Tested Environment: macOS Sequoia (Apple M4/M2)

Install Ollama

Install Ollama using Homebrew:

brew install ollama

Once the installation is complete, verify it by running:

ollama version

Manage the Ollama service using Homebrew

If the ollama command is not recognized, you may need to manually start the service.

Check active Homebrew services:

brew services list

Start the Ollama service:

brew services start ollama

Stop the Ollama service:

brew services stop ollama

Ollama commands

Commonly used Ollama commands, using the DeepSeek R1 (7B) model as an example.

Pull the model:

ollama pull deepseek-r1:7b

Run the model (automatically pulls if not already downloaded):

ollama run deepseek-r1:7b

Stop the model:

ollama stop deepseek-r1:7b

Delete the model:

ollama rm deepseek-r1:7b

References

Ollama on Homebrew

Ollama models

Tested Environment: macOS Sequoia (Apple M4/M2)

Install Open WebUI using Docker

This is an easier way since I don’t want to mess up my Python environment.

If Ollama is on your computer:

docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

If Ollama is on a Different Server (change the OLLAMA_BASE_URL to the server’s URL):

docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=https://example.com -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

With Nvidia GPU support:

docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda

Make sure to include the “-v open-webui:/app/backend/data” in the command to ensure your database is properly mounted.

Install Open WebUI with Bundled Ollama

With Nvidia GPU support:

docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama

For CPU Only:

docker run -d -p 3000:8080 -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama

References

Open Web UI

Install SSH server on Ubuntu:

sudo apt install openssh-server

Passwordless Login

Create SSH key on local machine (located at ~/.ssh by default):

ssh-keygen -t rsa -b 4096

Copy public key to server:

ssh-copy-id username@ip_address

Test:

ssh username@ip_address

References

https://linuxize.com/post/how-to-setup-passwordless-ssh-login/

Test environment: Ubuntu 20.04 LTS (Focal Fossa) 64-bit

Generate SSL Certificates

Navigate to the target directory, for example /root/sys/ssl, and generate private key:

openssl req -new -newkey rsa:2048 -nodes -out nassgeodata_gmu_edu.crt -keyout nassgeodata_gmu_edu.key -subj "/C=US/ST=Virginia/L=Fairfax/O=George Mason University/OU=CSISS/CN=nassgeodata.gmu.edu"

After requested the certificate, for example /root/sys/ssl/sslcerts/x509CO/nassgeo_csiss_gmu_edu_cert.cer, create symbolic link of the certificate file:

ln -s /root/sys/ssl/sslcerts/x509CO/nassgeo_csiss_gmu_edu_cert.cer /root/sys/ssl/nassgeodata_gmu_edu.cer 

Enable SSL

Configure Apache Web Server

Create symbolic link of SSL config file:

ln -s /etc/apache2/sites-available/default-ssl.conf /etc/apache2/sites-enabled/000-default-ssl.conf

Edit /etc/apache2/sites-enabled/000-default-ssl.conf and add:

SSLCertificateFile      /root/sys/ssl/nassgeodata_gmu_edu.cer
SSLCertificateKeyFile   /root/sys/ssl/nassgeodata_gmu_edu.key

Enable rewrite mod and restart Apache:

sudo a2enmod ssl
service apache2 restart

Redirect HTTP to HTTPS (Optional)

Edit /etc/apache2/sites-enabled/000-default.conf and add:

RewriteEngine On
RewriteCond %{HTTPS} !=on
RewriteRule ^/?(.*) https://%{SERVER_NAME}/$1 [R,L]

Enable rewrite mod and restart Apache:

sudo a2enmod rewrite
service apache2 restart

References

https://hallard.me/enable-ssl-for-apache-server-in-5-minutes/

https://www.tecmint.com/redirect-http-to-https-on-apache/

Test environment: Ubuntu 20.04 LTS (Focal Fossa) 64-bit

Installation

The current version of GDAL package can be checked using apt-cache policy gdal-bin:

gdal-bin:
  Installed: (none)
  Candidate: 3.0.4+dfsg-1build3
  Version table:
     3.0.4+dfsg-1build3 500
        500 http://us.archive.ubuntu.com/ubuntu focal/universe amd64 Packages

To update the GDAL package to the latest version, we can add the ubuntugis-unstable PPA to system:

sudo add-apt-repository ppa:ubuntugis/ubuntugis-unstable
sudo apt update

The source list is located at: /etc/apt/sources.list.d, the candidate of GDAL package should be updated to the latest version:

gdal-bin:
  Installed: (none)
  Candidate: 3.2.1+dfsg-1~focal0
  Version table:
     3.2.1+dfsg-1~focal0 500
        500 http://ppa.launchpad.net/ubuntugis/ubuntugis-unstable/ubuntu focal/main amd64 Packages
     3.0.4+dfsg-1build3 500
        500 http://us.archive.ubuntu.com/ubuntu focal/universe amd64 Packages

Install the latest GDAL package:

apt install gdal-bin

Check the version of installed package using gdalinfo --version and :

GDAL 3.2.1, released 2020/12/29

References

https://launchpad.net/~ubuntugis/+archive/ubuntu/ubuntugis-unstable

Installation

Create conda environment for PyWPS:

conda create -n "pywps" python=3.8

Install PyWPS and dependencies:

#Latest PyWPS
conda install -c conda-forge pywps

#Optional packages
conda install numpy
conda install gdal
conda install -c conda-forge pillow

Install and enable Apache WSGI for Python 3:

sudo apt-get install libapache2-mod-wsgi-py3
sudo a2enmod wsgi

Configuration

Create WPS Directory /usr/local/pywps and create pywps.wsgi:

#!/usr/bin/env python3

import sys
sys.path.append("/usr/local/anaconda3/envs/pywps/lib/python3.8/site-packages/")
sys.path.append("/usr/local/pywps/processes")

from pywps.app.Service import Service

# processes need to be installed in PYTHON_PATH
from processes.sleep import Sleep
from processes.ultimate_question import UltimateQuestion
from processes.centroids import Centroids
from processes.sayhello import SayHello
from processes.feature_count import FeatureCount
from processes.buffer import Buffer
from processes.area import Area

processes = [
    FeatureCount(),
    SayHello(),
    Centroids(),
    UltimateQuestion(),
    Sleep(),
    Buffer(),
    Area()
]

# Service accepts two parameters:
# 1 - list of process instances
# 2 - list of configuration files
application = Service(
    processes,
    ['/usr/local/pywps/pywps.cfg']

Create pywps.cfg:

According to https://pywps.readthedocs.io/en/master/configuration.html:

Create processes directory and logs directory:

mkdir processes
mkdir workdir
mkdir logs
chmod 777 logs
chmod 777 workdir
touch logs/pywps.log
chmod 777 logs/pywps.log

Create a configuration file /etc/apache2/sites-enabled/pywps.conf:

# PyWPS
WSGIDaemonProcess pywps home=/usr/local/pywps user=www-data group=www-data processes=2 threads=5
WSGIScriptAlias /pywps /usr/local/pywps/pywps.wsgi process-group=pywps
<Directory /usr/local/pywps/>
WSGIScriptReloading On
WSGIProcessGroup pywps
WSGIApplicationGroup %{GLOBAL}
Require all granted
</Directory>

Test

http://localhost/pywps?request=GetCapabilities&service=WPS

Issues

No module named gdal’’

Solution: install numpy may fix the gdal module issue.

No module named pywps.app.Service

This issue happens when the Python runtime is not being targeted to the conda environment. The message would be like:

[Wed Mar 17 11:35:24.851659 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116] mod_wsgi (pid=79590): Failed to exec Python script file '/usr/local/pywps/pywps.wsgi'.
[Wed Mar 17 11:35:24.851713 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116] mod_wsgi (pid=79590): Exception occurred processing WSGI script '/usr/local/pywps/pywps.wsgi'.
[Wed Mar 17 11:35:24.851782 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116] Traceback (most recent call last):
[Wed Mar 17 11:35:24.851827 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116]   File "/usr/local/pywps/pywps.wsgi", line 3, in <module>
[Wed Mar 17 11:35:24.851896 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116]     from pywps.app.Service import Service
[Wed Mar 17 11:35:24.851942 2021] [wsgi:error] [pid 79590:tid 139628017260288] [remote 10.28.140.89:50116] ImportError: No module named pywps.app.Service

Solution: add sys.path.append("/usr/local/anaconda3/envs/pywps/lib/python3.8/site-packages/" in pywps.wsgi.

Reference

https://github.com/geopython/pywps/

https://pywps.readthedocs.io/en/master/deployment.html

https://faun.pub/how-to-set-up-conda-virtual-environments-with-apache-mod-wsgi-flask-c2043711223e

https://stackoverflow.com/questions/45289299/python-3-4-importerror-no-module-named-gdal-arrayno-module-named-gdal-arra

Installation

GDAL

Import the minted package:

sudo add-apt-repository ppa:ubuntugis/ppa
sudo apt-get update
sudo apt-get install gdal-bin

Node

curl -sL https://deb.nodesource.com/setup_10.x | sudo -E bash -
sudo apt-get install -y nodejs
node --version
npm --version

Gulp

npm run gulp:

npm install --save-dev -g gulp

TerriaJS

Install TerriaJS:

git clone https://github.com/TerriaJS/TerriaJS.git
npm install

Install TerriaJS Map:

git clone https://github.com/TerriaJS/TerriaMap.git
npm install

Build:

npm run gulp release

Start:

npm start

Deploy and hide the port:

ProxyPass /terria/ http://localhost:3001/
ProxyPassReverse /terria/ http://localhost:3001/

Q&A

Github connection error while installing with npm:

npm ERR! git clone --template=/root/.npm/_git-remotes/_templates --mirror git://github.com/TerriaJS/pell.git /root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6: Cloning into bare repository '/root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6'...
npm ERR! git clone --template=/root/.npm/_git-remotes/_templates --mirror git://github.com/TerriaJS/pell.git /root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6: fatal: unable to connect to github.com:
npm ERR! git clone --template=/root/.npm/_git-remotes/_templates --mirror git://github.com/TerriaJS/pell.git /root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6: github.com[0: 140.82.113.4]: errno=Connection timed out
npm ERR! code 128
npm ERR! Command failed: git clone --template=/root/.npm/_git-remotes/_templates --mirror git://github.com/TerriaJS/pell.git /root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6
npm ERR! Cloning into bare repository '/root/.npm/_git-remotes/git-github-com-TerriaJS-pell-git-master-d15617c6'...
npm ERR! fatal: unable to connect to github.com:
npm ERR! github.com[0: 140.82.113.4]: errno=Connection timed out
npm ERR!
npm ERR!

npm ERR! A complete log of this run can be found in:
npm ERR!     /root/.npm/_logs/2020-06-11T03_58_52_831Z-debug.log

Solution:

git config --global url.https://github.com/.insteadOf git://github.com/

References

TerriaJS: Getting Started

Apache VirtualHost Proxy with a Subdirectory

NodeSource Node.js Binary Distributions

Fatal: Unable to connect to github.com #6285

Installation & Configuration

Note: This guide is based on MacTex.

Import package

Import the minted package:

\usepackage[cache=false,outputdir=build]{minted}

Add –shell-escape option

Make sure you are adding the --shell-escape option while building.

For example, if you are using the VS Code with latex-workshop, you have to add the --shell-escape option in the setting:

"latex-workshop.latex.tools": [
    {
        "args": [
            "--output-directory=build",
            "-interaction=nonstopmode",
            "-synctex=1",
            "--shell-escape",
            "%DOC%"
        ],
        "command": "pdflatex",
        "name": "pdflatex"
    }
    ...
]

Install pygments

If you see the error “You must have `pygmentize’ installed to use this package”. You have to install the pygments.

For mac user:

brew install pygments

References

LaTeXTools can’t find installed pygmentize package

如何在mac的latex上使用minted宏包？

umask

Check the umask value for the current directory:

umask

Change the default umask value for the current directory, for example:

umask u=,g=w,o=rwx

setfacl

setfacl -d -m g::rwx .

References

Set the default permissions for newly created files

Default File Permissions (umask)

How to set umask for a specific folder

Preparation

Check current database location, usually located at /var/lib/postgresql by default:

sudo -u postgres psql
SHOW data_directory;
\q

Stop CKAN and postgres service:

service apache2 stop
service nginx stop
service postgresql stop

Copy data from original location to target location:

cp -R /var/lib/postgresql NEW_LOCATION

Change permission:

chmod 700 -R NEW_LOCATION
chown -R postgres:postgres NEW_LOCATION

Target Database to New Location

Solution 1: Update Database Configuration

Edit /etc/postgresql/9.5/main/postgresql.conf and update the new location:

data_directory = '[NEW_LOCATION]'

Start the postgres service:

service postgresql start

Check the log of postgres to make sure it is successfully started:

tail -f /var/log/postgresql/postgresql-9.5-main.log

Remove the old database if needed:

rm -rf /var/lib/postgresql

Solution 2: Link to the new location

I tried to soft link the default directory to the new location but failed for some reason, might be caused by the permission issue:

ln -s NEW_LOCATION /var/lib/postgresql

It should work with the correct permission configuration but need to be further tested.

Automount Disk (for external disk only)

Get the UUID of the external disk:

blkid

Copy the UUID of the target disk, edit /etc/fstab and add the following line:

UUID=[UUID_OF_TARGET_DISK] [MOUNT_POINT] ext4  defaults       0  0

Q&A

I ran into this issue while checking the new database location with sudo -u postgres psql:

psql: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket

or

psql: could not connect to server: No such file or directory

Is the server running locally and accepting connections on 
Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?

A potential solution is removing postmaster.pid in the postgres database directory. I am not sure how did I solved this issue, maybe by removing the postmaster.pid or updating the permission.

According to https://stackoverflow.com/questions/42653690/psql-could-not-connect-to-server-no-such-file-or-directory-5432-error, the permission error would lead this error:

root@uddi:/etc/postgresql/9.5/main# pg_ctlcluster 9.5 main start
Error: Config owner (postgres:123) and data owner (root:0) do not match, and config owner is not root

Please make sure the owner of data directory and its subdirectories is postgres.

References

How To Move a PostgreSQL Data Directory to a New Location on Ubuntu 16.04

Editing fstab to automount partitions at startup