Hi All, I have upgraded the server from Database server to SQL. I have Citrix server on the same server as tHE SQL. When we run the Inv Batch Job it takes 7days to complete… which is a huge probelm. Does the Citrix server slow down the SQL and do I need to run both SQL and Citrix on seperate servers? I have ±25 users accessing the Nav Remotly and ± 25 users onsite. Your solutions and input would be greatly appreated… Server spec as follows: Dell XEON 2 X 1.8GIG 3GIG MEM RAID 5 72GIG <10 000RPM DRIVES> Windows Server 2000 SQL Server 2000 CITRIX Server DATABASE SIZE 40gigs
I would recommend using two servers: one for SQL and one for Citrix. I would recommend not using RAID 5 - this is a slow option. The SQL implementation is known to be slower than C/SIDE but the time for this job is excessive. Do you run other jobs at the same time? Have you made modifications to the processing? There are a number of known locking issues with the SQL version and these may be affecting the time taken to process the batch job. If you are not already talking to your Solution Centre contact them to discuss this serious problem. Good luck.
No other jobs are running on the server when running the batch job. It even has come to disconnecting the network card so as no one can access the server. Before running the batch Job, all locked tables are “Killed” as the batch Job will not run if tables are locked. We are talking to Microsft about this problem and have been waiting a couple of days now for a response. Thanks for your help!
You say you have upgraded from classic (Native) ? Is the whole process slow, putting on orders posting journals etc: I have read several posts on here about the need in SQL to optimise, IE: rebuild the keys after upgrading search for “SQL Speed”, How many Sales Orders in the batch? http://www.navision.net/forum/topic.asp?TOPIC_ID=10479 8GB Database 50 Users if they are all Sales Order Processing while you are trying to post that will slow it down with locking. I cannot see that it could take 7 days to post a batch of orders unless you have done major changes to the posting routines. You could also try batching the orders in small quantities if there are lots, after each batch the commit will be quicker and the tables will be freed between batches.