Hello Sniffer Folks, Please grab the following file:
<http://www.sortmonster.com/MessageSniffer/Betas/MessageSniffer2-3.0i9-Distribution.zip> This is the last interim/testing version before the next official release of Message Sniffer which will be version 2-3.1. This version has two major features: * The timing parameters for the persistent server have been adjusted. The maximum poll time for a persistent server is now 2 seconds. Also, the poll time for the persistent server now expands linearly. Client poll times still expand using a natural spiral (Fibonacci sequence). This provides for smoother response time adjustments and a faster "wake-up time" after a period of quiescence. * Clients will now attempt to open their results file a second time after a small delay (100ms) if the first attempt fails. The Win32 OS has a hard time keeping up with the aggressive new timing of this version of Message Sniffer. As a result we were seeing a failure rate of between 2% and 8% at random when a client instance would open it's result file after the persistent server had processed a message. Under these circumstances the clients would recover by loading the rulebase themselves and processing the message. This caused system loads to increase. We found that on our equipment we are able to completely mitigate this problem by simply detecting the error, waiting a short period, and then opening the file again. As a result the cost of a failed open attempt is now very low (100ms of sleeping time for the client effected.) Our testing on multiple versions of Linux did not show this problem, however the effect of the additional code is vanishingly small as long as no open errors occur - and if one does then it will be well handled. ---- The additional features that we had planned for the 2-3.1 release have been moved off to 2-3.2 in order to get this current version into official production and to minimize the complexity of this release. Work on the next version will begin immediately with 2-3.1i1 after a short break. ---- We look forward to your feedback on this latest interim/testing release. At this point we believe that there are no significant bugs in this version. It should be safe in a production environment. If there are no error reports or problems detected in our testing then this version will be re-published as the official release 2-3.1 early next week. Please let us know your experiences with this version as quickly as possible and keep us posted. As always, we will be here nearly 24/7 ;-) Thanks, _M Pete McNeil (Madscientist) President, MicroNeil Research Corporation Chief SortMonster (www.sortmonster.com) This E-Mail came from the Message Sniffer mailing list. For information and (un)subscription instructions go to http://www.sortmonster.com/MessageSniffer/Help/Help.html
