Get [panic:runtime error: index out of range] error after few days running

__Steps to reproduce:__

This issue seems difficult to reproduce as it might be related to data we have inside database and it appear after few days only.
I'll describe what is happening.

1. InfluxDB is started
2. We perform regular queries like this one: `SELECT last("users") FROM product.r_90d.usage WHERE time > now() - 380d GROUP BY time(1d), id fill(none)`
3. After few days running, the above query result in an error `[panic:runtime error: index out of range]` each time it is run.

__Expected behavior:__

A result is returned to the query

__Actual behavior:__

The following error is returned: `[panic:runtime error: index out of range]`

__Environment info:__

* System info: `Linux 4.15.0-50-generic x86_64`
* InfluxDB version: `InfluxDB v1.7.10 (git: 1.7 f46f63d4e2d9684a2dd716594ab609ccd32f0a5b)`
* Other relevant environment details: Container runtime, disk info, etc:
- Running in Kubernetes cluster, using Docker 18.06.1-ce as runtime engine.
- Using official docker image from https://hub.docker.com/_/influxdb
- Distribution is Ubuntu 18.04

__Config:__
```
reporting-disabled = true
bind-address = ":8088"

[meta]
dir = "/var/lib/influxdb/meta"

[data]
dir = "/var/lib/influxdb/data"
engine = "tsm1"
wal-dir = "/var/lib/influxdb/wal"
max-values-per-tag = 0
max-series-per-database = 3000000
index-version = "tsi1"
```

__Performance:__
I have generated a pprof, but I couldn't generate it when the bug is actually happening. I don't know if my pprof will be usefull then.
I'm ready to provide any additional information. The bug will certainly happen again in few days. Let me know if there are commands to run when it happens.