wekan/docs/Platforms/FOSS/Sandstorm/Export-from-Wekan-Sandstorm-grain-.zip-file.md

TODO: [Feature Request: command line utility to convert between Standalone Wekan MongoDB database, and Sandstorm grain .zip file](https://github.com/wekan/wekan/issues/1695)

This is useful for example if you get [Board not found error](https://github.com/wekan/wekan/issues/1430)

Using Ubuntu 18.04 or Mint 19.1 or Ubuntu for Windows Subsystem for Linux:

## 1) Install required packages:

```
sudo apt-get install zip unzip p7zip-full wget
```

## 2) If you don't have Meteor 1.6.0.1 installed, download it to your home directory, and unarchive it

```
cd
wget https://releases.wekan.team/dev/repair/meteor-repair.7z
7z x meteor-repair.7z
```

## 3) Install MongoDB 3.2.22

https://docs.mongodb.com/v3.2/tutorial/install-mongodb-on-ubuntu/

## 4) Download Wekan grain

Use Sandstorm arrow down button to download Wekan grain in .zip file to your home directory.

## 5) Unzip downloaded file

```
unzip "YOUR-GRAIN-NAME-HERE.zip"
```

## 6) Repair database, if it does not open

```
.meteor/packages/meteor-tool/1.6.0_1/mt-os.linux.x86_64/dev_bundle/mongodb/bin/mongod --dbpath "./YOUR-GRAIN-NAME-HERE/data/wiredTigerDb" --repair
```

## 7) (Note to self about future versions of Meteor and Wekan)

Currently Wekan uses Meteor 1.6.0.1. In .meteor/packages/meteor-tool/ directory is also some 1.8.x version for upcoming Wekan version using Meteor 1.8.x. This .meteor directory has been generated by installing meteor from www.meteor.com and then upgrading some existing or new app with for example `meteor update --release 1.6.0.1` so that correct version of Meteor is downloaded.

## 8) Stop MongoDB

```
sudo systemctl stop mongod
```

## 9) Copy repaired database to MongoDB raw files and start MongoDB

```
sudo su
cd /var/lib
mv mongodb mongodb-original
cp -pR "/home/YOUR-USERNAME/YOUR-GRAIN-NAME-HERE/data/wiredTigerDb" .
mv wiredTigerDb mongodb
chown mongodb:mongodb mongodb -R
systemctl start mongod
exit
```

## 10) Browse database with MongoDB CLI

```
mongo
show dbs
use meteor
show collections
db.users.find();
exit
```

## 11a) Dump database to MongoDB dump format

```
mongodump --db meteor
```

## 11b) Dump database to JSON text files

Save this to `dump-json.sh` and then `chmod +x dump-json.sh && ./dump.sh meteor filesdir`

Script is originally [from this Rocket.Chat issue](https://forums.rocket.chat/t/big-issue-with-custom-javascript/261/4)
```
#!/bin/bash

if [ ! $1 ]; then
        echo " Example of use: $0 database_name [dir_to_store]"
        exit 1
fi
db=$1
out_dir=$2
if [ ! $out_dir ]; then
        out_dir="./"
else
        mkdir -p $out_dir
fi

tmp_file="fadlfhsdofheinwvw.js"
echo "print('_ ' + db.getCollectionNames())" > $tmp_file
cols=`mongo $db $tmp_file | grep '_' | awk '{print $2}' | tr ',' ' '`
for c in $cols
do
    mongoexport -d $db -c $c -o "$out_dir/exp_${db}_${c}.json"
done
rm $tmp_file
```
It's also possible to restore JSON files like described [at stackoverflow](https://stackoverflow.com/questions/11255630/how-to-export-all-collection-in-mongodb) this way, but because userIDs etc can be different, there is no mapping to other users, so you may need to change JSON files some way:
```
for file in *.json; do c=${file#*exp_yourdbname_}; c=${c%.json}; mongoimport --db yourdbname --collection "${c}" --file "${file}"; done
```

## 12) Save attachments to files

[Download NoSQLBooster GUI webpage](https://nosqlbooster.com/downloads)

If you download AppImage for Linux, run it with:
```
chmod +x nosqlbooster4mongo*.AppImage
./nosqlbooster... (full filename here)
```
It can install NoSQLbooster to Linux Menu/Programming/NoSQLBooster for MongoDB.

With it connect to localhost:27017

Double click `cfs_gridfs.attachments.files` and right click filename and Download file.

<img src="https://wekan.fi/nosqlbooster.png" width="100%" alt="Wekan logo" />

## 13) Optional: Restore

a) Restore data to Standalone Wekan Snap

b) Figure out how to make Sandstorm grain .zip file

***

## OLD CONTENT, TO BE CLEANED UP SOMETIME

## 5) Copy database files from unzipped Wekan grain to MongoDB

In .zip file database files are in directory `data/wiredTigerDb/`

In Wekan Snap, database files are in directory `/var/snap/wekan/common` - if you have other data there, rename common directory to other name first. 

[Snap Backup and Restore](https://github.com/wekan/wekan-snap/wiki/Backup-and-restore)

[Docker Backup and Restore](Export-Docker-Mongo-Data)

## 6) Change database file permissions to root user

In Snap:
```
sudo chown root:root /var/snap/wekan/common -R
```

## 7) Start MongoDB

In Snap:
```
sudo snap start wekan.mongodb
```

## 8) Or, if you are brave, also start standalone wekan

Try does Standalone Wekan work with that database
```
sudo snap start wekan.mongodb
sudo snap start wekan.wekan
```
or both wekan and database at once
```
sudo snap start wekan
```

## 9) To see MongoDB data, install Robo3T GUI

https://robomongo.org

Wekan Snap: Connect it to address `localhost:27019`

Wekan Docker: If you have MongoDB exposed to outside Docker, Connect with Robo3T to address `localhost:27017`. Otherwise you first need to copy files from wekan-db Docker container to outside of Docker container, and restore files to local separately installed MongoDB database v3.2.20 or newer

[Snap Backup and Restore](https://github.com/wekan/wekan-snap/wiki/Backup-and-restore)

[Docker Backup and Restore](Export-Docker-Mongo-Data)

## 10) Browse data in Robo 3T

- On left, double click cards, checklists etc collections/tables to see their contents
- Currently default view to see is tree view. To see table view or JSON view, click small buttons above that white data area view display at near right border.

## 11) If you don't find data in Robo 3T, use hex editor GUI

In Linux, you can install Hex Editor for example with command:
```
sudo apt install ghex
```
or
```
sudo yum install ghex
```
Then it's usually in Linux desktop at Menu / Development / GHex.
Or you can start it in command line by writing `ghex`

You can open files from your unzipped Wekan Sandstorm grain directory `data/wiredTigerDb/` to see if there is still some data that is not yet overwritten with other data.

## 12) Moving data from Sandstorm to Standalone Wekan like Snap or Docker

12.1 Compare databases of Sandstorm and Docker Wekan with mongo GUI https://nosqlbooster.com/

12.2 Create new user to Standalone Wekan

12.3 Copy new user bcrypt password structure created bcrypt encrypted password from Sandstorm to Docker wekan structure. It's at users collection/table. Also add isAdmin=true boolean.

12.4 Login

12.5 Here is info about script how to save all from MongoDB to .JSON textfiles https://forums.rocket.chat/t/big-issue-with-custom-javascript/261/4
and most likely how to get them back to mongodb
so just copying all of those boards JSON files and restoring to MongoDB database would work
and editing JSON files with any plain text editor. For big files you can try JEdit.

12.6 Here is how to backup and restore mongodb in docker, snap etc https://github.com/wekan/wekan/wiki/Backup

12.7 All Wekan Docker settings are in this textfile https://raw.githubusercontent.com/wekan/wekan/devel/docker-compose-build.yml

12.8 Database name in Sandstorm is meteor, and in docker is wekan, so here is how to restore with different database name https://stackoverflow.com/questions/36321899/mongorestore-to-a-different-database

## Azure links

https://docs.microsoft.com/en-us/azure/container-service/dcos-swarm/container-service-swarm-mode-walkthrough

https://docs.docker.com/docker-cloud/cloud-swarm/create-cloud-swarm-azure/

https://docs.atlas.mongodb.com/reference/microsoft-azure/