{question}
How do I troubleshoot slowness or other issues encountered when trying to RESTORE DATABASE from a Network File System (NFS) location?
{question}
{answer}
SingleStore supports using RESTORE DATABASE command to restore from a network drive, as explained in our documentation:
Restore from a Local or Network Drive
This functionality requires sufficient network throughput, and health - slowness and similar symptoms can indicate issues with the underlying network infrastructure.
Possible Indications: Networking/Infrastructure Issues
There might be inconsistent performance between leaf nodes within a cluster. One way to check for this is to review the tracelogs (memsql.log) of each leaf node and look for messages including Finished recovery for database. If most leaf nodes have a bunch of these messages but one or a few have less or none at all, this might indicate network connectivity issues involving those particular leaf nodes.
Additionally, logging might include messages like the following:
nfs: server [YOUR_NFS_MOUNT] not responding, still trying
These can be signs of issues with the underlying networking infrastructure, including the nodes in the cluster and the NFS location. These should be investigated further, including with logging and tools outside of SingleStore.
{answer}