wcf-sql adapter bug for operation on same table
based on scenario you've described, dirty reads out (since want update on records retrieved). so you've got options here if looking configuration based solution. if data you’re polling on can near real time, can set polling interval in 8-10 minute range (or longer if business scenario supports it); set in pollingintervalinseconds on ‘binding’ tab of receive location (thus 420 - 600). i set retry intervals on send port inserting data in 1-3 minute range (and @ least 3 retries); if polling data returned can go on shorter timeframe retry happen quicker. you find retry interval , retry count on 'transport advanced options' of port configuration send port. in scenario, if blocked on insert because of polling, should retry insert before next polling interval. keep mind these settings complete shot in dark since unfamiliar number of rows on bulk insert , how many rows returned on polling statement. if data volumes small, can skew towards shorter times.
a couple of things keep in mind solution. first, not eliminate possibility of block, lowers chances occur reducing polling interval; though insert should succeed on retry. it creates scenario data received earlier can inserted after data received later. this because data dehyrated because of blocking, data comes after inserted before resumption of suspended message.
BizTalk Server > BizTalk Server General
Comments
Post a Comment