Hide Forgot
Description of problem: path_selector default hwtable entry for EMC CLARiiON is "round-robin 0" Version-Release number of selected component (if applicable): kernel: 2.6.32-71.18.2.el6.x86_64 device-mapper-multipath: device-mapper-multipath-0.4.9-31.el6_0.3.x86_64 How reproducible: N/A Steps to Reproduce: 1. 2. 3. Actual results: path_selector "round-robin 0" Expected results: path_selector "service-time 0" Additional info: EMC recommended configuration per EMC® Host Connectivity Guide for Linux P/N 300-003-865 REV A23 device { vendor "DGC" product ".*" product_blacklist "LUNZ" path_grouping_policy group_by_prio getuid_callout "/lib/udev/scsi_id --whitelisted --device=/dev/%n" path_selector "service-time 0" path_checker emc_clariion features "1 queue_if_no_path" hardware_handler "1 emc" prio emc failback immediate rr_weight uniform no_path_retry 60 rr_min_io 1000 }
Disk /dev/mapper/mpathb: 35.2 GB, 35163209728 bytes 255 heads, 63 sectors/track, 4275 cylinders, total 68678144 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x86336388 Device Boot Start End Blocks Id System /dev/mapper/mpathbp1 2048 68678143 34338048 83 Linux
Basic single thread I/O benchmark: mpathb (3600601f0d057000019fc7845f46fe011) dm-5 DGC,DISK size=33G features='1 queue_if_no_path' hwhandler='1 emc' wp=rw `-+- policy='service-time 0' prio=1 status=active |- 7:0:0:1 sdc 8:32 active ready running `- 6:0:0:1 sdf 8:80 active ready running # sync; echo 3 > /proc/sys/vm/drop_caches # sync; time `dd if=/dev/zero of=/san_disk/test.4096 bs=4096 count=262144; sync` 262144+0 records in 262144+0 records out 1073741824 bytes (1.1 GB) copied, 2.73627 s, 392 MB/s real 0m30.898s user 0m0.046s sys 0m1.536s
mpathb (3600601f0d057000019fc7845f46fe011) dm-5 DGC,DISK size=33G features='1 queue_if_no_path' hwhandler='1 emc' wp=rw `-+- policy='round-robin 0' prio=1 status=active |- 7:0:0:1 sdc 8:32 active ready running `- 6:0:0:1 sdf 8:80 active ready running # sync; echo 3 > /proc/sys/vm/drop_caches # sync; time `dd if=/dev/zero of=/san_disk/test.4096 bs=4096 count=262144; sync` 262144+0 records in 262144+0 records out 1073741824 bytes (1.1 GB) copied, 3.21566 s, 334 MB/s real 0m32.123s user 0m0.040s sys 0m1.555s
blacklist { devnode "^(ram|raw|loop|fd|md|dm-|sr|scd|st)[0-9]*" devnode "^hd[a-z]" devnode "^dcssblk[0-9]*" devnode "^sda[0-9]*" } defaults { user_friendly_names yes } devices } device { vendor "DGC" product ".*" product_blacklist "LUNZ" getuid_callout "/lib/udev/scsi_id --whitelisted --device=/dev/%n" prio_callout "/sbin/mpath_prio_emc /dev/%n" features "1 queue_if_no_path" hardware_handler "1 emc" path_selector "service-time 0" #path_selector "round-robin 0" path_grouping_policy group_by_prio failback immediate rr_weight uniform no_path_retry 60 rr_min_io 1000 path_checker emc_clariion prio emc } }
I'm confused. We take our default configs from the hardware vendors. As far as I know, EMC hasn't sent us a config with the default selector set to service time. I'm actually pushing to have remove the path_selector default from most of the configs in the devices section, unless the vendor specifically asks for it to stay. This will allow users to set the path selector want in the defaults section, without having to edit each device config separately. That being said, If someone from EMC wants to verify that EMC does want to use service-time as the default, I'll be more than happy to switch it.
Since RHEL 6.1 External Beta has begun, and this bug remains unresolved, it has been rejected as it is not proposed as exception or blocker. Red Hat invites you to ask your support representative to propose this request, if appropriate and relevant, in the next release of Red Hat Enterprise Linux.
I'm not changing this. I will be push a patch to remove the path_selector from the default configs on all devices configs unless the storage vendor specifically asks for it.