printer Print

Customer View


VNX/VNXe Family: Issues with native multipath on Red Hat Enterprise Linux

Article Number: 521715
Version: 4
Article Type: Break Fix
Last Published: 23 Mar 2020
Summary: Red Hat Linux servers loses paths and applications become unavailable and in some cases reboot server
Issue


Red Hat Enterprise Linux 6 and 7 lose devices by issues on paths and native multipath is not witching over to their alternate paths in a timely fashion, which causes production applications to become unavailable and in some cases rebooting.

Check multipath.conf and configuration its the same recommended on Host connectivity guide for Linux.

 Servers have more than one path to device / hba.

 
Cause
  • There are multiple parameters that affect error detection and failover times.
    dev_loss_tmo
    fast_io_fail_tmo
    checker_timeout

  • The dev_loss_tmo (rport) effects extended link timeout, in-flight io is held after a link down event for seconds before the driver gives up waiting for the port to come back. Default is 30-35s, so in-flight io can be held seconds before being killed off. After timeout expiration, rport is put in offline (down) state.

  • The fast_io_fail_tmo (rport) effects how long io is queued and held while rport is in blocked state.

  • checker_timeout specify the timeout to user for path checkers that issue scsi commands with an explicit timeout, in seconds; default taken from /sys/block/sd/device/timeout

  • polling_interval - interval between path checks in seconds.

Resolution
  • Add the following parameters to the defaults{} section in /etc/multipath.conf:
defaults {
        polling_interval 5
        fast_io_fail_tmo 5
        dev_loss_tmo 10
        checker_timeout 15
}
  • Reload multipathd:
# service multipathd reload
Notes
Attachments
Article Properties
First Published Thu May 31 2018
11:02:09 GMT
Primary Product
VNX/VNXe Family
Product
VMAX Family,Unity All-Flash,VNX/VNXe Family