Frequently Asked Questions - ArchiHUB
Frequently Asked Questions about ArchiHUB
Section titled “Frequently Asked Questions about ArchiHUB”🖥️ Requirements and Compatibility
Section titled “🖥️ Requirements and Compatibility”What are the minimum requirements to use ArchiHUB?
Section titled “What are the minimum requirements to use ArchiHUB?”ArchiHUB requires:
- Docker and Docker Compose installed
- Stable internet connection
- Minimum recommended: 4 GB RAM and 2 CPU cores
It works perfectly on modest servers. As a reference, ArchiHUB runs smoothly on a Raspberry Pi 4.
Does ArchiHUB work on Windows, macOS, and Linux?
Section titled “Does ArchiHUB work on Windows, macOS, and Linux?”Yes. Being based on Docker, it is completely cross-platform (Windows, macOS, Linux, including ARM64 architectures like Raspberry Pi).
❗ Common Errors
Section titled “❗ Common Errors”I saved a resource but I can’t see it in the cataloging module
Section titled “I saved a resource but I can’t see it in the cataloging module”ArchiHUB allows you to save or publish resources. If you just saved a resource and don’t see it in the cataloging module, it’s likely saved as a draft. In the left sidebar, enable “View drafts”. There you will find all your drafts.
Keep in mind that a resource will only be visible to other users once you publish it. Drafts can only be seen by you and users with the super-editor role.
Files appear with the “Unprocessed” label
Section titled “Files appear with the “Unprocessed” label”When you upload a file to ArchiHUB, it is not processed automatically by default. To enable automatic processing, review the Getting Started section of the documentation.
If you uploaded files before enabling automatic processing, you must process them manually. In the resource detail view, and if your user has the required permissions, you will see an option in the sidebar to process the files.
I processed the files but they are still marked as “Unprocessed”
Section titled “I processed the files but they are still marked as “Unprocessed””ArchiHUB uses independent processing nodes separate from the main application to handle file processing. If you triggered processing and the file remains marked as “Unprocessed”, you should contact the system administrator to verify that these nodes are online and functioning correctly.
My user does not exist or I cannot log in
Section titled “My user does not exist or I cannot log in”You should contact the system administrator to verify that your user exists in the database. If your installation uses LDAP, ensure that your user is included in the correct group.
My resource doesn’t appear in the search
Section titled “My resource doesn’t appear in the search”If you made changes that are not showing up in the search results, there may be an issue with indexing.
- Verify that the index is correctly configured (see the section at the end of this document).
- If you modified the form, you need to regenerate the index so the search engine recognizes the updated data structure.
I can’t see a resource created by another user
Section titled “I can’t see a resource created by another user”If another user created a resource and you cannot see it, it may be due to one of the following reasons:
- The resource was saved as a draft. If you do not have the
super-editorrole, you won’t be able to view it. - You do not have the required access levels. Check whether the resource has an access level compatible with your user, or whether it has been published as a public resource.
🎮 GPU Usage for Advanced Tasks
Section titled “🎮 GPU Usage for Advanced Tasks”Does ArchiHUB support GPU usage?
Section titled “Does ArchiHUB support GPU usage?”Yes, ArchiHUB allows the use of GPU (NVIDIA compatible with CUDA) for intensive processing tasks, such as automatic audio transcription with models like Whisper. This significantly accelerates processes that would otherwise depend only on the CPU.
What are the requirements to use GPU in ArchiHUB?
Section titled “What are the requirements to use GPU in ArchiHUB?”- An NVIDIA GPU with CUDA support installed and accessible.
- Configure dedicated processing nodes (Celery workers) on machines with GPU.
- Environment variable
CELERY_WORKER(any value) to identify the node as a worker. - Recommended concurrency:
CELERYD_CONCURRENCY=1to avoid overload.
Note: If no GPU node is available for a task, it will pause until one becomes available.
How do I configure a node for GPU tasks?
Section titled “How do I configure a node for GPU tasks?”- On the machine with GPU, define
CUDA_VISIBLE_DEVICES=0(or0,1for multiple GPUs) in yourdocker-compose.ymlfile (available at this link in theCELERY QUEUE SERVICE USING GPUsection)
🤖 Running Local AI Models with Ollama and GPU
Section titled “🤖 Running Local AI Models with Ollama and GPU”What is Ollama and how does it integrate with ArchiHUB?
Section titled “What is Ollama and how does it integrate with ArchiHUB?”Ollama is an open-source tool that allows running large language models (LLMs) locally, ensuring privacy and reducing dependence on cloud services. In ArchiHUB, it integrates to power AI assistants with local models, such as document analysis, summary generation, or image/transcription processing.
What are the requirements to use Ollama with GPU?
Section titled “What are the requirements to use Ollama with GPU?”- NVIDIA GPU compatible with CUDA (for acceleration; without GPU, it works on CPU but slower).
- Docker and the Ollama container enabled in
docker-compose.yml. - Environment variables in
.env:OLLAMA_HOST=archihub_ollama,OLLAMA_PORT=11434,OLLAMA_PATH=/path/to/ollama/data. - Disk space for models (can be several GB per model).
How do I configure Ollama with GPU support?
Section titled “How do I configure Ollama with GPU support?”- In
docker-compose.yml, enable thearchihub_ollamaservice with imageollama/ollama:latest:
- Expose port:
${OLLAMA_PORT}:${OLLAMA_PORT}. - Volumes:
${OLLAMA_PATH}:/root/.ollamato store models. - For GPU: Add in
environment:CUDA_VISIBLE_DEVICES: 0, and indeploy.resources.reservations.devices:
driver: nvidiacount: 1capabilities: [gpu]- Start services:
docker compose up -d. - Install a model:
docker exec -it <ollama_container_name> ollama pull <model>(e.g.,llama2orllavafor vision).
Note: Ollama automatically detects the GPU if configured. For multiple GPUs, adjust CUDA_VISIBLE_DEVICES=0,1.
How do I use Ollama in ArchiHUB?
Section titled “How do I use Ollama in ArchiHUB?”- Once configured, go to the Assistants section in ArchiHUB.
- Select “Ollama” as the AI provider.
- Assign a name to the assistant and choose the installed model.
- The assistant will be able to use local resources such as documents, images, or transcriptions for tasks like text generation or analysis.
Examples: Analysis of video transcriptions or identification of elements in images (with models like Llava).
Are there any issues using Ollama with GPU?
Section titled “Are there any issues using Ollama with GPU?”- Verify that the container is running (
docker ps) and the port is accessible. - If GPU is not detected, confirm NVIDIA drivers and restart the container.
- Large models take time to download; use
ollama listto verify.
For complete details, see the Ollama documentation in ArchiHUB.
🚀 Installation and Deployment
Section titled “🚀 Installation and Deployment”Can ArchiHUB be deployed in the cloud?
Section titled “Can ArchiHUB be deployed in the cloud?”Yes. You can use any provider (AWS, Google Cloud, Azure, DigitalOcean, Hetzner, etc.).
You also have available a public demo instance:
➜ https://archihub.bit-sol.com.co
How do I update ArchiHUB to the latest version?
Section titled “How do I update ArchiHUB to the latest version?”- Backup your database and configurations
- Stop the containers (
docker compose down) - Update the code:
git pullin your local folder (front and back) - Rebuild and start:
docker compose up -d --build - Verify that everything works correctly
🔒 Privacy and Security
Section titled “🔒 Privacy and Security”Where is my data stored?
Section titled “Where is my data stored?”All your data (projects, BIM files, documents) is stored locally on your server. ArchiHUB does not send information to external servers. You have complete control.
Can I use ArchiHUB in production environments?
Section titled “Can I use ArchiHUB in production environments?”Yes, it is designed and tested for production. For critical environments, we recommend ArchiHUB Pro with priority support and updates tailored to your needs.
🧑💻 Open Source and Contribution
Section titled “🧑💻 Open Source and Contribution”Is ArchiHUB open source?
Section titled “Is ArchiHUB open source?”Yes, the backend is 100% open-source under MIT license.
Official repository: https://github.com/ArchiHUB-App
The frontend is free to use under the Creative Commons Attribution – NonCommercial – NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license. For more details regarding the frontend license, go here.
How can I contribute?
Section titled “How can I contribute?”- Reporting bugs or suggestions on GitHub Issues
- Improving documentation
- Developing new features or plugins
- Helping other users in the community
- Giving ★ to the repository and sharing it
📚 Documentation and Support
Section titled “📚 Documentation and Support”Where can I find official help?
Section titled “Where can I find official help?”- Main documentation → archihub-app.github.io
- Technical documentation generated with DeepWiki → deepwiki.com/Archihub-App/archihub-backend
- Issues and community support → GitHub
- Premium support (response < 24h) → ArchiHUB Pro users only
🔧 Customization and Integrations
Section titled “🔧 Customization and Integrations”Can I customize the interface or add functionalities?
Section titled “Can I customize the interface or add functionalities?”Yes. You can:
- Modify the frontend (it’s a separate and completely customizable project) You must comply with the frontend license agreements
- Create your own interface using the complete API
- Develop your own plugins
- Request custom development (contact at bit-sol.com.co)
Does ArchiHUB support integration with AI tools?
Section titled “Does ArchiHUB support integration with AI tools?”Yes. It currently allows connecting external AI services through plugins and APIs. We are working on more powerful native integrations (automatic memory generation, advanced pattern detection, etc.).
🔍 Search and Index (Elasticsearch)
Section titled “🔍 Search and Index (Elasticsearch)”How do I activate advanced search with indexing?
Section titled “How do I activate advanced search with indexing?”- Make sure the
elasticsearchservice is enabled in yourdocker-compose.yml - In Settings → activate “Use index for searches”
- Save and restart ArchiHUB
Why does the index option turn off by itself?
Section titled “Why does the index option turn off by itself?”This happens when Elasticsearch is not running or cannot connect. Verify:
- That the container is active (
docker ps) - That there are no errors in the Elasticsearch logs
How do I reindex all my content?
Section titled “How do I reindex all my content?”Go to Settings → Regenerate the index for resource search → Click the button.
The process may take time depending on the amount of files. It is only necessary to do this after activating it for the first time or after major content changes.
📂 ArchiHUB Data Folders
Section titled “📂 ArchiHUB Data Folders”What permissions should ArchiHUB data folders have?
Section titled “What permissions should ArchiHUB data folders have?”Data folders must have read and write permissions for the user running the Docker containers.
For example, if you’re using Linux and running Docker as user archihub, make sure the folders have appropriate permissions:
chown -R archihub:archihub /path/to/originalchmod -R 755 /path/to/originalSimilarly for the other ArchiHUB data folders:
temporaluserfileswebfilesWhat folders should be backed up?
Section titled “What folders should be backed up?”It is recommended to back up the following folders to secure your data and configurations:
original: Contains the original uploaded files.userfiles: Contains user-generated files, such as configurations and customized data.webfiles: Contains files related to the web interface and static resources.
Additionally, it is important to back up the database used by ArchiHUB to store critical information about projects, users, and configurations:
path/to/data/mongodb: Contains the MongoDB database data used by ArchiHUB.
If you have the Elasticsearch service enabled, it is also recommended to back up its data folder:
path/to/data/elasticsearch: Contains the search index data used by ArchiHUB.
💼 Plans and Pro Version
Section titled “💼 Plans and Pro Version”What are the advantages of ArchiHUB Pro?
Section titled “What are the advantages of ArchiHUB Pro?”- Priority technical support (response < 24h)
- Guaranteed updates and security patches
- Advanced features (in development)
- Custom feature development (optional)
More information: https://bit-sol.com.co/es/projects/archihub/
Do you have a question that doesn’t appear here?
Write it on GitHub Discussions or contact us directly!