[overpass] apply_osc_to_db not handling dispatcher problems

[Igor Brejc <>]

Hi,

I notice that if the main dispatcher stops working for some reason (file locks etc.), the apply_osc_to_db.sh script does not really detect that, instead its log reports new diffs were applied in a rapid fashion without reporting any errors (cca. 1 second for a diff, while it usually takes around 15 seconds). 

After that, the DB gets inconsistent (missing nodes, ways etc.) and it's basically useless and needs to be recloned. 

Igor