UPDATE:
I opened phpmyadmin and see that in the websites where option “Restart on every main step” may be deactivated, the wp_options table has backwpup_cfg_jobsteprestart=''
whereas the websites that are refusing to untick this option DONT HAVE that value present.
Looks like some weird condition occuring in the installation of the plugin. I use to manage my WP sites with InfiniteWP remote management plugin, and those sites where I get this option to work ok were those where I personally upgraded the plugin (the multisite and one individual website).
In the other sites, I get this to working state using “Reset all settings to default” in order to make the plugin rewrite all the options in the wp_options table.
So, check the installation procedure, since there are glitches when it create the options in the db.
BTW, by doing this option reset and reconfigure the whole plugin, the websites where the backup was working ok, showed that the elapsed time to complete a backup job was reduced from 380 seconds to just 11!
You can disabel the option if WP_ALTERNATE_CRON active than it will not usesd because it can’t work.
The option is enabled by defauld, because there are to many hosters that not work without the option.
Nope, I don’t use WP_ALTERNATE_CRON anymore, because your previous version was colliding with that option and then I double checked that my sites not use this option anymore.
But please, take a minute and read the last paragraph of my second reply in this thread. The glitch is in how the plugin record its configuration data in wp_options: it’s forgetting this variable, backwpup_cfg_jobsteprestart.
I am also experiencing all of the above issues:
(1) Plugin stalls upon Run with:
ERROR: ZipArchive returns status: (ER_DELETED) Entry has been deleted
(2) Cannot uncheck “Restart the job on every main step on a working job “
Plugin failure.
I still get the checkmark returning after a full delete and fresh install of backwpup
Restart the job on every main step on a working job
Did you try resetting the options in the configuration page?
It solved the issue to me.
Anyway, this is an actual bug that Daniel has to fix.