A bit too late for this use case now, any way.
Most of the times it’s not required to stop Zammad during a rebuild at all. If you want to do that’s fine, you can start Zammad as soon as it did create the indexes.
After this output to be more precise:
root@zammad:~# zammad run rake searchindex:rebuild
delete pipeline (pipeline)... done
create pipeline (pipeline)... done
No sorry there is no such information.
In general reindexing will take longer the older and bigger your installation gets.
Most crucial factor in that moment is storage I/O.
Zammad does work during rebuild, the only thing that is not completely available is the search function / reports for the time being. This usually isn’t a too big issue for current tickets.
After about 2h the rebuild was done. I started the service about 5 minutes before the end with the assumption that simply the search would not work. But until I informed everyone that they can work carefully again, the rebuild was also finished