Can I add or remove devices to a fence_scsi or fence_mpath device without restarting the cluster or all the resources?

Solution Verified - Updated

Environment

  • Red Hat Enterprise Linux 6, 7, or 8 (with the High Availability or Resilient Storage Add-on)
  • Pacemaker
  • A fence_scsi or fence_mpath stonith device

Issue

  • When I add or remove storage devices in a cluster that uses fence_scsi or fence_mpath stonith devices, I have to restart the cluster or all the resources in order to update the SCSI persistent reservations.
  • Can I add or remove afence_scsior fence_mpath fence-agent without causing a restart of resources?

Resolution

For fence_scsi:

Red Hat Enterprise Linux 6


There are no plans to add this feature in RHEL 6.

Red Hat Enterprise Linux 7


There are no plans to add this feature in RHEL 7.

Red Hat Enterprise Linux 8

  • The issue (bugzilla bug: 2180706) has been resolved with the errata RHSA-2023:3082 with the following package(s): pcs-0.10.15-4.el8_8.1, pcs-snmp-0.10.15-4.el8_8.1 or later for RHEL 8.8.z.
  • The issue (bugzilla bug: 2179010) has been resolved with the errata RHBA-2023:6903 with the following package(s): pcs-0.10.17-2.el8, pcs-snmp-0.10.17-2.el8 or later.

Red Hat Enterprise Linux 9

  • The issue (bugzilla bug: 2180704) has been resolved with the errata RHSA-2023:2652 with the following package(s): pcs-snmp-0.11.4-7.el9_2, pcs-0.11.4-7.el9_2 or later for RHEL 9.2.z.
  • The issue (bugzilla bug: 2177996) has been resolved with the errata RHSA-2023:6316 with the following package(s): pcs-snmp-0.11.6-3.el9, pcs-0.11.6-3.el9 or later.

With the above errata new scsi devices can be added to the devices attribute for fence_scsi without causing a restart of all resources running on the same node where the stonith resource was running. A new command was added to add new scsi storage devices to an existing fence_scsi stonith devices without causing restart of all other resources.

  • pcs stonith update-scsi-devices <stonith id> set <path to scsi device>

For fence_mpath:

Red Hat Enterprise Linux 8

  • The issue tracked under bugzilla bug 2035332 for RHEL 8.4.z has been closed & deferred as the fix has been released & back-ported to RHEL 8.5.z or later.
  • The issue (bugzilla bug: 1872378) has been resolved with the errata RHSA-2021:4142 with the following package(s): pcs-0.10.10-4.el8, pcs-snmp-0.10.10-4.el8 or later.
  • The issue (bugzilla bug: 2023845) has been resolved with the errata RHSA-2022:7447 with the following package(s): pcs-0.10.14-5.el8, pcs-snmp-0.10.14-5.el8 or later.

Red Hat Enterprise Linux 9

  • The issue (bugzilla bug: 2024522) has been resolved with the errata RHSA-2022:7935 with the following package(s): pcs-snmp-0.11.3-4.el9, pcs-0.11.3-4.el9 or later.

With the above errata new multipath devices can be added to the devices attribute for fence_mpath without causing a restart of all resources running on the same node where the stonith resource was running. A new command was added to add new multipath storage devices to an existing fence_mpath stonith devices without causing restart of all other resources.

  • pcs stonith update-scsi-devices <stonith id> set <path to mpath device>

Note :

Workaround


These workarounds below can apply to `fence_scsi` or `fence_mpath`.
  • If the devices attribute of the stonith device is left blank (and thus automatically populated with a list of the physical volumes composing the clustered volume groups), a cluster restart is required to regenerate the list.
  • If the devices attribute is specified, updating the devices attribute triggers a restart of all cluster resources, since they all depend on the fence_scsi or fence_mpath device. For example, updating the fence configuration using the following command will require a restart of whole cluster or all resources:
# pcs stonith update <stonith id> devices=<path to devices>

Root Cause

When new shared storage devices are added to a Pacemaker cluster, either the whole cluster or all resources must be restarted in order to have fence_scsi or fence_mpath manage SCSI persistent reservations on those devices.

SBR
Product(s)
Components
Category
Tags

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.