[Bug] occasional crashes in v1.3.7: "panic: sync: WaitGroup is reused before previous Wait has returned"

[skladd]

InfluxDB: version 1.3.7 (from linux_armhf release tarball)
OS: Linux 4.1.41 on armv7l

I experienced these crashes several times, also for older versions. There are points written all the time, also deletes are performed regularly by cron jobs and scripts.

[I] 2017-11-13T08:58:56Z Snapshot for path /var/lib/influxdb/data/_internal/monitor/32 written in 1.303492817s engine=tsm1
[I] 2017-11-13T08:58:57Z beginning level 1 compaction of group 0, 2 TSM files engine=tsm1
[I] 2017-11-13T08:58:57Z compacting level 1 group (0) /var/lib/influxdb/data/_internal/monitor/32/000000010-000000001.tsm (#0) engine=tsm1
[I] 2017-11-13T08:58:57Z compacting level 1 group (0) /var/lib/influxdb/data/_internal/monitor/32/000000011-000000001.tsm (#1) engine=tsm1
[I] 2017-11-13T08:59:00Z compacted level 1 group (0) into /var/lib/influxdb/data/_internal/monitor/32/000000011-000000002.tsm.tmp (#0) engine=tsm1
[I] 2017-11-13T08:59:00Z compacted level 1 2 files into 1 files in 2.476008924s engine=tsm1
[I] 2017-11-13T09:01:34Z retention policy shard deletion check commencing service=retention
[I] 2017-11-13T09:31:34Z retention policy shard deletion check commencing service=retention
panic: sync: WaitGroup is reused before previous Wait has returned

goroutine 288 [running]:
sync.(*WaitGroup).Wait(0x114d2018)
	/usr/local/go/src/sync/waitgroup.go:133 +0x100
github.com/influxdata/influxdb/tsdb/engine/tsm1.(*Engine).disableLevelCompactions(0x114d2000, 0x11228100)
	/go/src/github.com/influxdata/influxdb/tsdb/engine/tsm1/engine.go:272 +0x68
github.com/influxdata/influxdb/tsdb/engine/tsm1.(*Engine).SetCompactionsEnabled(0x114d2000, 0x8eb200)
	/go/src/github.com/influxdata/influxdb/tsdb/engine/tsm1/engine.go:207 +0x60
github.com/influxdata/influxdb/tsdb.(*Shard).SetCompactionsEnabled(0x11228100, 0x10d62d00)
	/go/src/github.com/influxdata/influxdb/tsdb/shard.go:424 +0x48
github.com/influxdata/influxdb/tsdb.(*Store).monitorShards(0x10d561c0)
	/go/src/github.com/influxdata/influxdb/tsdb/store.go:1247 +0x1d8
created by github.com/influxdata/influxdb/tsdb.(*Store).Open
	/go/src/github.com/influxdata/influxdb/tsdb/store.go:150 +0x21c
[I] 2017-11-14T02:01:23Z InfluxDB starting, version 1.3.7, branch 1.3, commit 2d474a3089bcfce6b472779be9470a1f0ef3d5e4
[I] 2017-11-14T02:01:23Z Go version go1.8.4, GOMAXPROCS set to 2

I changed some settings to reduce the memory footprint, in case they do matter:

query-log-enabled = false
cache-max-memory-size = 16777216 # 16 MB instead of 1048576000
reporting-disabled = true
logging-enabled = false
cache-snapshot-memory-size = 8388608 # 8 MB instead of 26214400
max-concurrent-queries = 50
query-timeout = "60s"
log-queries-after = "30s"
max-select-series = 1024
log-enabled = false
max-connection-limit = 20

[skladd]

Thank you very much for fixing.
Is version 1.5.0 to be released soon?
Could you otherwise please integrate the fix in a 1.4.x release?