[sheepdog] [PATCH v4 00/12] add reweight support

Liu Yuan namei.unix at gmail.com
Mon May 27 14:53:10 CEST 2013


v4:
 - fix commit log
 - revert changes to default_read

v3:
 - inc epoch for reweighting event
 - add tests/064
 - fix 'node info' and 'md info' as 'df -h' style

v2:
 - fix reweighting cluster with gateway only node
 - use ext4 as mkfs format
 - remove duplicate check in cluster_recovery_completion

This patch set mainly add reweight support after sizes of nodes get changed by
plug/unplug.

The rationale for reweighting instead of auto-reweighting for every plug/unplug
event is that, people don't want plug/unplug or disk crashes in one node put the
the whole cluster in recovery. Instead, people can do manual reweighting at
their will because md event doesn't degrade the data reliability or availibility.

Also fixes 'node info' output

usage:
 $ collie cluster reweight # trigger reweighting and data rebalance across cluster




More information about the sheepdog mailing list