1830241 Members
10113 Online
109999 Solutions
New Discussion

Multipath

 
tarek_3
Frequent Advisor

Multipath

Hi.

We have a MS500 G2 storage attached to a DL380 using two Smart Array 642 SCSI running Red Hat Enterprise Server 4.

We have 1 LUN and this LUN has two paths in Linux:
/dev/cciss/c1d0
/dev/cciss/c2d0

When we tried to enable Multipath with mdadm mannualy, it works, but we cannot do this change persistent on boot.

Add /dev/md0 with command:
mdadm -C /dev/md0 -l multipath -n 2 /dev/cciss/c1d0p1 /dev/cciss/c2d0p1
WORK.

Reboot the system - doesn't work, and this message appears:
mdadm: failed to add /dev/cciss/c1d0p1 to /dev/md0: Device or resource busy
mdadm: /dev/md0 has been started with 2 drives and -1 spares.

Please, help me.
4 REPLIES 4
Jerome Henry
Honored Contributor

Re: Multipath

Hi hassoun,

Do you have /boot on /dev/md0 ?

J
You can lean only on what resists you...
cristian mazza
Advisor

Re: Multipath

you must set the partition to type fd "linux auto raid" in this way the boot process discover the partitions and assign it to md device.

also, you can modify the file/etc/mdadm.conf to insert the md device and the phisical disk.

regards, Cristian
Natasha3
Occasional Advisor

Re: Multipath

I have a server having two hdds (/dev/sdc and /dev/sdd) . Both hdds are same physical disk sent by two different paths to the sever.  I am trying to create a multipath raid device using the following command :
"mdadm --create /dev/md0 --level=multipath --raid-devices=2 /dev/sdd /dev/sdc "
But facing error 
" mdadm: RUN_ARRAY failed: Invalid argument  "

my /etc/mdadm.conf  file
DEVICE /dev/sdd /dev/sdc
ARRAY /dev/md0 metadata=1.2 name=myarray:0 UUID=261fb9c:976d0d97:30bc63ce:85e76e91



Can anyone help me with this?

Sunitha_Mod
Honored Contributor

Re: Multipath

Hello @Natasha3,

Thank you for posting! 

Since you have posted in an old topic and there is no response yet, we would recommend you create a new topic using the create "New Discussion" button, so the experts can check and assist you further.