The normal way to download data is via the eRODat web interface. This should be used if analysing individual objects or a small to medium sample of objects.
However, if you would like to copy the entire archive or a subset of the file types, we provide a simple mirroring script. Please make sure if you are copying the entire archive that you are doing so in a way which does not overload our server and stop other users accessing it. This will be more important immediately after the data release. Please think of other users. If you want to download data later, this should be less of an issue.
The mirroring shell script can be accessed here.
It has been tested with the bash shell, but should run under POSIX-compatible shells.
It can be interrupted and restarted later if necessary.
The script requires the wget and lftp command line programs to be installed (installable via your Linux distribution package manager or MacOS Homebrew/MacPorts).
If you have a buggy lftp version and get a certificate verification error, you may need to uncomment the line set ssl:verify-certificate no in the script to disable secure certificate verification (be aware that this disables checking the authenticity of the archive server).
Please report problems using the script in the eROSITA forum.
The script can be modified to choose which subset of data to download. We list these options and the defaults below:
| Setting | Product size (GiB) |
GET_EVENT_FILES=y | 38 |
GET_IMAGES=n | 159 |
GET_EXPOSURE_MAPS=n | 503 |
GET_BACKGROUND_IMAGES=n | 294 |
GET_SENSITIVITY_MAPS=n | 15 |
GET_SURVEY_AREA_TABLES=n | 1 |
GET_SOURCE_PRODUCTS=n | 469 |
GET_UPPER_LIMITS=n | 1668 |
| Total | 3144 |
The default setting just copies the event files.
For many users this will be sufficient.
Change n to y in the script to also copy a particular product type.
The script can be rerun later to download just those files which are missing.
The output location of the mirrored archive can be changed in the output variable.