Re: Strange intermittent WebDNA problems
This WebDNA talk-list message is from 2008
It keeps the original formatting.
numero = 70097
interpreted = N
texte = On Apr 9, 2008, at 11:26 AM, Alex McCombie wrote: > Lol Bob. Not suggesting a conspiracy. Just pointing out that multiple > servers that have run forever without a glitch seem to be effected > and some > at about or damn near the same time. > > I am having a similar issue but my OSX server is Webstar. > > Boy can I pick em or what! :-) > Now that Alex has spoken up to say he's running WebSTAR, I don't feel so shy. :-) I started having the "strange" problem a while ago too. I think the first instance was Mar. 14th at 7:16 AM Pacific Daylight Time (and I'm still wondering if the change to daylight time has something to do with it -- oh, I've neglected to check on the reports so far, do we have Australians with the problem?). Things just periodically stop working, manifested mainly by empty order files getting "stuck" in the Orders folder. It seems like the CPU usage of WebCatalog was high once, but lately I think I've been catching it before it goes up very much. I've tried removing the empty order files and that doesn't seem to fix things, I end up rebooting and then things work okay again. I once or twice saw the "can't open order file" error message, and have also seen error messages that I think happen because the WebDelivery folder gets swept by a trigger before the order ever gets processed. Besides the orders not working, I think there are other things that begin to go south, but it's hard to tell much from the log files and ErrorLog. Software Update was last run on Mar 7, but I believe there are reports from non-Macintosh servers, so that's probably not a likely source of the problem. Here are my vitals: * [version] and [platform] unix-Macintosh OS X WebSTAR V Plug-in version 5.1c on OSX 10.4.11 on a PowerPC G4 * Database size (MB) The largest are Orders.db and Lineitems.db, but they were actually reduced in size significantly before the problems started. There are numerous other databases, but none that have more than about 5000 records. store/WebMerchant/Orders.db 16666 records, using 13,583K of RAM 90 fields store/WebMerchant/LineItems.db 24168 records, using 3,163K of RAM 12 fields * Load ("heavy" being transactions per seconds or something) Load is moderate to light. * Is safe-write on or not? Safe-write is on. * Are you finding mal-formed record sets when looking at the .db? Haven't seen any. * Does this database interrelate with other .db's? Can't identify one database as a specific problem. The problem seems to be with running orders, so I guess that means Orders.db and LineItems.db, not sure what to say about interrelating with other db's. * Do you find any difference of using [exclusivelock..] as apposed to not using it? Haven't experimented with that. * Are you sure that the /etc/init.d/WebCatalogCtl is started? (seeing two WebCatalog proccesses when doing "top -u apache" Two processes are running. WebCatalog doesn't seem to stop running, it just begins to fail in processing orders. * Are you doing any encrypting values to the database!? No. * what date did the problem start? Looks like March 14th was the first time. -- Dana ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list
. To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/
Associated Messages, from the most recent to the oldest:
On Apr 9, 2008, at 11:26 AM, Alex McCombie wrote: > Lol Bob. Not suggesting a conspiracy. Just pointing out that multiple > servers that have run forever without a glitch seem to be effected > and some > at about or damn near the same time. > > I am having a similar issue but my OSX server is Webstar. > > Boy can I pick em or what! :-) > Now that Alex has spoken up to say he's running WebSTAR, I don't feel so shy. :-) I started having the "strange" problem a while ago too. I think the first instance was Mar. 14th at 7:16 AM Pacific Daylight Time (and I'm still wondering if the change to daylight time has something to do with it -- oh, I've neglected to check on the reports so far, do we have Australians with the problem?). Things just periodically stop working, manifested mainly by empty order files getting "stuck" in the Orders folder. It seems like the CPU usage of WebCatalog was high once, but lately I think I've been catching it before it goes up very much. I've tried removing the empty order files and that doesn't seem to fix things, I end up rebooting and then things work okay again. I once or twice saw the "can't open order file" error message, and have also seen error messages that I think happen because the WebDelivery folder gets swept by a trigger before the order ever gets processed. Besides the orders not working, I think there are other things that begin to go south, but it's hard to tell much from the log files and ErrorLog. Software Update was last run on Mar 7, but I believe there are reports from non-Macintosh servers, so that's probably not a likely source of the problem. Here are my vitals: * [version] and [platform] unix-Macintosh OS X WebSTAR V Plug-in version 5.1c on OSX 10.4.11 on a PowerPC G4 * Database size (MB) The largest are Orders.db and Lineitems.db, but they were actually reduced in size significantly before the problems started. There are numerous other databases, but none that have more than about 5000 records. store/WebMerchant/Orders.db 16666 records, using 13,583K of RAM 90 fields store/WebMerchant/LineItems.db 24168 records, using 3,163K of RAM 12 fields * Load ("heavy" being transactions per seconds or something) Load is moderate to light. * Is safe-write on or not? Safe-write is on. * Are you finding mal-formed record sets when looking at the .db? Haven't seen any. * Does this database interrelate with other .db's? Can't identify one database as a specific problem. The problem seems to be with running orders, so I guess that means Orders.db and LineItems.db, not sure what to say about interrelating with other db's. * Do you find any difference of using [exclusivelock..] as apposed to not using it? Haven't experimented with that. * Are you sure that the /etc/init.d/WebCatalogCtl is started? (seeing two WebCatalog proccesses when doing "top -u apache" Two processes are running. WebCatalog doesn't seem to stop running, it just begins to fail in processing orders. * Are you doing any encrypting values to the database!? No. * what date did the problem start? Looks like March 14th was the first time. -- Dana ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/
Dana Keil
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
calculating ShipCost depending on weight (1997)
weiered lookup (2004)
Need relative path explanation (1997)
OS X Problem... (2000)
Running 2 two WebCatalog.acgi's (1996)
Resume Catalog ? (1997)
Web DNA... (2007)
RE: MS Commerce Comparison (long) (1998)
ftp to webstar (2001)
default error on NT? (1997)
E-mailer error codes (1997)
No incentive ... (2000)
Every other record deleted (2005)
remotely creating and populating a stock inventory db -almost there! (1999)
Quickie question on the email templates (1997)
.hdr files (2004)
WebCat2 - storing unformatted date data? (1997)
Help!!!! Purchases not going through! (1997)
strange [Shell] things / Linux.2 (2000)
[AppendFile] problem (WebCat2b13 Mac .acgi) (1997)