I was looking at the current processes in the MSSQL database and noticed something odd. A process that was logged for one user was trying to use a sql statement intended for another database. I also cruised the log files and noticed that one of the taf files had repeatedly logged long access times and periodic KILL commands from witango. Ah ha, I thought. I found the bad taf. So I opened the taf in the editor and looked at it hard, but could see no problem. The taf was doing 2 searches, one was returning all the rows in a basket, the other was summing all the rows of the the prices in that table. But everytime I hit the taf, it bombed everything and locked both machines.
I was going nuts. This is a taf I have used for the past 2 years on at least 6 other sites with absolutely no problems. Just on a whim, I right clicked the search for the sum search and brought up the properties. Under the deployment tab, the box was checked "Same As Development" but the greyed out fields showed a different database name, username and password. So I unchecked the "Same As Development" box and filled in the correct database then tested the taf.
Now instead of bombing, I got an error telling me that "null" could not log in. So I highlighted the search and reset the data source using the "Set Data Source" and everything is up and running well.
I have 6 sites that are virtually the same. Different artwork and databases but the way the baskets work and taf files are the all same. I did these sites for a client that wanted these sites set up like this. When I created these sites, I finished the first one then copied all the files for the next one, created the new database, then would go through each taf and hightlight all their database searches, inserts and updates then "Set Data Source" to the new database. I have since noticed that out of maybe 100 search actions (not tafs, but actions within tafs), there will be 5 or 6 actions that even though the "Same As Development" box is checked, the greyed out database name, username and password under deployment for the search properties will have a different database listed. None of these databases are listed in the dependencies directory.
If anyone could shed some light as to what is going on, I would appreciate it. I thought that when I set the data source, they all would change identically. Also, why would this suddenly be a problem when the taf was running correctly for the past years? I have been doing regular hotfixes and updates to the servers(last one was last week), but have not changed these taf files.
Thanks for a reply!!!
Yes, it is called "threadPoolSize" But do this test in small incrementsBen Johansen - http://www.pcforge.com Authorized Witango Reseller http://www.pcforge.com/WitangoGoodies.htm Latest downloads & List Archives @ http://www.witango.ws -----Original Message----- From: [EMAIL PROTECTED] [mailto:owner-witango-talk@;witango.com] On Behalf Of Web Dude Sent: Monday, October 28, 2002 1:21 PM To: Multiple recipients of list witango-talk Subject: RE: Witango-Talk: HELP What the heck?? I will give this a try. Do you mean the thread count in the witango configuration? I currently have this set at the default level of 20. I have never really had to adjust this in the past. I wish I could find the offending sql statement! Anyway of doing that? I am currently running around 400 taf files. Is it possible an increase in traffic could cause this? If so, how high should i set the thread count? Sorry, I am just trying to get this straightened out. I am going nuts. p.s. Killed the transaction server and sill froze around 1/2 hour later. Obviously that was not it.I wonder if you have a sql statement that is causing a deadlock on the SQL Server which in turn is hold the thread, which then witango would open another thread until running out of the thread count.>Maybe you could increase tread count, due to a sql statement taking longer than expected due to increase in data. Just some thoughts Ben Johansen - http://www.pcforge.com Authorized Witango Reseller http://www.pcforge.com/WitangoGoodies.htm Latest downloads & List Archives @ http://www.witango.ws -----Original Message----- From: [EMAIL PROTECTED] [mailto:owner-witango-talk@;witango.com] On Behalf Of Web Dude Sent: Monday, October 28, 2002 12:02 PM To: Multiple recipients of list witango-talk Subject: Witango-Talk: HELP What the heck?? Okay... 2 production servers. Server 1 NT 4.0 SP 6a IIS 4 Witango 2000 sp1 40 sites 1 gig RAM All security updates in place Server 2 NT 4.0 SP 6a SQL 7.0 1 gig RAM All security updates in place. Last Friday I was on server 2 just surfing when suddenly the computer pretty much froze. I went to the task manager (which took forever) and noticed the cpu being hogged by sqlserver.exe. I went to server 1 and noticed in task manager(which also took forever) that the witango server was hogging all the cpu. I managed to stop the witango server and immediately the cpu was freed. I restarted the witango server and about 15 minutes later, everything froze again on both servers. I haven't coded in the past week, so I don't think there is a coding problem in witango. Most of the sites have been up and running for the past 2 years, some have been up for 4 years or more. BUT suddenly I am having all sorts of problems. It seems the servers (both) are freezing 8 to 10 times a day. There are no email servers or email clients on these servers, so I don't think I have a virus (but not sure). I have used netstat extensively to see if I can find a wierd connection. I did notice however, that microsft transaction server seemed to have been started at about the time that I started getting these freezes. Not sure if this is related or not. I did reboot both machines several times that day. I am pretty sure that MSDTS is installed by default, even though I do not use it. I have looked at configuration files for witango, and all seems to be the way I left them. I have turned on logging to see if I can catch an offending taf file being executed, but it seems that the freezes are random. I have noticed, however, that whenever a freeze happens, I seem to get anywhere from 10 to 30 system reapers in the log file. Not sure if this is related. I have since shut down microsoft transaction server on server 1 (not even sure that I need it), but left it up on server 2. I need to know: Do I need the transaction server? Any other ways of trying to figure this out (it is driving me nuts)? Anything I should be looking for in the logs? Any Suggestions at all? -- ________________________________________________________________________TO UNSUBSCRIBE: send a plain text/US ASCII email to[EMAIL PROTECTED]with unsubscribe witango-talk in the message body ________________________________________________________________________TO UNSUBSCRIBE: send a plain text/US ASCII email to[EMAIL PROTECTED]with unsubscribe witango-talk in the message body-- ________________________________________________________________________ TO UNSUBSCRIBE: send a plain text/US ASCII email to [EMAIL PROTECTED] with unsubscribe witango-talk in the message body ________________________________________________________________________ TO UNSUBSCRIBE: send a plain text/US ASCII email to [EMAIL PROTECTED] with unsubscribe witango-talk in the message body
--
________________________________________________________________________
TO UNSUBSCRIBE: send a plain text/US ASCII email to [EMAIL PROTECTED]
with unsubscribe witango-talk in the message body
