook-cassandra-us-east-1-us-east-1a-0   Terminating   pvc-77d3d697-edd5-4\r\n633-8b81-a8474eb94043   5Gi        RWO            rook-ceph-block   26d\r\nrook-cassandra   rook-cassandra-data-rook-cassandra-us-east-1-us-east-1a-1   Terminating   pvc-c6bc0341-ba1d-4\r\n419-b280-c143edf240bb   5Gi        RWO            rook-ceph-block   26d\r\nrook-cassandra   rook-cassandra-data-rook-cassandra-us-east-1-us-east-1a-2   Terminating   pvc-c4261458-7d46-4\r\n33a-a05f-7c94e93b7034   5Gi        RWO            rook-ceph-block   26d\r\n<\/code><\/pre>\n<p>\u76ee\u524d\u6765\u8bf4\uff0cCeph\u5065\u5eb7\u8be6\u60c5\u7684\u8f93\u51fa\u60c5\u51b5\u5c31\u50cf\u662f\u4e00\u4e2a\u9519\u8bef\u7684\u8fde\u73af\u9a6c\u620f\u8868\u6f14\u3002<\/p>\n<p>\u7531\u4e8e\u5220\u9664\u4e86\u6240\u6709\u7684PV\uff0c\u6240\u4ee5\u6211\u4f1a\u5c1d\u8bd5\u4e00\u4e9b\u4e0d\u540c\u7684\u65b9\u6cd5\u3002<\/p>\n<pre class=\"post-pre\"><code># ceph health detail\r\nHEALTH_ERR 1\/397 objects unfound (0.252%); 920 scrub errors; Too many repaired reads on 5 OSDs; Possible data \r\ndamage: 1 pg recovery_unfound, 19 pgs inconsistent; Degraded data redundancy: 3\/1191 objects degraded (0.252%)\r\n, 1 pg degraded; 1 pgs not deep-scrubbed in time; 1 pgs not scrubbed in time; 4 slow ops, oldest one blocked f\r\nor 1626835 sec, daemons [osd.1,osd.2,osd.5] have slow ops.\r\n[WRN] OBJECT_UNFOUND: 1\/397 objects unfound (0.252%)\r\n    pg 2.1a has 1 unfound objects\r\n[ERR] OSD_SCRUB_ERRORS: 920 scrub errors\r\n[WRN] OSD_TOO_MANY_REPAIRS: Too many repaired reads on 5 OSDs\r\n    osd.3 had 16 reads repaired\r\n    osd.1 had 18 reads repaired\r\n    osd.7 had 31 reads repaired\r\n    osd.2 had 32 reads repaired\r\n    osd.4 had 34 reads repaired\r\n[ERR] PG_DAMAGED: Possible data damage: 1 pg recovery_unfound, 19 pgs inconsistent\r\n    pg 2.1 is active+clean+inconsistent+failed_repair, acting [7,0,6]\r\n    pg 2.3 is active+clean+inconsistent+failed_repair, acting [3,5,6]\r\n    pg 2.5 is active+clean+inconsistent, acting [6,4,1]\r\n    pg 2.7 is active+clean+inconsistent+failed_repair, acting [1,7,4]\r\n    pg 2.8 is active+clean+inconsistent+failed_repair, acting [3,2,0]\r\n    pg 2.a is active+clean+inconsistent+failed_repair, acting [2,4,7]\r\n    pg 2.11 is active+clean+inconsistent+failed_repair, acting [7,1,4]\r\n    pg 2.18 is active+clean+inconsistent+failed_repair, acting [3,1,2]\r\n    pg 2.1a is active+recovery_unfound+degraded, acting [5,6,4], 1 unfound\r\n    pg 2.1b is active+clean+inconsistent+failed_repair, acting [6,5,3]\r\n    pg 2.1c is active+clean+inconsistent+failed_repair, acting [2,1,0]\r\n    pg 2.1d is active+clean+inconsistent+failed_repair, acting [3,6,5]\r\n    pg 2.1e is active+clean+inconsistent+failed_repair, acting [4,6,3]\r\n    pg 2.1f is active+clean+inconsistent+failed_repair, acting [4,1,2]\r\n    pg 11.6 is active+clean+inconsistent, acting [3,0,1]\r\n    pg 11.9 is active+clean+inconsistent, acting [4,1,7]\r\n    pg 11.c is active+clean+inconsistent, acting [7,4,1]\r\n    pg 11.14 is active+clean+inconsistent, acting [1,4,3]\r\n    pg 11.1b is active+clean+inconsistent, acting [1,2,0]\r\n    pg 11.1d is active+clean+inconsistent, acting [1,11,3]\r\n[WRN] PG_DEGRADED: Degraded data redundancy: 3\/1191 objects degraded (0.252%), 1 pg degraded\r\n    pg 2.1a is active+recovery_unfound+degraded, acting [5,6,4], 1 unfound\r\n[WRN] PG_NOT_DEEP_SCRUBBED: 1 pgs not deep-scrubbed in time\r\n    pg 2.1a not deep-scrubbed since 2021-06-24T13:46:01.226873+0000\r\n[WRN] PG_NOT_SCRUBBED: 1 pgs not scrubbed in time\r\n    pg 2.1a not scrubbed since 2021-06-30T19:52:31.104830+0000\r\n[WRN] SLOW_OPS: 4 slow ops, oldest one blocked for 1626835 sec, daemons [osd.1,osd.2,osd.5] have slow ops.\r\n<\/code><\/pre>\n<p>\u5bf9\u4e8e\u627e\u4e0d\u5230\u7684pg\uff0c\u6211\u4eec\u5c06\u901a\u8fc7mark_unfound_lost\u6765\u5904\u7406\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"c\"># ceph pg 2.1a mark_unfound_lost delete<\/span>\r\npg has 1 objects unfound and apparently lost marking\r\n<\/code><\/pre>\n<p>\u5728RedHat\u7684\u6587\u6863\u4e2d\uff0c\u5bf9\u4e8eOSD_TOO_MANY_REPAIRS\u7684\u60c5\u51b5\uff0c\u4ecb\u7ecd\u4e86\u4e00\u4e2a\u540d\u4e3aclear_shards_repaired\u7684\u9009\u9879\uff0c\u4f3c\u4e4e\u53ef\u4ee5\u5f3a\u5236\u6e05\u9664\u8be5\u6d88\u606f\u3002\u7136\u800c\uff0c\u5728\u4e00\u822c\u7684ceph\u4e2d\u4f3c\u4e4e\u6ca1\u6709\u63d0\u4f9b\u8fd9\u4e2a\u9009\u9879\u3002<\/p>\n<p>\u6211\u4f1a\u5c1d\u8bd5\u4fee\u590d OSD\uff08\u5bf9\u8c61\u5b58\u50a8\u8bbe\u5907\uff09\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"c\"># ceph osd repair all<\/span>\r\ninstructed osd<span class=\"o\">(<\/span>s<span class=\"o\">)<\/span> 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 to repair\r\n<\/code><\/pre>\n<p>\u901a\u8fc7\u8fd9\u6837\u7684\u52aa\u529b\uff0c\u9519\u8bef\u6570\u91cf\u6709\u6240\u51cf\u5c11\uff0c\u53d8\u6210\u4e86\u4ee5\u4e0b\u7684\u72b6\u6001\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"c\"># ceph health detail<\/span>\r\nHEALTH_ERR 864 scrub errors<span class=\"p\">;<\/span> Too many repaired reads on 5 OSDs<span class=\"p\">;<\/span> Possible data damage: 6 pgs inconsistent<span class=\"p\">;<\/span> 3 sl\r\now ops, oldest one blocked <span class=\"k\">for <\/span>1627269 sec, daemons <span class=\"o\">[<\/span>osd.1,osd.2] have slow ops.\r\n<span class=\"o\">[<\/span>ERR] OSD_SCRUB_ERRORS: 864 scrub errors\r\n<span class=\"o\">[<\/span>WRN] OSD_TOO_MANY_REPAIRS: Too many repaired reads on 5 OSDs\r\n    osd.3 had 16 reads repaired\r\n    osd.1 had 18 reads repaired\r\n    osd.7 had 31 reads repaired\r\n    osd.2 had 32 reads repaired\r\n    osd.4 had 35 reads repaired\r\n<span class=\"o\">[<\/span>ERR] PG_DAMAGED: Possible data damage: 6 pgs inconsistent\r\n    pg 11.6 is active+clean+inconsistent, acting <span class=\"o\">[<\/span>3,0,1]\r\n    pg 11.9 is active+clean+inconsistent, acting <span class=\"o\">[<\/span>4,1,7]\r\n    pg 11.c is active+clean+inconsistent, acting <span class=\"o\">[<\/span>7,4,1]\r\n    pg 11.14 is active+clean+inconsistent, acting <span class=\"o\">[<\/span>1,4,3]\r\n    pg 11.1b is active+clean+inconsistent, acting <span class=\"o\">[<\/span>1,2,0]\r\n    pg 11.1d is active+clean+inconsistent, acting <span class=\"o\">[<\/span>1,11,3]\r\n<span class=\"o\">[<\/span>WRN] SLOW_OPS: 3 slow ops, oldest one blocked <span class=\"k\">for <\/span>1627269 sec, daemons <span class=\"o\">[<\/span>osd.1,osd.2] have slow ops.\r\n<\/code><\/pre>\n<p>\u8fc7\u4e86\u4e00\u6bb5\u65f6\u95f4\uff0cPG_DAMAGED\u5168\u90e8\u6d88\u5931\u4e86\uff0c\u53ea\u5269\u4e0b\u4e86OSD_TOO_MANY_REPAIRS\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"c\"># ceph health detail<\/span>\r\nHEALTH_WARN Too many repaired reads on 5 OSDs<span class=\"p\">;<\/span> 3 slow ops, oldest one blocked <span class=\"k\">for <\/span>1627335 sec, daemons <span class=\"o\">[<\/span>osd.1,\r\nosd.2] have slow ops.\r\n<span class=\"o\">[<\/span>WRN] OSD_TOO_MANY_REPAIRS: Too many repaired reads on 5 OSDs\r\n    osd.3 had 16 reads repaired\r\n    osd.1 had 18 reads repaired\r\n    osd.7 had 31 reads repaired\r\n    osd.2 had 32 reads repaired\r\n    osd.4 had 35 reads repaired\r\n<span class=\"o\">[<\/span>WRN] SLOW_OPS: 3 slow ops, oldest one blocked <span class=\"k\">for <\/span>1627335 sec, daemons <span class=\"o\">[<\/span>osd.1,osd.2] have slow ops.\r\n\r\n\r\n<span class=\"c\"># ceph status<\/span>\r\n  cluster:\r\n    <span class=\"nb\">id<\/span>:     454d7e30-a1f5-4369-b1e5-6b9b2700054c\r\n    health: HEALTH_WARN\r\n            Too many repaired reads on 5 OSDs\r\n            3 slow ops, oldest one blocked <span class=\"k\">for <\/span>1627340 sec, daemons <span class=\"o\">[<\/span>osd.1,osd.2] have slow ops.\r\n \r\n  services:\r\n    mon: 3 daemons, quorum a,b,c <span class=\"o\">(<\/span>age 2w<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 2w<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-b<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 16 osds: 16 up <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>, 16 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>\r\n    rgw: 1 daemon active <span class=\"o\">(<\/span>my.store.a<span class=\"o\">)<\/span>\r\n \r\n  task status:\r\n \r\n  data:\r\n    pools:   11 pools, 177 pgs\r\n    objects: 343 objects, 184 MiB\r\n    usage:   18 GiB used, 29 TiB \/ 29 TiB avail\r\n    pgs:     177 active+clean\r\n \r\n  io:\r\n    client:   5.7 KiB\/s rd, 426 B\/s wr, 6 op\/s rd, 2 op\/s wr\r\n<\/code><\/pre>\n<p>\u6211\u5c06\u5c1d\u8bd5\u4f7f\u7528\u547d\u4ee4 &#8220;osd purge &lt;id|osd.id&gt; [&#8211;force] [&#8211;yes-i-really-mean-it]&#8221; \u8fdb\u884c\u6e05\u9664\u64cd\u4f5c\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"c\"># ceph osd down osd.1<\/span>\r\nmarked down osd.1. \r\n<span class=\"c\"># ceph osd down osd.2<\/span>\r\nmarked down osd.2. \r\n\r\n<span class=\"c\"># ceph status<\/span>\r\n  cluster:\r\n    <span class=\"nb\">id<\/span>:     454d7e30-a1f5-4369-b1e5-6b9b2700054c\r\n    health: HEALTH_WARN\r\n            Too many repaired reads on 4 OSDs\r\n            2 slow ops, oldest one blocked <span class=\"k\">for <\/span>1627677 sec, osd.2 has slow ops\r\n \r\n  services:\r\n    mon: 3 daemons, quorum a,b,c <span class=\"o\">(<\/span>age 2w<span class=\"o\">)<\/span>\r\n    mgr: a<span class=\"o\">(<\/span>active, since 2w<span class=\"o\">)<\/span>\r\n    mds: myfs:1 <span class=\"o\">{<\/span><span class=\"nv\">0<\/span><span class=\"o\">=<\/span>myfs-b<span class=\"o\">=<\/span>up:active<span class=\"o\">}<\/span> 1 up:standby-replay\r\n    osd: 16 osds: 16 up <span class=\"o\">(<\/span>since 76s<span class=\"o\">)<\/span>, 16 <span class=\"k\">in<\/span> <span class=\"o\">(<\/span>since 2w<span class=\"o\">)<\/span>\r\n    rgw: 1 daemon active <span class=\"o\">(<\/span>my.store.a<span class=\"o\">)<\/span>\r\n \r\n  task status:\r\n \r\n  data:\r\n    pools:   11 pools, 177 pgs\r\n    objects: 343 objects, 184 MiB\r\n    usage:   18 GiB used, 29 TiB \/ 29 TiB avail\r\n    pgs:     177 active+clean\r\n \r\n  io:\r\n    client:   3.5 KiB\/s rd, 511 B\/s wr, 4 op\/s rd, 3 op\/s wr\r\n<\/code><\/pre>\n<h1>\u5982\u679c\u5b89\u88c5\u4e86Rook\/Ceph\uff0c\u5c06\u751f\u6210\u8bb8\u591aosd\u3002<\/h1>\n<p>\u5728\u65b0\u6784\u5efa\u7684k8s\u96c6\u7fa4\u4e2d\u5f15\u5165Rook\/Ceph\u65f6\uff0c\u539f\u672c\u5e94\u8be5\u53ea\u6709\u6bcf\u53f0\u670d\u52a1\u5668\u4e00\u4e2aBlueStore\uff08HDD\uff09\uff0c\u4f46\u5374\u521b\u5efa\u4e86\u516d\u4e2aOSD\u3002<\/p>\n<p>USB\u9a71\u52a8\u5668\u63d2\u5165\u7684AutoInstall\u5728\u5f15\u5bfc\u64cd\u4f5c\u7cfb\u7edf\u65f6\u88ab\u4fdd\u7559\u3002\u6211\u66fe\u62c5\u5fc3ISO9660\u6620\u50cf\u4f1a\u88ab\u89c6\u4e3aCD-ROM\u800c\u88ab\u5ffd\u7565\uff0c\u4f46\u7cfb\u7edf\u5c06\u5176\u8bc6\u522b\u4e3a\/dev\/sdc\uff0c\u5e76\u6210\u529f\u542f\u52a8\u4e86OSD\u6a21\u5757\u3002<\/p>\n<p>\u5f53\u7136\uff0c\u7531\u4e8eUSB\u5b58\u50a8\u5668\u88ab\u56de\u6536\uff0c\u56e0\u6b64\/dev\/sdc\u672c\u8eab\u4e0d\u5b58\u5728\uff0c\u5bfc\u81f4OSD\u65e0\u6cd5\u542f\u52a8\u5e76\u51fa\u73b0\u9519\u8bef\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph get pod <span class=\"nt\">-l<\/span> osd\r\nNAME                               READY   STATUS                  RESTARTS          AGE\r\nrook-ceph-osd-0-5557b5c888-lj9x8   1\/1     Running                 2 <span class=\"o\">(<\/span>16h ago<span class=\"o\">)<\/span>       6d15h\r\nrook-ceph-osd-1-66bcb89669-ch5bx   1\/1     Running                 2 <span class=\"o\">(<\/span>16h ago<span class=\"o\">)<\/span>       6d15h\r\nrook-ceph-osd-2-9b8b5ff5b-44kbz    1\/1     Running                 2 <span class=\"o\">(<\/span>16h ago<span class=\"o\">)<\/span>       6d15h\r\nrook-ceph-osd-3-59bcb68f97-g9nl9   0\/1     Init:CrashLoopBackOff   177 <span class=\"o\">(<\/span>24s ago<span class=\"o\">)<\/span>     14h\r\nrook-ceph-osd-4-6745cf4d86-7vnht   0\/1     Init:CrashLoopBackOff   192 <span class=\"o\">(<\/span>3m4s ago<span class=\"o\">)<\/span>    6d15h\r\nrook-ceph-osd-5-f6cc674f4-6sp5q    1\/1     Running                 2 <span class=\"o\">(<\/span>16h ago<span class=\"o\">)<\/span>       6d15h\r\nrook-ceph-osd-6-f776bf965-xqbrg    0\/1     Init:CrashLoopBackOff   175 <span class=\"o\">(<\/span>3m20s ago<span class=\"o\">)<\/span>   14h\r\n<\/code><\/pre>\n<h2>\u73af\u5883<\/h2>\n<ul class=\"post-ul\">\n<li style=\"list-style-type: none;\">\n<ul class=\"post-ul\">Rook v1.9.10 (ceph 16.2.10)<\/ul>\n<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<ul class=\"post-ul\">Kubernetes v1.23.7<\/ul>\n<h2>\u5de5\u4f5c\u6b65\u9aa4<\/h2>\n<p>\u7531\u4e8e\u6211\u4eec\u77e5\u9053\u8fd9\u4e2aOSD\u662f\u4e0d\u5fc5\u8981\u7684\uff0c\u4e0d\u9700\u8981\u8fdb\u884c\u4efb\u4f55BlueStore\u521d\u59cb\u5316\uff0c\u4e5f\u6ca1\u6709\u6784\u5efaPGs\uff0c\u56e0\u6b64\u6211\u4eec\u5c06\u5728\u6ca1\u6709\u53d7\u5f71\u54cd\u7684PV\u7684\u524d\u63d0\u4e0b\u8fdb\u884c\u64cd\u4f5c\u3002<\/p>\n<p>\u6309\u7167\u5b98\u65b9\u6307\u5357\u300aCeph OSD\u7ba1\u7406\u300b\u548c\u5b98\u65b9\u6307\u5357\u300aCeph\u914d\u7f6e\u300b\uff0c\u7b80\u5355\u5730\u5220\u9664\u4e0d\u9700\u8981\u7684OSD\uff08ID: 3,4,6\uff09\u3002<\/p>\n<ol>\n<li style=\"list-style-type: none;\">\n<ol>\u64cd\u4f5c\u5458\u505c\u6b62<\/ol>\n<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<ol>\n<li style=\"list-style-type: none;\">\n<ol>\u4eceToolBox\u6267\u884cceph osd down<\/ol>\n<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<ol>\n<li style=\"list-style-type: none;\">\n<ol>\u4eceToolBox\u6267\u884cceph osd out\/crush remove\/auth del\/rm<\/ol>\n<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<ol>\n<li style=\"list-style-type: none;\">\n<ol>\u5220\u9664\u4e0eOSD\u5bf9\u5e94\u7684Deployment\u5bf9\u8c61\uff08\u5982\u679cremoveOSDsIfOutAndSafeToRemove\u6ca1\u6709\u8bbe\u7f6e\u4e3atrue\uff09<\/ol>\n<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<ol>\u64cd\u4f5c\u5458\u6062\u590d<\/ol>\n<h3>\u64cd\u4f5c\u5458\u505c\u6b62\u3002<\/h3>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph scale deployment rook-ceph-operator <span class=\"nt\">--replicas<\/span><span class=\"o\">=<\/span>0\r\n<\/code><\/pre>\n<h3>\u4eceToolBox\u4e2d\u6267\u884cceph osd down\u3002<\/h3>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph <span class=\"nb\">exec<\/span> <span class=\"nt\">-it<\/span> rook-ceph-tools-7d59b6df78-tlqrb <span class=\"nt\">--<\/span> bash\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd down osd.3\r\nosd.3 is already down. \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd down osd.4\r\nosd.4 is already down. \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd down osd.6\r\nosd.6 is already down\r\n<\/code><\/pre>\n<h3>\u6267\u884c ToolBox \u4e2d\u7684 ceph osd out\/crush remove\/auth del\/rm\u3002<\/h3>\n<pre class=\"post-pre\"><code><span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd out osd.6\r\nmarked out osd.6. \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd out osd.4\r\nosd.4 is already out. \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd out osd.3\r\nosd.3 is already out.\r\n\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd crush remove osd.3\r\nremoved item <span class=\"nb\">id <\/span>3 name <span class=\"s1\">'osd.3'<\/span> from crush map\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd crush remove osd.4\r\nremoved item <span class=\"nb\">id <\/span>4 name <span class=\"s1\">'osd.4'<\/span> from crush map\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd crush remove osd.6\r\nremoved item <span class=\"nb\">id <\/span>6 name <span class=\"s1\">'osd.6'<\/span> from crush map\r\n\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph auth del osd.3\r\nupdated                             \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph auth del osd.4 \r\n^Pupdated                           \r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph auth del osd.6 \r\nupdated\r\n\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd <span class=\"nb\">rm <\/span>osd.3\r\nremoved osd.3\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd <span class=\"nb\">rm <\/span>osd.4\r\nremoved osd.4\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span>ceph osd <span class=\"nb\">rm <\/span>osd.6\r\nremoved osd.6\r\n\r\n<span class=\"o\">[<\/span>rook@rook-ceph-tools-7d59b6df78-tlqrb \/]<span class=\"nv\">$ <\/span><span class=\"nb\">exit\r\nexit<\/span>\r\n<span class=\"nv\">$ <\/span>\r\n<\/code><\/pre>\n<h3>\u5220\u9664\u5bf9\u5e94\u7684Deployment\u5bf9\u8c61<\/h3>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete deploy rook-ceph-osd-3\r\ndeployment.apps <span class=\"s2\">\"rook-ceph-osd-3\"<\/span> deleted\r\n<span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete deploy rook-ceph-osd-4\r\ndeployment.apps <span class=\"s2\">\"rook-ceph-osd-4\"<\/span> deleted\r\n<span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph delete deploy rook-ceph-osd-6\r\ndeployment.apps <span class=\"s2\">\"rook-ceph-osd-6\"<\/span> deleted\r\n<\/code><\/pre>\n<h3>Operator\u7684\u56de\u5f52<\/h3>\n<pre class=\"post-pre\"><code><span class=\"nv\">$ <\/span><span class=\"nb\">sudo <\/span>kubectl <span class=\"nt\">-n<\/span> rook-ceph scale deployment rook-ceph-operator <span class=\"nt\">--replicas<\/span><span class=\"o\">=<\/span>1\r\ndeployment.apps\/rook-ceph-operator scaled\r\n<\/code><\/pre>\n<h2>\u786e\u8ba4\u7ed3\u679c<\/h2>\n<p>\u6211\u5c06\u8fdb\u5165ToolBox\uff0c\u7136\u540e\u8fdb\u884cceph\u72b6\u6001\u7684\u786e\u8ba4\u3002<\/p>\n<pre class=\"post-pre\"><code>  cluster:\r\n    id:     8c9c83ef-4694-42b9-8cc7-cbb4a529e384\r\n    health: HEALTH_OK\r\n \r\n  services:\r\n    mon: 3 daemons, quorum a,b,c (age 17h)\r\n    mgr: a(active, since 17h), standbys: b\r\n    osd: 4 osds: 4 up (since 17h), 4 in (since 41m); 2 remapped pgs\r\n    rgw: 1 daemon active (1 hosts, 1 zones)\r\n \r\n  data:\r\n    pools:   9 pools, 113 pgs\r\n    objects: 2.92k objects, 7.1 GiB\r\n    usage:   16 GiB used, 3.6 TiB \/ 3.6 TiB avail\r\n    pgs:     125\/8769 objects misplaced (1.425%)\r\n             111 active+clean\r\n             1   active+remapped+backfilling\r\n             1   active+remapped+backfill_wait\r\n \r\n  io:\r\n    client:   7.5 KiB\/s rd, 204 KiB\/s wr, 1 op\/s rd, 1 op\/s wr\r\n    recovery: 6.5 MiB\/s, 6 objects\/s\r\n<\/code><\/pre>\n<p>ceph osd tree \u7684\u8f93\u51fa\u5982\u4e0b\u6240\u793a\u3002<\/p>\n<pre class=\"post-pre\"><code>ID  CLASS  WEIGHT   TYPE NAME         STATUS  REWEIGHT  PRI-AFF\r\n-1         3.63879  root default                               \r\n-5         0.90970      host s10rx61                           \r\n 1    hdd  0.90970          osd.1         up   1.00000  1.00000\r\n-9         0.90970      host s10rx62                           \r\n 5    hdd  0.90970          osd.5         up   1.00000  1.00000\r\n-7         0.90970      host s10rx63                           \r\n 2    hdd  0.90970          osd.2         up   1.00000  1.00000\r\n-3         0.90970      host s10rx64                           \r\n 0    hdd  0.90970          osd.0         up   1.00000  1.00000\r\n<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"<p>\u9996\u5148 \u5728\u786e\u8ba4\u4e86\u4ece\u505c\u7535\u4e2d\u6062\u590d\u7684Kubernetes\u96c6\u7fa4\u72b6\u6001\u65f6\uff0c\u6211\u53d1\u73b0\u53ea\u6709\u4e00\u4e2aPod\uff08gitbucket\uff09\u4ecd\u5904\u4e8eC [&hellip;]<\/p>\n","protected":false},"author":8,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-50352","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.5 (Yoast SEO v21.5) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d - Blog - Silicon Cloud<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.silicloud.com\/zh\/blog\/\u4ecerook-ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\" \/>\n<meta property=\"og:description\" content=\"\u9996\u5148 \u5728\u786e\u8ba4\u4e86\u4ece\u505c\u7535\u4e2d\u6062\u590d\u7684Kubernetes\u96c6\u7fa4\u72b6\u6001\u65f6\uff0c\u6211\u53d1\u73b0\u53ea\u6709\u4e00\u4e2aPod\uff08gitbucket\uff09\u4ecd\u5904\u4e8eC [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.silicloud.com\/zh\/blog\/\u4ecerook-ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog - Silicon Cloud\" \/>\n<meta property=\"article:published_time\" content=\"2023-09-14T06:48:08+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-05-03T23:48:24+00:00\" \/>\n<meta name=\"author\" content=\"\u96c5, \u609f\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u96c5, \u609f\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"17 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/\",\"name\":\"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d - Blog - Silicon Cloud\",\"isPartOf\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\"},\"datePublished\":\"2023-09-14T06:48:08+00:00\",\"dateModified\":\"2024-05-03T23:48:24+00:00\",\"author\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/f044a4b7fa4ee2701702942002419ca6\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.silicloud.com\/zh\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/\",\"name\":\"Blog - Silicon Cloud\",\"description\":\"\",\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/f044a4b7fa4ee2701702942002419ca6\",\"name\":\"\u96c5, \u609f\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/e71a913e914f1aad1efc391f92084294bac54bc782acd289638580134cf667a6?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/e71a913e914f1aad1efc391f92084294bac54bc782acd289638580134cf667a6?s=96&d=mm&r=g\",\"caption\":\"\u96c5, \u609f\"},\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/author\/yawu\/\"},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#local-main-organization-logo\",\"url\":\"\",\"contentUrl\":\"\",\"caption\":\"Blog - Silicon Cloud\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d - Blog - Silicon Cloud","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.silicloud.com\/zh\/blog\/\u4ecerook-ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\/","og_locale":"zh_CN","og_type":"article","og_title":"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d","og_description":"\u9996\u5148 \u5728\u786e\u8ba4\u4e86\u4ece\u505c\u7535\u4e2d\u6062\u590d\u7684Kubernetes\u96c6\u7fa4\u72b6\u6001\u65f6\uff0c\u6211\u53d1\u73b0\u53ea\u6709\u4e00\u4e2aPod\uff08gitbucket\uff09\u4ecd\u5904\u4e8eC [&hellip;]","og_url":"https:\/\/www.silicloud.com\/zh\/blog\/\u4ecerook-ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d\/","og_site_name":"Blog - Silicon Cloud","article_published_time":"2023-09-14T06:48:08+00:00","article_modified_time":"2024-05-03T23:48:24+00:00","author":"\u96c5, \u609f","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"\u96c5, \u609f","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"17 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/","url":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/","name":"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d - Blog - Silicon Cloud","isPartOf":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website"},"datePublished":"2023-09-14T06:48:08+00:00","dateModified":"2024-05-03T23:48:24+00:00","author":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/f044a4b7fa4ee2701702942002419ca6"},"breadcrumb":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.silicloud.com\/zh\/blog\/"},{"@type":"ListItem","position":2,"name":"\u4eceRook\/Ceph\u53d1\u751f\u7684\u4e0d\u4e00\u81f4\u4e2d\u6062\u590d"}]},{"@type":"WebSite","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website","url":"https:\/\/www.silicloud.com\/zh\/blog\/","name":"Blog - Silicon Cloud","description":"","inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/f044a4b7fa4ee2701702942002419ca6","name":"\u96c5, \u609f","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/e71a913e914f1aad1efc391f92084294bac54bc782acd289638580134cf667a6?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/e71a913e914f1aad1efc391f92084294bac54bc782acd289638580134cf667a6?s=96&d=mm&r=g","caption":"\u96c5, \u609f"},"url":"https:\/\/www.silicloud.com\/zh\/blog\/author\/yawu\/"},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e4%bb%8erook-ceph%e5%8f%91%e7%94%9f%e7%9a%84%e4%b8%8d%e4%b8%80%e8%87%b4%e4%b8%ad%e6%81%a2%e5%a4%8d\/#local-main-organization-logo","url":"","contentUrl":"","caption":"Blog - Silicon Cloud"}]}},"_links":{"self":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/50352","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/comments?post=50352"}],"version-history":[{"count":2,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/50352\/revisions"}],"predecessor-version":[{"id":97050,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/50352\/revisions\/97050"}],"wp:attachment":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/media?parent=50352"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/categories?post=50352"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/tags?post=50352"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}