`Feature`: Validate if pinning map is already present to avoid rebalancing

## General VMs are migrated although the cpu_threshold is set to 25% (while each node is < 5% cpu). Furthermore balanciness is set to 30. ## Weighting 7.5 because it is about the core of ProxLB ## Config [proxlb.yaml](https://github.com/user-attachments/files/24523950/proxlb.yaml) ## Log [log.txt](https://github.com/user-attachments/files/24524069/log.txt) Important lines (filtered!): ``` /app/proxlb $ python3 main.py -c /etc/proxlb/proxlb.yaml | tee log.txt| grep -i -E "Guest balancing|INFO|threshold" 2026-01-09 09:34:03,428 - ProxLB - INFO - Using config path: /etc/proxlb/proxlb.yaml 2026-01-09 09:34:03,435 - ProxLB - INFO - API connection to host px103 succeeded. 2026-01-09 09:34:13,333 - ProxLB - INFO - Node px103 found in cluster. Applying pinning. 2026-01-09 09:34:13,333 - ProxLB - INFO - Node px102 found in cluster. Applying pinning. 2026-01-09 09:34:17,415 - ProxLB - INFO - Node px103 found in cluster. Applying pinning. 2026-01-09 09:34:17,415 - ProxLB - INFO - Node px102 found in cluster. Applying pinning. 2026-01-09 09:34:17,433 - ProxLB - DEBUG - Guest balancing is ok. Highest cpu usage node px102 is below the defined threshold of 25% with a value of 0.176842986449014%. 2026-01-09 09:34:17,433 - ProxLB - DEBUG - Guest balancing is ok. Highest value: 0.176842986449014, lowest value: 0.137112474095162 balanced by cpu and used. 2026-01-09 09:34:17,433 - ProxLB - DEBUG - Guest balancing is ok. Highest cpu usage node px102 is below the defined threshold of 25% with a value of 0.176842986449014%. 2026-01-09 09:34:17,433 - ProxLB - DEBUG - Guest balancing is ok. Highest value: 0.176842986449014, lowest value: 0.137112474095162 balanced by cpu and used. 2026-01-09 09:34:17,434 - ProxLB - DEBUG - Guest balancing is ok. Highest cpu usage node px102 is below the defined threshold of 25% with a value of 0.16758683150754564%. 2026-01-09 09:34:17,434 - ProxLB - DEBUG - Guest balancing is ok. Highest value: 0.16758683150754564, lowest value: 0.140204028250065 balanced by cpu and used. 2026-01-09 09:34:17,435 - ProxLB - DEBUG - Guest balancing is ok. Highest cpu usage node px102 is below the defined threshold of 25% with a value of 0.15808416992559013%. 2026-01-09 09:34:17,435 - ProxLB - DEBUG - Guest balancing is ok. Highest value: 0.15808416992559013, lowest value: 0.140204028250065 balanced by cpu and used. 2026-01-09 09:34:17,435 - ProxLB - DEBUG - Guest balancing is ok. Highest cpu usage node px102 is below the defined threshold of 25% with a value of 0.15808416992559013%. 2026-01-09 09:34:17,435 - ProxLB - DEBUG - Guest balancing is ok. Highest value: 0.15808416992559013, lowest value: 0.140204028250065 balanced by cpu and used. 2026-01-09 09:34:17,437 - ProxLB - INFO - Balancing: Starting to migrate VM guest test103 from px102 to px101. 2026-01-09 09:34:17,502 - ProxLB - INFO - Balancing: Starting to migrate VM guest pdm102 from px102 to px103. 2026-01-09 09:34:17,573 - ProxLB - INFO - Balancing: Starting to migrate VM guest test100 from px102 to px101. 2026-01-09 09:34:42,043 - ProxLB - INFO - Daemon mode active: Next run in: 5 minutes. ``` ## Meta The setup I try to configure is supposed to balance by **used cpu**, with some pinning. The pinning works :-), but as far as I can see, ProxLB isn't honoring the settings of the balancing itself. The pinning is configured by defining 'poolx': <img width="438" height="176" alt="Image" src="https://github.com/user-attachments/assets/caeb5a80-18de-4ec1-9425-0c70233d595f" /> (BTW, I'm not trying to annoy you by opening as many cases as I possibly can, but I think this extension/program/addon is a really nice thing and it's almost there where it should be (imho), so I want it to work and thus I'm willing to spend time creating cases and following up the suggestions). Version: feature/414-enfoced-pinning Installed from: Github/gyptazy/ProxLB Running as: Container, local on Proxmox.