Ok I followed the instructions and checking used_serial.db put out about 100 lines of
On tree page x cell y: Rowid z out of order
errors. I’ve recreated the database according to the instructions and reran the node. These must’ve been very old errors because there weren’t any abrupt shutdowns in a good while.
On that point, can you please confirm whether control + C in the console is the right way to shut down the FreeBSD node? I don’t know of any other way.
Yes, that’s the only way at the moment.
1 Like
I just experiencing the same issue and I am pretty sure that it is not a problem due to the disk. I stopped the node performed a thorough disk check and there were no errors.
The disk is attached to the system through a SATA port (the system is an ARM Odroid HC2) - so I would say is a pretty robust configuration and it was running for months withouth problems. The only relevant aspect is that the node already reached the max available storage capacity two months ago.
I feel that is a problem in the node program because the node was running from the time I switched from the beta to the latest tag - several weeks ago. Yesterday the node restarted (by itself, the system has an uptime of 30 days) and when it was up again that database disk image malformed error appeared. I am still waiting to have orders.db and used_serial.db rebuild process to be completed. It takes long time being those files are quite large (400MB the former and 115MB the latter).
Here is the the moment the error surfaced in the log:
This database could be malformed long before that version, it’s just checked on that time.