Count Events, Group by date field

[hogan24]

I have data that looks like this that I'm pulling from a db. Each row is pulling in as one event:
trxn_id create_dt_tm
123456 2013-11-22 11:01:22.xxx
123457 2013-11-22 11:01:23.xxx
123458 2013-11-22 11:01:24.xxx
123459 2013-11-22 11:02:22.xxx
123460 2013-11-22 11:02:22.xxx

I'd like the results to look like this in a timechart:
Time Count
2013-11-22 11:01 3
2013-11-22 11:02 2
etc

When I do something like this below, I'm getting the results in minute but they are grouped by the time in which they were indexed.
"index=main | timechart count(edi_trxn_detail_id) span=1m"

How do I tell splunk to group by the create_dt_tm of the transaction and subsequently by minute? Thanks.

[hogan24 · Accepted solution]

I think I may have figured this one out through suggestions and trial and error...

create_dt_tm | convert timeformat="%m/%d/%y %H:%M" ctime(create_dt_tm) as Minute | stats count as "Transactions" by Minute

This gives me the following which is what I was looking for. Thanks to those who helped!
Minute Transactions
11/24/13 10:00 8
11/24/13 10:01 6
11/24/13 10:02 4
etc

[somesoni2]

Try following

index=main sourcetype=yoursourcetype| eval create_dt_tm=strptime(create_dt_tm,"%Y-%m-%d %H:%M") | stats count by create_dt_tm | eval create_dt_tm=strftime(create_dt_tm,"%Y-%m-%d %H:%M")

[rechteklebe]

You could extract the values until the minutes in create_dt_tm field by using Rex Command.
With the New field you can simply make a timechart span=1m count by "newfield".