Your code would fail in almost any database, because of the use of non-aggregated functions in the having clause.

In general, if you want to filter data, you should do so before aggregation. In SQL Server, I would suggest:

select max(me_date)
from table1
where me_date < dateadd(day, (1 - day(getdate())), convert(date, getdate())) and
      me_date >= dateadd(month, -1, dateadd(day, (1 - day(getdate())), convert(date, getdate())));

The above is index and optimizer friendly. If you don't care about that, then a simpler formulation is:

where datediff(month, me_date, getdate()) = 1

SELECT EOMONTH(GETDATE()-1) AS 'Last Month';

Click here for more info.

We can't use having without group by clause, hence the error.

I tested The following query; it works fine in MySQL

Select distinct me_date from <table_name> where month(me_date) not in (select max(month(me_date)) from <table_name> ) order by me_date describe limit 1;

Alternatives of month() function would be datepart(month, me_date) Or, Extract (month from me_date) Or, Strftime('%m', me_date)

Regards!

If you want to get the max(me_date) from previous month of max(me_date) of table1 then try this:

select max(me_date)
from table1
where datediff(month, me_date, (select max(me_date) from table1)) = 1;