- Original Message -
> From: "Adarsh Sharma"
> Thanks Johan, but i mentioned before that adding auto increment
> column doesn't solve the issue & it causes errors in the multi
> threaded application.
If it causes errors, you have other problems than this.
> Multiple clients calls this
Thanks Johan, but i mentioned before that adding auto increment column
doesn't solve the issue & it causes errors in the multi threaded
application.
Multiple clients calls this insert procedure simultaneously, so it fails
the transactions if two or more clients reads the same ID value.
I need t
- Original Message -
> From: "Johnny Withers"
>
> I'm not sure, It seems to me the proper way to do would be to insert
> into table1, get the insert ID, then insert into table2 using that ID,
> this is pretty standard stuff.
>
> Not sure why, in this case, he cannot do that.
last_insert
I'm not sure, It seems to me the proper way to do would be to insert into
table1, get the insert ID, then insert into table2 using that ID, this is
pretty standard stuff.
Not sure why, in this case, he cannot do that.
-JW
On Wed, Feb 22, 2012 at 8:54 AM, Rhino wrote:
> I miised the first mess
You can also handle this with transactions:
CREATE TABLE `seq` (
`seq_num` int(10) unsigned NOT NULL DEFAULT '1000'
) ENGINE=InnoDB DEFAULT CHARSET=latin1
#Initialize sequence numbers
INSERT INTO seq(seq_num) VALUES(1000);
#Get next sequence number
START TRANSACTION;
UPDATE seq SET seq_num=L
I agree with the testicular remedy, but in the case of the iron codpiece, I
can think of another approach which may work for you. It still uses Select,
but reads a one-row table, so it shouldn't hurt performance much. The table
serves no other purpose than storing the next available PK; call the ta
- Original Message -
> From: "Adarsh Sharma"
>
> Today I noticed some duplicacy in my c_id column which is not
Yes, that's what you get if you don't use auto_increments.
> I need multiple client select cid from 2 tables & insert data with
> adding 1 to previous C_id in isolated manner.
