Troubleshooting
Checkpoint Errors Troubleshooting
1. Check the log files for any checkpoint errors.
• archiver failure: iiacp.log in $II_SYSTEM/ingres/files
• ckpdb/rollforwarddb failure: rerun with '#x' to generate a ckpdb.log or rfpdb.log file.
a) Check if the rclone command is generating errors. The rclone command will be displayed above the errors.
b) Run the same command manually (drop the --quiet from the line) and check the errors. A verbose log can be generated by adding '-vv' to the rclone command line.
c) For any authorization errors, verify your access keys.
d) For any other status errors, verify the status of your bucket with your s3 provider.
2. If ckpdb is not working on Windows, run the following command:
jsutil ckpdb <database>
If this works and ckpdb doesn't work, then your service has the wrong variables set in regedit or you must restart the service again.
3. If the archiver is failing with rclone errors, and rclone is working in your command prompt:
• Identify the database where the errors are generated in II_SYSTEM/ingres/files/iiacp.log, and run infodb on the database.
• If the database is using the archiver's expected bucket, try restarting the archiver (assuming you have the correct RCLONE environment variables currently set).
• If the database is NOT using the appropriate bucket, you should disable journaling on that database with 'alterdb <db> -disable_journaling' and restart the archiver.
• If the archiver is having errors processing journals for a database (which causes the archiver to shut down), as a last resort you can disable journaling for the database, restart the archiver, and then re-enable journals at the next checkpoint.
Note: The archiver can only access one specific cloud bucket while it is running. It is not possible to have separate databases using different cloud buckets in the same installation.
Last modified date: 12/19/2024