cerc-io/lotus
10
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 27 Wiki Activity

keep retrying the proof until we run out of sectors to skip

Browse Source 
If we have a bunch of corrupted but not missing sectors on disk, we may need to
retry many times before we get a proof to pass. Simply giving up doesn't help anyone.
This commit is contained in:
Steven Allen 2020-10-28 18:10:29 -07:00
parent 6cdbf5216e
commit 6985af5e88
1 changed files with 14 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
storage/wdpost_run.go
View File

@ -510,10 +510,10 @@ func (s *WindowPoStScheduler) runPost(ctx context.Context, di dline.Info, ts *ty
skipCount := uint64(0) skipCount := uint64(0)
postSkipped := bitfield.New() postSkipped := bitfield.New()
var postOut []proof2.PoStProof somethingToProve := false
somethingToProve := true
for retries := 0; retries < 5; retries++ { // Retry until we run out of sectors to prove.
for retries := 0; ; retries++ {
var partitions []miner.PoStPartition var partitions []miner.PoStPartition
var sinfos []proof2.SectorInfo var sinfos []proof2.SectorInfo
for partIdx, partition := range batch { for partIdx, partition := range batch {
@ -567,7 +567,6 @@ func (s *WindowPoStScheduler) runPost(ctx context.Context, di dline.Info, ts *ty
if len(sinfos) == 0 { if len(sinfos) == 0 {
// nothing to prove for this batch // nothing to prove for this batch
somethingToProve = false
break break
} }
@ -585,24 +584,31 @@ func (s *WindowPoStScheduler) runPost(ctx context.Context, di dline.Info, ts *ty
return nil, err return nil, err
} }
var ps []abi.SectorID postOut, ps, err := s.prover.GenerateWindowPoSt(ctx, abi.ActorID(mid), sinfos, abi.PoStRandomness(rand))
postOut, ps, err = s.prover.GenerateWindowPoSt(ctx, abi.ActorID(mid), sinfos, abi.PoStRandomness(rand))
elapsed := time.Since(tsStart) elapsed := time.Since(tsStart)
log.Infow("computing window post", "batch", batchIdx, "elapsed", elapsed) log.Infow("computing window post", "batch", batchIdx, "elapsed", elapsed)
if err == nil { if err == nil {
// Proof generation successful, stop retrying if len(postOut) == 0 {
params.Partitions = append(params.Partitions, partitions...) return nil, xerrors.Errorf("received no proofs back from generate window post")
}
// Proof generation successful, stop retrying
somethingToProve = true
params.Partitions = partitions
params.Proofs = postOut
break break
} }
// Proof generation failed, so retry // Proof generation failed, so retry
if len(ps) == 0 { if len(ps) == 0 {
// If we didn't skip any new sectors, we failed
// for some other reason and we need to abort.
return nil, xerrors.Errorf("running window post failed: %w", err) return nil, xerrors.Errorf("running window post failed: %w", err)
} }
// TODO: maybe mark these as faulty somewhere?
log.Warnw("generate window post skipped sectors", "sectors", ps, "error", err, "try", retries) log.Warnw("generate window post skipped sectors", "sectors", ps, "error", err, "try", retries)
@ -617,12 +623,6 @@ func (s *WindowPoStScheduler) runPost(ctx context.Context, di dline.Info, ts *ty
continue continue
} }
if len(postOut) == 0 {
return nil, xerrors.Errorf("received no proofs back from generate window post")
}
params.Proofs = postOut
posts = append(posts, params) posts = append(posts, params)
} }