[Aaron Markley]

I’m on Rick’s team and assisting.  

We’re calling odbctcl inside of a TCL script which is inside a thread as a UPOC.  We are using this script to insert multiple fields into a single MSSQL table.  This script runs once per file, with one state’s messages per file.  Once the ODBC insert happens, the application which uses the database encrypts the payload and sends it via HTTPS to external receivers.

As of the date of our upgrade, it appears that one of the database fields (which contains the main message payload) sometimes contains messages from multiple state files.  A database record could be sent to Washington (for example), but in the payload field, our receiving contact reported seeing messages for Washington and Illinois.  We have verified that in SMAT, the messages look correct and are not mixed by the time they hit the ODBC script.

[Aaron Markley]

Adding the indicated statement (odbc SQLSetStmtAttr $hstmt SQL_ATTR_PARAMSET_SIZE 1 0) prior to the SQLExecDirect command doesn’t appear to resolve these malloc errors we are seeing.

[Aaron Markley]

Tony,

 Our organization has created an ODBC connection to the SQL database of our own implementation of PHINMS.  Results are sent from our lab system in a batch process early every morning, which are processed and routed to this connection.  

Each message sent to the PHINMS database is then sent from PHINMS as a single record, so we have decided to compile messages from certain states into common files, each of which are then passed to PHINMS as individual records (for example, we compile all Massachusetts messages for one day into a single file, which is passed to PHINMS as one record).  Your method may differ based on how you are sending and who you’re sending to.

Let me know if you have other questions.

-Aaron

[Author]

Replies