panic while droping big measurement with about 1 million series

__Steps to reproduce:__
List the minimal actions needed to reproduce the behavior.

1. drop measurement "measurement_name"

__Expected behavior:__
drop success

__Actual behavior:__
influxdb server panic

__Environment info:__

* System info: Linux 3.10.0-327.el7.x86_64 x86_64
* InfluxDB version: InfluxDB v1.7.7 (git: 1.7 f8fdf652f348fc9980997fe1c972e2b79ddd13b0)
* Other relevant environment details: physical machine， centos 7.2

__Config:__
[influxdb_conf.zip](https://github.com/influxdata/influxdb/files/3777869/influxdb_conf.zip)

__Logs:__
[log.zip](https://github.com/influxdata/influxdb/files/3777928/log.zip)

__Performance:__
get profiles fail, return 404 page not found
[vars.txt](https://github.com/influxdata/influxdb/files/3777935/vars.txt)
[iostat.txt](https://github.com/influxdata/influxdb/files/3777966/iostat.txt)

Start by reproducing the drop of measurement "measurement_name" with about one million series on InfluxDB v1.7.7, then inspect log.zip and influxdb_conf.zip for the panic context and relevant configuration. Use vars.txt and iostat.txt for supporting performance data. Done means the measurement drop succeeds without the InfluxDB server panicking.

Written by the indexing model from the issue text.