White Papers

Dell HPC NFS Storage Solution High Availability Configurations with Large Capacities
60
Make sure the host port identifiers on each R710 match the sas_address from
/sys/class/sas_phy/phy-1:0/sas_address and /sys/class/sas_phy/phy-2:0/sas_address.
6. On each R710, run the command rescan_dm_devs to detect all the virtual disks.
7. On each R710, cat /proc/partitions and multipath ll should show all the LUNs on the
storage.
Reference: Configuration: Device Mapper Multipath for Linux,
http://support.dell.com/support/edocs/systems/md3200/en/OM/HTML/config_n.htm
8. Make sure each vitual disk on two R710s has the same wwid. For example, check the output of
multipath ll | grep MD32xx, and every device with same name should have same wwids, as
below:
mpatha (36842b2b0004c256400000db24e95a162) dm-4 DELL,MD32xx (from active)
mpatha (36842b2b0004c256400000db24e95a162) dm-4 DELL,MD32xx (from passive)
If the wwids do not match, execute the following commands one by one:
a. On machine named as “passive”:
# service multipathd stop
# multipath F
b. Copy /etc/bindings from machine named as “active” to “passive”
c. On “passive”:
# service multipathd start
NOTE:
Under certain conditions, on solutions with many arrays (288 TB and bigger), multipathd may
start before all the virtual disks from the MD3200 are detected and ready to be used, resulting in
missing links under /dev/mapper and other problems related to those missing links.
A way to check for this problem is to compare the number of virtual disks or LUNs your NSS has
with the number of /dev/mapper links. As an example, a 288 TB solution must have 8 links:
# ls -l /dev/mapper | grep mpath | wc l
8
If this problem occurs (number of links < 8), a workaround is to send the service a reload after a
couple minutes, to make sure all the virtual disks are already visible to multipathd.
Edit the file /etc/rc.local on both servers and append the line:
sleep 120; service multipathd reload
Please note that this workaround will slow down the server boot time by 120 seconds