Learning-based fault localization has achieved promising results. However, as software and tests are constantly evolving, models trained on old data become ineffective on new data. Particularly, in the context of system testing for large-scale software, each iteration generates a large volume of new data. This makes retraining the model from scratch incur an unacceptable time overhead, while merely fine-tuning on new data leads to catastrophic forgetting. Continual learning offers an effective method for models to avoid catastrophic forgetting during this iterative process. However, existing continual learning methods are not specifically designed for fault localization or for large-scale software system testing scenarios, which leads to their direct application yielding sub-optimal effectiveness. In response, we propose Canto, a novel continual learning framework specifically designed for large-scale software fault localization. Canto first extracts fine-grained program semantics from logs, then utilizes fault characteristics to enhance the weights of certain semantics. Finally, Canto uses an unsupervised algorithm to obtain corresponding embeddings and selects representative exemplars based on clustering. Subsequently, Canto mixes the representative exemplars with new samples for training and adjusts the loss weight according to the model’s degree of mastery over the sample. This allows the model to focus more on samples that are not yet well-mastered during the training process, thereby enabling it to learn new faults while mitigating the forgetting of old ones. In extensive evaluations against 6 continual learning baselines, Canto demonstrates superior performance, improving overall effectiveness by 17.30% to 45.23%.