Message-ID: <29344109.1075842017871.JavaMail.evans@thyme> Date: Fri, 16 Mar 2001 08:16:56 -0800 (PST) From: bob.shults@enron.com To: chris.edmonds@truequote.com Subject: Truequote-EnronOnline Test Status Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-From: Shults, Bob X-To: Chris Edmonds X-cc: stew@enform.com, richard.toubia@truequote.com, dennis.crum@truequote.com, Zipper, Andy X-bcc: X-Folder: \ExMerge - Zipper, Andy\TQ X-Origin: ZIPPER-A X-FileName: andy zipper 6-26-02.PST Chris, Here is our status as of yesterday and answers to Richard's comments: Primary issue: 4 to 5 second latency between Enron Stack/EnronOnline TEST changes to TrueQuote browser update. Updates from the Enron Stack to the True Quote interface bridge are instantaneous. Other issues: 1. Products hit/lifted on Truequote do not show price/volume updates from EnronOnline after transaction is complete on EnronOnline. 2. Multiple product activation results in a "stepped" display of products on Truequote browser as opposed to all at once on EnronOnline. 3. EnronOnline sends a bundle of information (e.g. products activated, prices changed, etc.) to the Truequote server. Truequote server takes longer to process the incoming data than the EnronOnline server will stay connected. Truequote has possibly resolved this issue by sending an immediate response for all incoming data in order to prevent the EnronOnline server from cutting the connection. Oracle error: Kevin Meredith forwarded the email containing the text for the Oracle error to Kenneth Lim at 7:21AM. Kenneth responded with an answer as to why we were receiving the error for transaction attempts by 7:36AM. I informed Stewart Clark as to the status of the issue surrounding the Oracle error in our first conversation at approximately 9:00AM. Kenneth's response: Approximately 6:20pm Wed night, there was an update to the TEST database which caused this error, we resolved this error along with our internal EnronOnlineTEST site with the resynching of our servers. Price Update on any transaction attempt: This has always been the case with EnronOnline to ensure the correct price is showing Chris please call me on my cell 713 825-6372 "Chris Edmonds" on 03/15/2001 11:00:44 AM To: cc: "Stewart Clark \(E-mail\)" , "Richard Toubia \(E-mail\)" , "Dennis Crum \(E-mail\)" Subject: FW: TQ-EnronOnline test Status Bob: Here is my latest update from my team. How does it compare to your current status? In a conversation this morning with Stew, he brought up an interesting point regarding the proposed performance of the connection. Since we are using the public internet VPN tunnel, we are encrypting the data to /from EOL during this test. The encryption process is overhead on the transfer between the systems. When we begin using the dedicated T1s, the plan is not to use the encryption between the sites. This will eliminate the associated overhead and increase performance. I will also call you to follow-up. Best regards, Chris -----Original Message----- From: Toubia, Richard [mailto:Richard.Toubia@truequote.com] Sent: Thursday, March 15, 2001 11:10 AM To: Edmonds, Chris; Crum, Dennis Cc: Clark, Stewart D.; Paxton, Chris; O'Neill, Maria Subject: TQ-EOL test Status Per your request, here are the testing status and issues: Attempting Transactions: EOL is sending an ORACLE Error which is failing any transaction attempt. A Bridge Server error is suspected. A copy of the Error was emailed to Kevin Meredith and Jason Althaus yesterday. No response yet. EOL is sending a price update for any transaction attempt, Filled or Killed. This was not expected by our team. (I think the stack manager is set up to automatically regenerate a price, a different price if a transaction takes place or same price if transaction is failed) The price update triggers our Deal Execution Engine which attempts another transaction, and we have a loop. Performance issues: We are looking into improving performance on our side (such as better Data Base sweeping and STOR procedures). In addition, EOL is testing their speed on a LAN, we are testing over the internet with a moderate-performance Server. In addition, the first two issues have to be resolved before appropriate performance evaluation can be meaningful. During testing we came across some issues: Bad TRP for BOM + 1 in Gas. This was fixed EOL tried sending wrong Prices, wrong volumes or wrong products to test how our system will handle them. These wrong products/ prices/volumes were disregarded and not posted at all (which is what we are looking for) Enform team are writing some code and changing other to accommodate some of the issues that transpired during testing. Regards, Richard