Add option '--copies' to 'cluster snapshot load'

Bug #1279370 reported by sirio81
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
sheepdog
Fix Committed
Undecided
Unassigned

Bug Description

First this but has to be fixed: 1272305.
This will make 'cluster snapshot load' to format the cluster with the same redundance scheme as it was saved.

Rational:

in case of disaster recovery or to restore a cluster snapshot on a separate testing cluster, it may be necessary to set a different redundance scheme.

E.g.

production cluster running with --copies 8:4.
I need to restore it's snapshot on a testing cluster with few (or even one) node.
Another case is that the production cluster it self got 5 or more servers damages.
In this example, I couldn't restore on a cluster with less than 8 nodes.
In any case, I think it's a good thing to be able to choose the number of copies when restoring a 'cluster snapshot'.

New syntax:
dog cluster snapshot {save|list|load} [-a address] [-c copies] [-p port] [-h] <tag|idx> <path>

Tags: whishlist
Liu Yuan (namei-unix)
Changed in sheepdog-project:
status: New → Fix Committed
Revision history for this message
sirio81 (sirio81) wrote :

Is the fix committed in master or devel branch?

Note:
I don't see the new option in the help message

Sheepdog daemon version 0.8.0_101_gbf21156

dog cluster snapshot load --help
Usage: dog cluster snapshot {save|list|load} [-a address] [-p port] [-h] <tag|idx> <path>
Available subcommands:
  save save snapshot to localpath
  list list snapshot of localpath
  load load snapshot from localpath
Options:
  -a, --address specify the daemon address (default: localhost)
  -p, --port specify the daemon port
  -h, --help display this help and exit

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.