A distributed network file system with Linux-like inode and block arrangement, RAID5-style parity distribution, data integrity through hashing, and fault tolerance with automated recovery.
- Distributed Storage: Data and parity blocks are distributed across multiple servers (RAID5 style).
- Linux-like Filesystem Structure: Inode and block-based file management.
- Data Integrity: Hashing for block-level integrity checks.
- Fault Tolerance: Parity allows recovery from single server failures.
- Automated Recovery: Failed servers can be rebuilt from parity data.
- Extensible CLI: Supports common file operations and advanced block management.
classDiagram
class Client {
+connect()
+executeCommand(cmd)
}
class Server {
+storeBlock()
+retrieveBlock()
+repair()
}
class Inode {
+inodeNumber
+blockPointers
}
class Block {
+blockID
+data
+checksum
}
Client "1" -- "many" Server : connects to
Server "1" -- "many" Block : manages
Inode "1" -- "many" Block : references
Client <.. Inode : interacts
- Python 3.x
- Install dependencies:
pip install -r requirements.txt
Start each server instance (use unique ports):
python3 server.py -nb 256 -bs 128 -port <server_port>(Repeat for each server—at least 4 recommended for RAID5 demonstration)
In a new terminal, after servers are running:
python3 client.py -port 8000 -cid 0 -startport <first_server_port> -ns <num_servers> -nb <total_data_blocks>Commands include:
- Basic:
cd,ls,mkdir,create,rm,cat,exit - Advanced:
lnh,lns,append,slice,mirror,showblock,showinode,repair,load,save
To recover a failed server, restart it and run:
# Use the repair command in the client CLI
repairNote: For best results, ensure all servers and clients are started as described, and refer to the CLI help for command details.