Excellent. I have the similar situation of handling about 80 above servers. You script of join sj and sh solve most of my delima of checking server by server and consolidate all into one table.

I made following changes:

1. declare @daydiff int           select @daydiff = 1 'in case I am only interested in last night job

 where cast(left(cast(run_date      as char(8)), 4)    as int)= datepart(yyyy, getdate())
   and cast(substring(cast(run_date as char(8)), 5, 2) as int)= datepart(month,getdate())
   and cast(right(cast(run_date     as char(8)), 2)    as int)= datepart(day,  getdate()-@dayDiff)
   and sh.step_name <> '(Job outcome)'

2. I filtered out the step_name = '(Job outcome)'

3. I union all servers

4. Finally, I create 2 cursor loops to find any existing databases on 80 servers either missing the job set up or with failure status. It works great in either case.

thanks

-D

One thing:

why the name of sj and sh data type are sysname? which causes centain problems in datalength() function...

-D