问题

在发生故障后将磁盘添加回集群。

  • 2018年11月12日
  • 3回复
  • 9的观点
O
徽章 ＋1
我很抱歉，如果这个问题已经在其他地方得到了回答，但我一直找不到它，所以它可能没有。我相信其他人也会让自己陷入这种境地。

我有一个基于旧超级微工具包的3节点Nutanix集群。每个节点配置3块硬盘:1块240GB ssd硬盘、1块250GB硬盘和1块1.0TB硬盘。我从Node 1中删除了250GB的HDD，以演示在提取磁盘时仍然可以正常工作。

那天剩下的时间里，我把它放在一边，发现它已经从集群中失败了。我认为这很正常。然后我把它推回去。我希望能做些什么让它再次被认可，但我不知道那是什么。

它被主机识别为/dev/sda:-

代码: 
# lsscsi root@NTNX-baff460a-A ~
[0:0:0:0] disk ATA WDC wd2502abyss -0 3B03 /dev/sda
[0:0:1:0] disk ATA WDC WD10EZEX-08W 1A02 /dev/sdb . exe /dev/sdb . exe
[1:0:0:0]硬盘ATA KINGSTON SA400S3 61E1 /dev/sdc
[4:0:0:0] disk Kingston DataTraveler 3.0 /dev/sdd


它也被CVM识别为/dev/sda:-

代码: 
nutanix@NTNX-baff460a-A-CVM: 192.168.2.31: ~ lsscsi美元
[1:0:0:0] cd/dvd QEMU QEMU dvd - rom 2.5+ /dev/sr0
[2:0:0:0] /dev/sda . /dev/sda . /dev/sda . /dev/sda . /dev/sda . /dev/sda . /dev/sda . /dev/sda
[2:0:0:1] disk ATA WDC WD10EZEX-08W 2.5+ /dev/sdc
[2:0:0:2] disk ATA KINGSTON SA400S3 2.5+ /dev/sdb . /


但是，它不会出现在df -f中。
代码: 
nutanix@NTNX-baff460a-A-CVM: 192.168.2.31: ~ $ df - h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 9.8G 5.2G 4.6G 54% /
devtmpfs 4.9G 0 4.9G 0% /dev
tmpfs 512M 4.0K 512M 1% /dev/shm
tmpfs 4.9G 452K 4.9G 1% /运行
/sys/fs/cgroup . tmpfs 4.9G 0 4.9G 0%
/dev/loop0 240M 2.5M 221M 2% /tmp
/dev/sdb3 40G 15G 25G 37% /home
/dev/sdc1 916G 139M 907G 1% /home/nutanix/data/stargate-storage/disks/50014ee265101d36
/dev/sdb4 161G 31G 128G 20% /home/nutanix/data/stargate-storage/disks/50026B77822DAFFD
tmpfs 1000M 0 1000M 0% /run/user/1000



我试过运行' add-all-free-disks '命令，但这只是显示存储池的现有大小和已经在其中的磁盘，所以我猜它没有发现250GB磁盘是一个可以添加到集群。
代码: 
Ncli > sp edit name=default-storage-pool-54685878474571 add-all-free-disks=true

Id: 000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 7
Uuid: e9c42d6c - c2ea - 4 - o - 9616 - 71327 - f70a89c
名称:默认-存储池- 54685878474571
磁盘数:8
硬盘编号:000575d0-1c84-5942-5000-003048f5d55c:: 33,000575d0 -1c84-5942-5000-00
3048 f5d55c:: 34岁000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 35岁,000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 36岁,
000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 37岁000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 41岁,000575 d0-1c84-59
42 - 5000 - 003048 - f5d55c: 42岁,000575 d0 - 1 - c84 - 5942 - 5000 - 003048 - f5d55c:: 43
容量(物理):2.96 TiB(3,249,836,931,478字节)
已使用(物理):30.59 GiB(32,849,920,000字节)
空闲(物理):2.93 TiB(3,216,97,011,1478字节)


如果我必须从集群中删除节点，重新映像它，然后再次将其添加进去，以便再次将磁盘添加进去，我将这样做，但我希望通过这个过程可以学到一些东西。通过探索这些选项，我已经学到了一些东西。

谢谢!
欧文。
    \r\n
    \r\nI have a 3 node Nutanix Cluster based on old SuperMicro kit. Each node has 3 disks: 1 x 240GB SSD1, 1 x 250GB HDD and 1 x 1.0TB HDD. I removed the 250GB HDD from Node 1 to demonstrate that things continue working when a disk is pulled.
    \r\n
    \r\nI left it out for the rest of the day and found that it had been failed out of the cluster. This is normal I believe. I then pushed it back in. I was expecting to have to do something to get it to be recognised again but I can\u2019t find out what that is.
    \r\n
    \r\nIt\u2019s recognised by the host as \/dev\/sda:-
    \r\n
    \r\n
    code:<\/b>
    [root@NTNX-baff460a-A ~]# lsscsi
    [0:0:0:0]  disk  ATA   WDC WD2502ABYS-0 3B03 \/dev\/sda 
    [0:0:1:0]  disk  ATA   WDC WD10EZEX-08W 1A02 \/dev\/sdb 
    [1:0:0:0]  disk  ATA   KINGSTON SA400S3 61E1 \/dev\/sdc 
    [4:0:0:0]  disk  Kingston DataTraveler 3.0    \/dev\/sdd 
    <\/pre><\/div>
    \r\n
    \r\nIt\u2019s also recognised by the CVM as \/dev\/sda:-
    \r\n
    \r\n
    code:<\/b>
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ lsscsi
    [1:0:0:0]  cd\/dvd QEMU   QEMU DVD-ROM   2.5+ \/dev\/sr0 
    [2:0:0:0]  disk  ATA   WDC WD2502ABYS-0 2.5+ \/dev\/sda 
    [2:0:0:1]  disk  ATA   WDC WD10EZEX-08W 2.5+ \/dev\/sdc 
    [2:0:0:2]  disk  ATA   KINGSTON SA400S3 2.5+ \/dev\/sdb 
    <\/pre><\/div>
    \r\n
    \r\nHowever, it doesn\u2019t show up with df -f.
    \r\n
    code:<\/b>
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ df -h
    Filesystem   Size Used Avail Use% Mounted on
    \/dev\/sdb1    9.8G 5.2G 4.6G 54% \/
    devtmpfs    4.9G   0 4.9G  0% \/dev
    tmpfs      512M 4.0K 512M  1% \/dev\/shm
    tmpfs      4.9G 452K 4.9G  1% \/run
    tmpfs      4.9G   0 4.9G  0% \/sys\/fs\/cgroup
    \/dev\/loop0   240M 2.5M 221M  2% \/tmp
    \/dev\/sdb3    40G  15G  25G 37% \/home
    \/dev\/sdc1    916G 139M 907G  1% \/home\/nutanix\/data\/stargate-storage\/disks\/50014ee265101d36
    \/dev\/sdb4    161G  31G 128G 20% \/home\/nutanix\/data\/stargate-storage\/disks\/50026B77822DAFFD
    tmpfs     1000M   0 1000M  0% \/run\/user\/1000
    <\/pre><\/div>
    \r\n
    \r\n
    \r\nI\u2019ve tried running the \u2018add-all-free-disks\u2019 command but that just shows the existing size of the storage pool and the disks that are already in it so I\u2019m guessing it is not finding that the 250GB disk is one that can be added to the Cluster.
    \r\n
    code:<\/b>
    ncli> sp edit name=default-storage-pool-54685878474571 add-all-free-disks=true

      Id            : 000575d0-1c84-5942-5000-003048f5d55c::7
      Uuid           : e9c42d6c-c2ea-4ead-9616-71327f70a89c
      Name           : default-storage-pool-54685878474571
      Disk Count        : 8
      Disk Ids         : 000575d0-1c84-5942-5000-003048f5d55c::33, 000575d0-1c84-5942-5000-00
    3048f5d55c::34, 000575d0-1c84-5942-5000-003048f5d55c::35, 000575d0-1c84-5942-5000-003048f5d55c::36, 
    000575d0-1c84-5942-5000-003048f5d55c::37, 000575d0-1c84-5942-5000-003048f5d55c::41, 000575d0-1c84-59
    42-5000-003048f5d55c::42, 000575d0-1c84-5942-5000-003048f5d55c::43
      Capacity (Physical)    : 2.96 TiB (3,249,836,931,478 bytes)
      Used (Physical)      : 30.59 GiB (32,849,920,000 bytes)
      Free (Physical)      : 2.93 TiB (3,216,987,011,478 bytes)
    <\/pre><\/div>
    \r\n
    \r\nIf I have to remove the node from the cluster, re-image it and add it in again to get the disk in again I\u2019ll do so but I\u2019m hoping to learn something by going through this process. I\u2019ve already learned a few things by exploring the options.
    \r\n
    \r\nThanks!
    \r\nOwen.","quoteUsername":"Owen Hopkins","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
    本主题已关闭供评论

    3回复

    J
    Userlevel 7
    徽章 + 25
    那个磁盘在删除列表里吗?
    O
    徽章 ＋1
    啊哈!是的,它是。我不知道这事-

    代码: 
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ ncli disk list-tombstone-entries

    序列号:50014ee2048288f1



    我运行命令，它不在列表中了:-

    代码: 
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ ncli disk rm-tombstone-entry serial-number=50014ee2048288f1
    磁盘'50014ee2048288f1'从墓碑条目中删除



    我再次运行了添加所有空闲磁盘命令，但它只是产生了相同的结果。磁盘个数为8个，容量相同。除了把他们从墓碑列表中删除，还有什么别的办法吗?
    \r\n
    \r\n
    code:<\/b>
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ ncli disk list-tombstone-entries

      Serial number       : 50014ee2048288f1
    <\/pre><\/div>
    \r\n
    \r\n
    \r\nI ran the command and it's not in the list anymore:-
    \r\n
    \r\n
    code:<\/b>
    nutanix@NTNX-baff460a-A-CVM:192.168.2.31:~$ ncli disk rm-tombstone-entry serial-number=50014ee2048288f1
    Disk '50014ee2048288f1' removed from tombstone entries
    <\/pre><\/div>
    \r\n
    \r\n
    \r\nI ran the add-all-free-disks command again but it just produced the same result. A count of 8 disks and the same capacity. Is there something else to do as well as removing them from the tombstone list?","quoteUsername":"Owen Hopkins","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
    J
    Userlevel 7
    徽章 + 25
    这个信息看起来有点过时，但可能一些reenable和clean命令将在当前的gen CVM上发挥作用。

    http://nutanix.blogspot.com/2013/05/troubleshooting-bad-hard-disk-on-nutanix.html
    \r\n
    \r\nhttp:\/\/nutanix.blogspot.com\/2013\/05\/troubleshooting-bad-hard-disk-on-nutanix.html","quoteUsername":"jrack","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">
    Baidu