Increasing the datasourcelife= will cause Witango to drop/create datasource connections less. I've found on windows that the server runs slightly better when it has to do less management of datasources (longer life). But I still prefer to keep this number as low as possible since I believe that refreshing the connections regularly is a good thing. Your platform/software might have very different outcomes.
Robert -----Original Message----- From: Roland Dumas [mailto:[EMAIL PROTECTED] Sent: Friday, March 05, 2004 5:54 PM To: [EMAIL PROTECTED] Subject: Re: Witango-Talk: Crashes on OS X My second urgent problem is increase in frequency of witango crashes. It is crashing now 4-6 times a day. Needless to say, all the variables are dumped as it restarts, losing user sessions, etc. The crash log is very consistent. Always an IODBC thing going on. I look and there is absolutely no correlation with what was going on just prior to the crash. Neither the web log nor the witango log shows anything like a pattern that could lead to an application, a database, a record, etc., and a mirror of the site runs as a test server and has not seen a single crash of this kind. My current theory is that it's related to these kinds of events: 05/03/2004 01:23:59 0 [Expired] ODBC datasource ridgerw Expiry of datasources. Is it possible that something in that event could be tripping up witango? How does one change the expiry anyway? Is that a value set somewhere? On Feb 9, 2004, at 8:34 AM, Roland Dumas wrote: > Can anyone tell me if this is evidence of RAM problems? The crashed > thread records tend to be rather consistent in what they're writing > and where they are writing it, as far as my uneducated eyes can tell. > >> On Feb 6, 2004, at 10:05 AM, Ben Johansen wrote: >> >>> Disclaimer: Not a Mac Guy ;-) >>> >>> This sounds a lot like a RAM module going bad. >>> >>> This can be intermittent for several reasons >>> -the ram chip is fail intermittently >>> -the failed location on the ram chip is in a high mem address and >>> only >>> is reached under higher loads. >>> >>> One thing you can do when looking at those funny number is to see if >>> you >>> see a consistent funny number :) >> >> OK: you can tell me if this constitutes a consistent funny number. I >> grabbed the crash log, clipped out the crashed thread and here are a >> few. Definitely recurring numbers in the "in" (is that a memory >> location or something?) >> >> Thread 10 Crashed: >> #0 0x94f92e68 in _iodbcdm_driverload >> #1 0x94f94f64 in SQLDriverConnect >> #2 0x0016b8e0 in 0x16b8e0 >> #3 0x00047570 in 0x47570 >> #4 0x000685b8 in 0x685b8 >> #5 0x00024ec0 in 0x24ec0 >> #6 0x00025278 in 0x25278 >> #7 0x0001949c in 0x1949c >> #8 0x00019e24 in 0x19e24 >> #9 0x00017ac8 in 0x17ac8 >> #10 0x0001a910 in 0x1a910 >> #11 0x0014e09c in 0x14e09c >> #12 0x90020c28 in _pthread_body >> >> Thread 2 Crashed: >> #0 0x94f92e68 in _iodbcdm_driverload >> #1 0x94f94f64 in SQLDriverConnect >> #2 0x0016b8e0 in 0x16b8e0 >> #3 0x00047570 in 0x47570 >> #4 0x000685b8 in 0x685b8 >> #5 0x00024ec0 in 0x24ec0 >> #6 0x00025278 in 0x25278 >> #7 0x0001949c in 0x1949c >> #8 0x00019e24 in 0x19e24 >> #9 0x0001830c in 0x1830c >> #10 0x0007cba0 in 0x7cba0 >> #11 0x0009dd18 in 0x9dd18 >> #12 0x0009d780 in 0x9d780 >> #13 0x00019750 in 0x19750 >> #14 0x00019e54 in 0x19e54 >> #15 0x00017ac8 in 0x17ac8 >> #16 0x0001a910 in 0x1a910 >> #17 0x0014e09c in 0x14e09c >> #18 0x90020c28 in _pthread_body >> >> Thread 14 Crashed: >> #0 0x94f92e68 in _iodbcdm_driverload >> #1 0x94f94f64 in SQLDriverConnect >> #2 0x0016b8e0 in 0x16b8e0 >> #3 0x00047570 in 0x47570 >> #4 0x000685b8 in 0x685b8 >> #5 0x00024ec0 in 0x24ec0 >> #6 0x00025278 in 0x25278 >> #7 0x0001949c in 0x1949c >> #8 0x00019e24 in 0x19e24 >> #9 0x00017ac8 in 0x17ac8 >> #10 0x0001a910 in 0x1a910 >> #11 0x0014e09c in 0x14e09c >> #12 0x90020c28 in _pthread_body >> >> Thread 20 Crashed: >> #0 0x94f92e68 in _iodbcdm_driverload >> #1 0x94f94f64 in SQLDriverConnect >> #2 0x0016b8e0 in 0x16b8e0 >> #3 0x00047570 in 0x47570 >> #4 0x000685b8 in 0x685b8 >> #5 0x00024ec0 in 0x24ec0 >> #6 0x00025278 in 0x25278 >> #7 0x0001949c in 0x1949c >> #8 0x00019e24 in 0x19e24 >> #9 0x00017ac8 in 0x17ac8 >> #10 0x0001a910 in 0x1a910 >> #11 0x0014e09c in 0x14e09c >> #12 0x90020c28 in _pthread_body >> >> ______________________________________________________________________ >> __ >> TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf >> > > _______________________________________________________________________ > _ > TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf > ________________________________________________________________________ TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf ________________________________________________________________________ TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
