Instance segmentation on 3D point clouds has been attracting increasing attention due to its wide applications, especially in scene understanding areas. However, most existing methods require training data to be fully annotated. Manually preparing ground-truth labels at point-level is very cumbersome and labor-intensive. To address this issue, we propose a novel weakly supervised method RWSeg that only requires labeling one object with one point. With these sparse weak labels, we introduce a unified framework with two branches to propagate semantic and instance information respectively to unknown regions, using self-attention and random walk. Furthermore, we propose a Cross-graph Competing Random Walks (CGCRW) algorithm which encourages competition among different instance graphs to resolve ambiguities in closely placed objects and improve the performance on instance assignment. RWSeg can generate qualitative instance-level pseudo labels. Experimental results on ScanNet-v2 and S3DIS datasets show that our approach achieves comparable performance with fully-supervised methods and outperforms previous weakly-supervised methods by large margins. This is the first work that bridges the gap between weak and full supervision in the area.