db connect 2.3 runtime error

[kenkenou]

splunk enterprise free license:6.3.3
db connect:2.3
driver:mysql 5.1
DB:mysql

when i excuse dbxquery command or set db input, i got the error message below.

ERROR MESSAGE:
External search command 'dbxquery' returned error code 1. Script output = "RuntimeError: Failed to run query: "SELECT * FROM XXXX.XXXX", params: "None", caused by: TypeError('string indices must be integers',). "mm

| dbxquery query="SELECT%20*%20FROM%20%60XXXX%60.%60XXXX%60" connection="XXXX"

Please help!!!

[gyarici]

I am not sure if it is buggy behavior or not, but it seems it could be come from db connect python script to determine indexes of the string in the db table.

Example:

>>>string = 'hi'
>>>print string[0]

Output is 'h' as a first string.

If you run the similar command , but string index is string instead of integer,

  >>>string = 'hi'
   >>>print string['hi']

Output is :
  File "<pyshell#12>", line 1, in <module>
    print string['hi']
TypeError: string indices must be integers, not str

You can have the same error, Therefore it is more likely incompatibility of the column types with db-connect.

As I mentioned it before, blob object is causing such kind of issue as well. Make sure that there is no blob object in your table. If you have, please take a look the link for it.
link text

[kenkenou · Accepted solution]

It has been solved.
After replace the "*" to the columns it works.
for example "selcect * from XXX.XXX" -> "select columnA, columnB from XXX.XXX"

but, why the * doesn't work...

[hartfoml]

SOOOO why doesn't the "*" work?????

[gyarici]

It sometimes happens, if there is a binary content data in one of the columns are not supposed to read/browse from db .Thus most of the times, you should exclude such kind of columns in order to get proper output.

[hartfoml]

Sounds logical but there is no binary data in my output only about 60 colunm's and 1200 rows of text readable data