Shinobi inaccessible when cron.js is restarted
When cron.js is restarted I am unable to login to the Shinobi interface, the login just times out. top shows msqld as the process utilising the most CPU.
I have a 6 month retention and whilst the videos are deleted the database entries are not, until cron.js is restarted. The database entry pertaining to those records do not serve any purpose if the videos are deleted and thus I have no reason to keep them.
The database entries are only purged when cron.js is restarted and camera.js is running. This has caused issues when upgrading from one version to the next as cron.js is restarted, as per the recommendation, which starts pruning the database entries thus incorrectly creating a perceived error that the upgrade has caused an issue.
Running 'show processlist' within mysql shows the following:
When initating the cron.js restart, the following log lines are appended:
If cron.js hasn't been restarted in a while the found records could easily be into the thousands or tens of thousands.
Do I require to set up any further configuration with cron.js to manage the deletion of these or should cron.js be handling this automatically?
Aside from being unable to login to Shinobi whilst this prune happens, without restarting cron.js so as to initate the prune, the space utilised bar is incorrect as it is displaying used space for videos that no longer exist but are still within the database.