## My Environment

* __ArangoDB Version__:        <!-- e.g. 3.3.19 or self-compiled devel branch -->
* __Storage Engine__:             <!-- MMFiles / RocksDB -->
* __Deployment Mode__:       <!-- Single Server | Master/Slave | Active Failover | Cluster | DC2DC -->
* __Deployment Strategy__:   <!-- Manual Start | Manual Start in Docker | ArangoDB Starter | ArangoDB Starter in Docker | Kubernetes | DCOS -->
* __Configuration__:               <!-- cluster setup details, notable server settings, etc. -->
* __Infrastructure__:               <!-- AWS | Azure | ... | own -->
* __Operating System__:        <!-- Ubuntu 18.04 | Windows 10 | MacOS 10.13.4 | DCOS 1.10 | ... -->
* __Total RAM in your machine__:        <!-- e.g. 32Gb. If more machines are in use, provide info for each of your machines -->
* __Disks in use__:        <!-- SSD | HDD. If more machines are in use, provide info for each of your machines -->
* __Used Package__:              <!-- Debian or Ubuntu .deb | SUSE or RedHat .rpm | Docker - official Docker library | other -->

## Component, Query & Data

__Affected feature__:
<!-- e.g. Installation | Foxx | AQL query using web interface | arangosh | with driver | ... -->


__AQL query (if applicable)__:


__AQL explain (if applicable)__:
<!-- output of  db._explain("<my aql query>") -->


__Dataset__:
<!-- description, or if possible, please share an example dataset to reproduce the issue either as Gist with an arangodump, or an arangosh script with db.collection.save({my: "values"}) statements -->


__Size of your Dataset on disk__:
<!-- size of your dataset on disk-->


__Replication Factor & Number of Shards (Cluster only)__:
<!-- please list these settings for each collection in question -->


## Steps to reproduce

1. 
2. 
3. 

__Problem__:


__Expected result__: