declare @a varbinary(100)
declare @b varbinary(100)
declare @f float
set @f=123.456
set @a=EncryptByPassPhrase(‘vival’, cast(@f as varbinary))
set @b=DecryptByPassPhrase(‘vival’, @a)
select @b
select cast(cast(@b as decimal(10,4)) as float)

getting error at last line,
Arithmetic overflow error converting varbinary to data type numeric.

plz tell what’s d mess?

Answer to first question:
Certificates will have to be transferred to different server physically first and setup with SQL Server before using its name to decrypt the data.

Answer to second question:
Yes there is way to insert the data with encryption on.

I will soon write separate article explaining both of the above methods.

Regards,
Pinal Dave ( http://www.SQLAuthority.com )

The most interesting fact about your articles is they are written in very simple language which can be undertsood easily by a person which dont have much knowledge about SQL.

Thanks for the article

I have few questions regarding your article :

1. In your example we encrypted symmetric key using a certificate and then we encrypted the column in the table using that symmetric key and when decrypting the data we are just giving the name of the symmetric key and then name of the certificate, no where we mentioned the password, for some one to decrypt the data all he needs to know is name of the symmetric key and name of the certificate used to encrypt data and then he is good to go.

Please explain if I misunderstood.

2. we first inserted the data and then encrypted it and after encryption we are deleting that column, because it doesn’t make sense to keep that column. what if we want to insert more data in that column aftre it is encrypted, do we have to decrypt the existing column and then insert the data and the again encrypt it ?

or is there any way that we can insert the data into the encrypted column with out decrypting and inserting and encrypting it again.

Please clarify.

one request: please give a similar example for Assymetric key.

Thanks.