zrepl/cmd/prune.go
Christian Schwarz e951beaef5 Simplify CLI by requiring explicit job names.
Job names are derived from job type + user-defined name in config file
CLI now has subcommands corresponding 1:1 to the config file sections:

        push,pull,autosnap,prune

A subcommand always expects a job name, thus executes exactly one job.

Dict-style syntax also used for PullACL and Sink sections.

jobrun package is currently only used for autosnap, all others need to
be invoked repeatedly via external tool.
Plan is to re-integrate jobrun in an explicit daemon-mode (subcommand).
2017-07-08 11:13:50 +02:00

140 lines
3.2 KiB
Go

package cmd
import (
"fmt"
"github.com/spf13/cobra"
"github.com/zrepl/zrepl/util"
"github.com/zrepl/zrepl/zfs"
"os"
"sort"
"time"
)
var pruneArgs struct {
job string
dryRun bool
}
var PruneCmd = &cobra.Command{
Use: "prune",
Short: "perform pruning",
Run: cmdPrune,
}
func init() {
PruneCmd.Flags().StringVar(&pruneArgs.job, "job", "", "job to run")
PruneCmd.Flags().BoolVarP(&pruneArgs.dryRun, "dryrun", "n", false, "dry run")
RootCmd.AddCommand(PruneCmd)
}
func cmdPrune(cmd *cobra.Command, args []string) {
if len(args) < 1 {
log.Printf("must specify exactly one job as positional argument")
os.Exit(1)
}
job, ok := conf.Prunes[args[0]]
if !ok {
log.Printf("could not find prune job: %s", args[0])
os.Exit(1)
}
log.Printf("Beginning prune job:\n%s", job)
ctx := PruneContext{job, time.Now(), pruneArgs.dryRun}
err := doPrune(ctx, log)
if err != nil {
log.Printf("Prune job failed with error: %s", err)
os.Exit(1)
}
}
type PruneContext struct {
Prune *Prune
Now time.Time
DryRun bool
}
type retentionGridAdaptor struct {
zfs.FilesystemVersion
}
func (a retentionGridAdaptor) Date() time.Time {
return a.Creation
}
func (a retentionGridAdaptor) LessThan(b util.RetentionGridEntry) bool {
return a.CreateTXG < b.(retentionGridAdaptor).CreateTXG
}
func doPrune(ctx PruneContext, log Logger) error {
if ctx.DryRun {
log.Printf("doing dry run")
}
prune := ctx.Prune
// ZFSListSnapsFiltered --> todo can extend fsfilter or need new? Have already something per fs
// Dedicated snapshot object? Adaptor object to FilesystemVersion?
filesystems, err := zfs.ZFSListMapping(prune.DatasetFilter)
if err != nil {
return fmt.Errorf("error applying filesystem filter: %s", err)
}
for _, fs := range filesystems {
fsversions, err := zfs.ZFSListFilesystemVersions(fs, prune.SnapshotFilter)
if err != nil {
return fmt.Errorf("error listing filesytem versions of %s: %s", fs, err)
}
if len(fsversions) == 0 {
continue
}
adaptors := make([]util.RetentionGridEntry, len(fsversions))
for fsv := range fsversions {
adaptors[fsv] = retentionGridAdaptor{fsversions[fsv]}
}
sort.SliceStable(adaptors, func(i, j int) bool {
return adaptors[i].LessThan(adaptors[j])
})
ctx.Now = adaptors[len(adaptors)-1].Date()
describe := func(a retentionGridAdaptor) string {
timeSince := a.Creation.Sub(ctx.Now)
const day time.Duration = 24 * time.Hour
days := timeSince / day
remainder := timeSince % day
return fmt.Sprintf("%s@%dd%s from now", a.ToAbsPath(fs), days, remainder)
}
keep, remove := prune.RetentionPolicy.FitEntries(ctx.Now, adaptors)
for _, a := range remove {
r := a.(retentionGridAdaptor)
log.Printf("remove %s", describe(r))
// do echo what we'll do and exec zfs destroy if not dry run
// special handling for EBUSY (zfs hold)
// error handling for clones? just echo to cli, skip over, and exit with non-zero status code (we're idempotent)
if !ctx.DryRun {
err := zfs.ZFSDestroy(r.ToAbsPath(fs))
if err != nil {
// handle
log.Printf("error: %s", err)
}
}
}
for _, a := range keep {
r := a.(retentionGridAdaptor)
log.Printf("would keep %s", describe(r))
}
}
return nil
}