Enabling Multi-pathing on ESVA with Red Hat Enterprise Linux 6 Device Mapper Application Note Abstract This document describes how to enable multi-pathing configuration using the Device Mapper service available in Red Hat Enterprise Linux 6.1.
Table of Contents Product Models Covered by This Document... 3 Device Mapper Multi-path Prerequisites... 3 ESVA Introduction... 4 Key Features of Device Mapper Multi-pathing... 5 Configurations for ESVA FC-host Systems... 6 Configurations for ESVA iscsi-host Systems... 9 2
Product Models Covered by This Document This application note applies to the following product models: ESVA Fibre Channel (FC)-host systems ESVA iscsi-host systems For more information regarding individual product models, please visit www.infortrend.com. Device Mapper Multi-path Prerequisites Hardware and software prerequisites include the following: Operating system (OS) version: Red Hat Enterprise Linux 6.1 Host Bus Adapters (HBA) and SAN switches Supported arrays: ESVA FC-host and iscsi-host systems Ensure the device-mapper-multipath rpm has been installed 3
ESVA Introduction Infortrend s ESVA (Enterprise Scalable Virtualized Architecture) is a leading storage solution designed for mid-range enterprise SAN. At affordable prices, it meets mission-critical storage demands for performance, scalability and reliability with advanced hardware designs and comprehensive data services. On the innovative Enterprise Scalable Virtualized Architecture, various features, including storage virtualization, thin provisioning, distributed load balancing, automated storage tiering, automatic data migration, prioritized volume accessibility, snapshot and replication, are consolidated to realize optimal business benefits. With ESVA systems, users can optimize ROI, simplify storage infrastructure, and maximize application productivity. For more information about Infortrend s ESVA storage systems, please visit our website: http://www.infortrend.com. 4
Key Features of Device Mapper Multi-pathing Multi-pathing with Device Mapper offers two key features: Data path redundancy: Offering failover in an active/passive configuration, Device Mapper multi-pathing can help users eliminate downtime caused by data path failure. If any element of an I/O path (the cable, switch, or controller) fails, I/Os will be directed to alternate data paths. Enhanced performance: Device Mapper multi-pathing can be configured in active/active mode, whereby workloads can be balanced by spreading I/O over available data paths in a round-robin fashion. In addition, in some configurations Device Mapper multi-pathing can detect loading on I/O paths and dynamically re-balance workloads. For more information, please refer to official Red Hat Enterprise Linux documentation at: https://access.redhat.com/knowledge/docs/red_hat_enterprise_linux/ 5
Configurations for ESVA FC-host Systems Step 1: Uninstall Infortrend s EonPath driver if it has previously been installed on your application server. Step 2: Ensure the device-mapper-multipath rpm has been installed. Step 3: Configure your ESVA system and complete relevant mapping processes, parameter tuning, etc. Step 4: Change the last section of the configuration file as shown in the example text box below. The file can be found in /etc/multipath.conf. If you are already using a valid configuration file and do not have a device subsection for the ESVA arrays, please add the appropriate device subsection for the respective arrays. If you already have a device subsection for ESVA arrays which has different parameters from the entries below, modify it accordingly. devices { device { vendor path_grouping_policy getuid_callout path_checker "IFT" group_by_prio "/lib/udev/scsi_id --whitelisted --device=/dev/%n" readsector0 path_selector "round-robin 0" hardware_handler "0" failback 15 rr_weight uniform no_path_retry 12 prio alua } } Step 5: Reboot your system. Step 6: Restart the multi-pathing service: /etc/init.d/multipathd restart Stopping multipathd daemon [ OK ] Stopping multipathd daemon [ OK ] 6
Step 7: Check the availability of your multi-path devices. Check the device list: multipath -ll Devices should appear as follows: mpath1 (3600d023100037dfb000000007e2ae926) dm-2 IFT,F60-2830 [size=20g][features=0][hwhandler=0] \_round-robin 0 [prio=1][active] \_4:0:0:0 sdb 8:16 [active][ready] \_round-robin 0 [prio=1][enabled] \_5:0:0:0 sdc 8:32 [active][ready] Check disk partitions, you should be able to see fault-tolerant links dm-2 and dm-3: more /proc/partitions major minor #blocks name 8 0 244198584 sda 8 1 104391 sda1 8 2 244091610 sda2 253 0 242057216 dm-0 253 1 2031616 dm-1 8 16 20971520 sdb 8 17 20964793 sdb1 8 32 59180032 sdc 8 33 59180016 sdc1 253 2 20971520 dm-2 253 3 20964793 dm-3 Note: If you can not find multi-path devices, check the following: 1. Check blacklist devices to see if the ESVA arrays have been indicated in the blacklist, and then restart the multi-path devices. Device Mapper s default is to blacklist all devices. Add # marks in front of blacklist lines to disable the blacklist setting. 2. Enter five spaces after the vendor name to cover the range of Device Mapper version variables, e.g., IFT. Restart multipath service. 7
Step 8: You may then start formatting and using these devices. mkfs /dev/dm-0 mkfs /dev/dm-1 mkfs /dev/dm-2 mkfs /dev/dm-3. 8
Configurations for ESVA iscsi-host Systems Step 1: Uninstall Infortrend s EonPath driver if it has previously been installed on your application server. Step 2: Ensure the device-mapper-multipath rpm has been installed. Step 3: Configure your ESVA system and complete relevant mapping processes, parameter tuning, etc. Step 4: Enable the OS-embedded iscsi initiator as shown in the following example: rpm -ivh iscsi-initiator-utils-6.2.0.868-0.7.el5.i386.rpm Step 5: Check the iscsi initiator version: iscsiadm V Step 6: Enable initiators to find targets (logical drives mapped to the host) by specifying iscsi host port addresses: iscsiadm -m discovery -t sendtargets -p 192.168.10.100:3260 iscsiadm -m discovery -t sendtargets -p 192.168.10.101:3260 iscsiadm -m discovery -t sendtargets -p 192.168.10.102:3260 iscsiadm -m discovery -t sendtargets -p 192.168.10.103:3260 Step 7: Login these iscsi targets: Iscsiadm m node --loginall=all (can be abbreviated as L all) Step 8: Check if all targets are present in the OS file system: cat /proc/partitions Step 9: Change the last section of the configuration file as shown in the example text box below. The file can be found in /etc/multipath.conf. If you are already using a valid configuration file and do not have a device subsection for the ESVA arrays, please add the appropriate device subsection for the respective arrays. If you already have a device subsection for ESVA arrays which has different parameters from the entries below, modify it accordingly. 9
devices { device { vendor path_grouping_policy getuid_callout path_checker "IFT" group_by_prio "/lib/udev/scsi_id --whitelisted --device=/dev/%n" readsector0 path_selector "round-robin 0" hardware_handler "0" failback 15 rr_weight uniform no_path_retry 12 prio alua } } Step 10: Locate and open an iscsi configuration file at /etc/iscsi/iscsid.conf (the file can also be iscsi.conf) and change the following parameters: node.conn[0].timeo.noop_out_interval = 30 (default is 5) node.comm[0].timeo.noop_out_timeout = 90 (default is 5) Step 11: Reboot your system. Step 12: Restart the multi-path service. /etc/init.d/multipathd restart Stopping multipathd daemon [ OK ] Stopping multipathd daemon [ OK ] Step 13: Check the availability of your multipath devices. Check the device list: multipath -ll Devices should appear as follows: mpath1 (3600d023100037dfb000000007e2ae926) dm-2 IFT, E60-2130 [size=20g][features=0][hwhandler=0] \_round-robin 0 [prio=1][active] \_4:0:0:0 sdb 8:16 [active][ready] \_round-robin 0 [prio=1][enabled] \_5:0:0:0 sdc 8:32 [active][ready] 10
Check disk partitions. You should be able to see fault-tolerant links dm-2 and dm-3: more /proc/partitions major minor #blocks name 8 0 244198584 sda 8 1 104391 sda1 8 2 244091610 sda2 253 0 242057216 dm-0 253 1 2031616 dm-1 8 16 20971520 sdb 8 17 20964793 sdb1 8 32 59180032 sdc 8 33 59180016 sdc1 253 2 20971520 dm-2 253 3 20964793 dm-3 Note: If you can not find multi-path devices, check the following: 1. Check blacklist devices to see if the ESVA arrays have been indicated in the blacklist, and then restart the multi-path devices. Device Mapper s default is to blacklist all devices. Add # marks in front of blacklist lines to disable the blacklist setting. 2. Enter five spaces after the vendor name to cover the range of Device Mapper version variables, e.g., IFT. Restart multipath service. Step 14: You may then start formatting and using these devices. mkfs /dev/dm-0 mkfs /dev/dm-1 mkfs /dev/dm-2 mkfs /dev/dm-3.. 11 Copyright 2012 Infortrend Technology, Inc. All rights reserved. Infortrend, ESVA, EonStor, EonNAS and EonPath are trademarks or registered trademarks of Infortrend. All other marks and names mentioned herein may be trademarks of their respective owners. The information contained herein is subject to change without notice. The content provided as is, without express or implied warranties of any kind. AN_EV_2012003_GL_1.1