Skip to content

Fix/lifecycler/stuck in leaving #1921

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@
* [FEATURE] Added `global` ingestion rate limiter strategy. Deprecated `-distributor.limiter-reload-period` flag. #1766
* [FEATURE] Added support for Microsoft Azure blob storage to be used for storing chunk data. #1913
* [BUGFIX] Fixed unnecessary CAS operations done by the HA tracker when the jitter is enabled. #1861
* [BUGFIX] Fixed #1904 ingesters getting stuck in a LEAVING state after coming up from an ungraceful exit. #1921

## 0.4.0 / 2019-12-02

Expand Down
6 changes: 6 additions & 0 deletions pkg/ring/lifecycler.go
Original file line number Diff line number Diff line change
Expand Up @@ -497,6 +497,12 @@ func (i *Lifecycler) initRing(ctx context.Context) error {
return ringDesc, true, nil
}

// If the ingester failed to clean it's ring entry up in can leave it's state in LEAVING.
// Move it into ACTIVE to ensure the ingester joins the ring.
if ingesterDesc.State == LEAVING && len(ingesterDesc.Tokens) == i.cfg.NumTokens {
ingesterDesc.State = ACTIVE
}

// We exist in the ring, so assume the ring is right and copy out tokens & state out of there.
i.setState(ingesterDesc.State)
tokens, _ := ringDesc.TokensFor(i.ID)
Expand Down
57 changes: 57 additions & 0 deletions pkg/ring/lifecycler_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -418,3 +418,60 @@ func TestTokensOnDisk(t *testing.T) {
require.Equal(t, expTokens, actTokens)
}
}

// JoinInLeavingState ensures that if the lifecycler starts up and the ring already has it in a LEAVING state that it still is able to auto join
func TestJoinInLeavingState(t *testing.T) {
var ringConfig Config
flagext.DefaultValues(&ringConfig)
codec := GetCodec()
ringConfig.KVStore.Mock = consul.NewInMemoryClient(codec)

r, err := New(ringConfig, "ingester", IngesterRingKey)
require.NoError(t, err)
defer r.Stop()

cfg := testLifecyclerConfig(ringConfig, "ing1")
cfg.NumTokens = 2
cfg.MinReadyDuration = 1 * time.Nanosecond

// Set state as LEAVING
err = r.KVClient.CAS(context.Background(), IngesterRingKey, func(in interface{}) (interface{}, bool, error) {
r := &Desc{
Ingesters: map[string]IngesterDesc{
"ing1": {
State: LEAVING,
Tokens: []uint32{1, 4},
},
},
Tokens: []TokenDesc{
{
Ingester: "ing2",
Token: 2,
},
{
Ingester: "ing2",
Token: 3,
},
},
}

return r, true, nil
})
require.NoError(t, err)

l1, err := NewLifecycler(cfg, &nopFlushTransferer{}, "ingester", IngesterRingKey)
l1.Start()
require.NoError(t, err)

// Check that the lifecycler was able to join after coming up in LEAVING
test.Poll(t, 1000*time.Millisecond, true, func() interface{} {
d, err := r.KVClient.Get(context.Background(), IngesterRingKey)
require.NoError(t, err)
desc, ok := d.(*Desc)
return ok &&
len(desc.Ingesters) == 1 &&
desc.Ingesters["ing1"].State == ACTIVE &&
len(desc.Ingesters["ing1"].Tokens) == cfg.NumTokens &&
len(desc.Tokens) == 2
})
}