Cluster-wide Device Special Files (cDSFs)

HP-UX, Linux, Tru64 UNIX, NonStop, OpenVMS, MS Windows Server, DataCenter and Virtualization Solutions, etc

Moderator: xyevolve

Forum rules
1. 本版是定位于惠普软件系统及解决方案的技术讨论区。
2. 本版鼓励发帖共同讨论技术问题,不鼓励站内信件私下交流,独知知不如众知知。
3. 本版允许转贴或引用他人的作品,但必须声明原作者信息。
4. 本版禁止发表出售、求购、或其他非技术讨论等帖子。
5. 本版禁止灌水,包括但不限于任何与所讨论主题无关的回复,无意义字符,直接复制其他回复等。
6. 本站附件禁止用于商业目的,请在下载后24小时内删除,本站不对其造成的结果负任何责任。

Cluster-wide Device Special Files (cDSFs)

Post # 1 by MUDBOY » 2014-08-04 10:28

Because DSF names may be duplicated between one host and other, it is possible for different storage devices to have the same name on different nodes in a cluster, and for the same piece of storage to be addressed by different names.

Cluster-wide Device Special Files (cDSFs) ensure that each storage device used by the cluster has a unique device file name. Cluster-wide device special files (cDSFs) are persistent device special files applied across a set of nodes. That is, they ensure that the same piece of storage has the same device file name on all of the nodes that share it; no matter how many paths there are to the device, the same cluster DSF is used to address it. If the device is moved, the same cDSF still addresses it.

How to Create the Cluster-wide Device Special Files (cDSFs):

Important: Only configure cDSF if you are building a 2 or more nodes cluster. This is not applicable when building a single node cluster

Execute the following command:

# cmsetdsfgroup -c
Cluster device file group configuration set to 'clusternode1 clusternode2'.



Note: The option -c creates a cDSF group from the members of the existing local Serviceguard cluster. The command must be run on a member of the Serviceguard cluster. This option eliminates the need to explicitly list each node of the Serviceguard cluster on the command line.

Check cluster nodes are listed in the output of the following command

# cmsetdsfgroup -q
clusternode1
clusternode2



Run the following command(the disks can be identified using the command ioscan -m lun):

# ioscan -F -m cluster_dsf /dev/disk/disk#
# ioscan -F -m cluster_dsf /dev/disk/disk##




How to migrate/convert VG with vgcdsf

# strings /etc/lvmtab_p
/dev/vgpkg1
A0000000000000001Tue Feb 1 13:21:27 201103b950fa-2d66-11e0-bf02-00156004e156
/dev/disk/disk3
/dev/disk/disk5

# vgcdsf /dev/vgpkg1
Successfully converted persistent DSF(s) in VG /dev/vgpkg1.

# strings /etc/lvmtab_p
/dev/vgpkg1
A0000000000000001Tue Feb 1 13:21:27 201103b950fa-2d66-11e0-bf02-00156004e156
/dev/cdisk/disk1
/dev/cdisk/disk2


提倡分享之精神, 技术问题请在相关版面直接发帖, 站内短信恕不答复.
I don't answer private messages on topics of public interest, because it defeats the whole purpose of a message board.
User avatar
MUDBOY
创始人
 
Posts: 3742
Joined: 2010-12-28

Re: Cluster-wide Device Special Files (cDSFs)

Post # 2 by MUDBOY » 2015-07-01 23:59

对于新添加的node,可使用命令cmsetdsfgroup -a -n <node_name>

对于vgcdsf的转换,我一般先把cluster加入后,然后再执行vgcdsf /dev/<vg_name>命令转换。

然后vgexport出来map文件
root@node1# vgexport -p -m /tmp/<vg_name>.mapfile -s -v /dev/<vg_name>

把mapfile用scp拷贝到其他node
root@node1# scp /tmp/<vg_name>.mapfile node2:/tmp

在node2上执行vgimport
root@node2# vgimport -m /tmp/<vg_name>.mapfile -s -v -N /dev/<vg_name>

注意,如果主机某块硬盘有问题,那么因为vgimport里面的-s参数会扫描全部磁盘,造成命令无响应。这时我可以去掉s参数并指定disk名字
root@node2# vgimport -m /tmp/<vg_name>.mapfile -v /dev/<vg_name> /dev/cdisk/disk1 /dev/cdisk/disk2

提倡分享之精神, 技术问题请在相关版面直接发帖, 站内短信恕不答复.
I don't answer private messages on topics of public interest, because it defeats the whole purpose of a message board.
User avatar
MUDBOY
创始人
 
Posts: 3742
Joined: 2010-12-28


Return to Software & Soluations

Who is online

Users browsing this forum: No registered users and 2 guests