what are the possible reasons for 'Arithmetic overflow error converting expression to data type int' in SQL server

Question

When this command is run, there is SQL server error

'Arithmetic overflow error converting expression to data type int'.

SELECT sum(size) FROM [dbname]..sysfiles sf, [dbname]..sysfilegroups sfg WHERE sfg.groupname = 'PRIMARY' AND sf.groupid = sfg.groupid

Because the file size is less than 16TB, bigint is not required. Still I tried to execute cast(size as bigint), the error persists.

SELECT sum(cast(size as bigint)) FROM [dbname]..sysfiles sf, [dbname]..sysfilegroups sfg WHERE sfg.groupname = 'PRIMARY' AND sf.groupid = sfg.groupid

So the question is, what are the reasons for arithmetic overflow error other than exceeding the file size limit of 16TB.

Answer 1

-- casting using DECIMAL(38,2) throws exception, Size needs to be casted as bigint

select [SizeInMB] = CAST( ((SUM(Size)* 8) / 1024.0) AS DECIMAL(38,2) ) from sys.master_files

-- This fix is sufficient, but cannot limit to 2 decimal digits

select [SizeInMB1] = ((SUM(CAST(Size AS BIGINT))* 8) / 1024.0) from sys.master_files

-- 2 decimal digits are not available in fix, as divide by 1024.0 converts to default 6 digits

select [SizeInMB2] = ((SUM(CAST(Size AS DECIMAL(38,2)))* 8) / 1024.0) from sys.master_files

-- so 2 casts are required to eliminate Arithmetic overflow exception and limiting 2 decimal digits

select [SizeInMB3] = CAST( ((SUM(CAST(Size AS BIGINT))* 8) / 1024.0) AS DECIMAL(38,2) ) from sys.master_files