Could not get database status

Troubleshooting
13

Database backups would have been implemented by whoever managed or manages this CryoSPARC instance. It is possible that there are no backups.
Under the assumption there are no backups, I recommend (run commands on the CryoSPARC master node under the Linux account that “owns” your CryoSPARC installation):

  1. Shut down CryoSPARC
    cryosparcm stop

  2. Ensure there are no remaining CryoSPARC processes:
    ps x | grep -e cryosparc -e mongo

  3. Assuming the existing database path /data/cryo/cryosparc_database, set aside the existing database directory:

    mv /data/cryo/cryosparc_database /data/cryo/cryosparc_database.20230607

  4. Start CryoSPARC
    cryosparcm start
    This command should create a blank database. You will have to:

  • create CryoSPARC login(s) (how?).
  • connect worker(s). Is this a “standalone” (workstation-type) CryoSPARC instance? What is the path to the particle scratch space, if there is any?
  1. Configure database backups.
  2. Regularly monitor available space for the database.
  3. Attach project directories. Caution: Deletion of cs.lock files from CryoSPARC project directories is strongly discouraged under most circumstances. This particular recovery scenario is an exception. During the creation of the new database, a new instance ID was created, which may be incompatible with the cs.lock that were compatible with the old, presumably corrupted database. Before attaching each project directory to the CryoSPARC instance with an new database, delete the cs.lock file in the project directory. I repeat: cs.lock files should only be deleted under exceptional circumstances.