Open tcrst opened 4 days ago
This problem usually occurs when the main indexer or ord fails during indexing somehow and fills the log_file.txt over and over again with same block commands. Probably the size of log_file.txt has grown a lot on your node and currently main_index.js needs to read all of that file to memory in order to work.
Even though it is possible to fix the file, we currently do not have a fix script. You can remove the repeating blocks from the file by hand, or you can restart the indexing but set the memory flag to something around or bigger than 16GB.
Finally, I suggest using the restore.py script, it'll download the latest snapshot from S3 and start from there.
The main_index process is running out of heap memory even after increasing `--max-old-space-size to 64GB; what is the recommended memory allocation and any further optimization suggestions?
Also getting this in the logs