CLI Guide

Table 8. ds summary field descriptions (continued)
Field Description
● requires-resume-after-data-loss-failure -
There have been at least two concurrent failures, and data
has been lost. For example, a director fails shortly after the
inter-cluster link fails, or when two directors fail at almost
the same time. Use the consistency-group resume-
after-data-loss-failure command to select a
winning cluster and allow I/O to resume.
● cluster-departure - Not all the visible clusters are in
communication.
● requires-resume-after-rollback - A cluster has
detached its peer cluster and rolled back the view
of data, but is awaiting the consistency-group
resume-after-rollback command before resuming
I/O. Displayed:
○ At the winning side when a detach rule fires, or shortly
after the consistency-group choose-winner
command picks a winning cluster.
requires-resume-at-loser - Displayed on the
losing side when the inter-cluster link heals after an
outage. After the inter-cluster link is restored, the
losing cluster discovers that its peer was declared
the winner and resumed I/O. Use the consistency-
group resume-at-loser command to make the
view of data consistent with the winner, and to resume
I/O at the loser.
restore-link-or-choose-winner - I/O is
suspended at all clusters because of a cluster
departure, and cannot automatically resume. This can
happen if:
○ There is no detach-rule
○ If the detach-rule is 'no-automatic-winner', or
○ If the detach-rule cannot fire because its conditions are
not met.
For example, if more than one cluster is active at
the time of an inter-cluster link outage, the 'active-
cluster-wins' rule cannot take effect. When this
detail is present, I/O will not resume until either the
inter-cluster link is restored, or the user intervenes
to select a winning cluster with the consistency-
group choose-winner command.
unhealthy-devices - I/O has stopped in this
consistency group because one or more volumes is
unhealthy and cannot perform I/O.
will-rollback-on-link-down - If there were a
link-down now, the winning cluster would have to roll
back the view of data in order to resume I/O.
Examples
Display summary information when no devices are unhealthy:
VPlexcli:/distributed-storage> ds summary
Slot usage summary:
Total 912 slots used by distributed device logging segments.
Distributed devices health summary:
Commands
165