Re: [WebDNA] An unknown error occured // Deadlock avoided
This WebDNA talk-list message is from 2011
It keeps the original formatting.
numero = 106925
interpreted = N
texte = I changed the subject to better describe what I was posting. Sorry that you and perhaps others missed it originally. Sincerely, Kenneth Grome > Ken, I see you started another thread with the actual > code to discuss it. Thanks for that. This thread has > grown too long and it my personal opinion is a bit to > fractured for any real use. > > I never meant to get into this discussion to start with, > but felt I had to say something. Which I have. I am > removing myself from this thread. > > Best of luck. > > On Mon, Jul 11, 2011 at 8:48 AM, Kenneth Grome
wrote: > >> You guys are all missing the point. The guys wants it. > > > > No one missed the point. > > > > I suggested a solution that will absolutely work, and > > it will give PigHead the data he wants. > > > > When you eliminate the db accesses as I suggested, you > > also eliminate the deadlock issue. Simple. > > > >> That all being said, that code looks like it could be > >> done with a two lines of code. > > > > If the site is so busy that the heavy traffic is > > causing the deadlock issue, replacing append with > > replace may not help. > > > > In this case the better solution may be to write to a > > log file rather than to the dbs. Since writing to a > > log file will absolutely avoid deadlock issues it is > > arguably a better approach -- especially if your goal > > is to solve the problem definitively the first time. > > > >> I cannot see how you know if someone has gone > >> offline or not based on the above code. > > > > Agreed. > > > > So just write to a log file that is formatted as a db > > and then it can be included in the PigHead's special > > admin template. Then search for only the hits > > recorded during the past 5 minutes or one hour or > > whatever PigHead's particular opinion is about how > > long people remain on the site after their last hit. > > > >> So if you need that simply add a > >> timestamp to the db and run a different script, I > >> recommend from a trigger or cron job, and have it > >> remove anyone that is idle for more than X minutes. > > > > My suggestion above does the same thing more easily > > while avoiding all db accesses, and it is guaranteed > > to eliminate the deadlock issue without fail. > > > > Look, I don't have any problem using dbs (or using > > replace instead of append) when dbs are actually > > necessary. But dbs are absolutely not necessary here, > > so why not avoid them altogether? Then you KNOW > > you're going to eliminate the problem! > > > > Sincerely, > > Kenneth Grome > > ------------------------------------------------------- > > -- This message is sent to you because you are > > subscribed to the mailing list . > > To unsubscribe, E-mail to: > > archives: http://mail.webdna.us/list/talk@webdna.us > > Bug Reporting: support@webdna.us > > --------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > archives: http://mail.webdna.us/list/talk@webdna.us > Bug Reporting: support@webdna.us
Associated Messages, from the most recent to the oldest:
I changed the subject to better describe what I was posting. Sorry that you and perhaps others missed it originally. Sincerely, Kenneth Grome > Ken, I see you started another thread with the actual > code to discuss it. Thanks for that. This thread has > grown too long and it my personal opinion is a bit to > fractured for any real use. > > I never meant to get into this discussion to start with, > but felt I had to say something. Which I have. I am > removing myself from this thread. > > Best of luck. > > On Mon, Jul 11, 2011 at 8:48 AM, Kenneth Grome wrote: > >> You guys are all missing the point. The guys wants it. > > > > No one missed the point. > > > > I suggested a solution that will absolutely work, and > > it will give PigHead the data he wants. > > > > When you eliminate the db accesses as I suggested, you > > also eliminate the deadlock issue. Simple. > > > >> That all being said, that code looks like it could be > >> done with a two lines of code. > > > > If the site is so busy that the heavy traffic is > > causing the deadlock issue, replacing append with > > replace may not help. > > > > In this case the better solution may be to write to a > > log file rather than to the dbs. Since writing to a > > log file will absolutely avoid deadlock issues it is > > arguably a better approach -- especially if your goal > > is to solve the problem definitively the first time. > > > >> I cannot see how you know if someone has gone > >> offline or not based on the above code. > > > > Agreed. > > > > So just write to a log file that is formatted as a db > > and then it can be included in the PigHead's special > > admin template. Then search for only the hits > > recorded during the past 5 minutes or one hour or > > whatever PigHead's particular opinion is about how > > long people remain on the site after their last hit. > > > >> So if you need that simply add a > >> timestamp to the db and run a different script, I > >> recommend from a trigger or cron job, and have it > >> remove anyone that is idle for more than X minutes. > > > > My suggestion above does the same thing more easily > > while avoiding all db accesses, and it is guaranteed > > to eliminate the deadlock issue without fail. > > > > Look, I don't have any problem using dbs (or using > > replace instead of append) when dbs are actually > > necessary. But dbs are absolutely not necessary here, > > so why not avoid them altogether? Then you KNOW > > you're going to eliminate the problem! > > > > Sincerely, > > Kenneth Grome > > ------------------------------------------------------- > > -- This message is sent to you because you are > > subscribed to the mailing list . > > To unsubscribe, E-mail to: > > archives: http://mail.webdna.us/list/talk@webdna.us > > Bug Reporting: support@webdna.us > > --------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > archives: http://mail.webdna.us/list/talk@webdna.us > Bug Reporting: support@webdna.us
Kenneth Grome
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:
founditems (2000)
Getting URL's entered manually (1997)
Duplicates (1998)
TCPCONNECT ---- POST? (2004)
blank orders (2001)
Hiding HTML and breaking the page (1997)
RE: Signal Raised (1997)
Using [detete] (1998)
NT [delete] (1998)
[Webcat 2]Next (1997)
WebCatalog for guestbook ? (1997)
WebDelivery downloads alias, not original ? (1997)
Math Function (1997)
Signal Raised (1997)
2cd pair.. [math] and European Date format (2004)
is your webDNA site 'froogle' ready ??? (2003)
[append] problem (1997)
WebCatalog2 Feature Feedback (1996)
Variable prices and Formulas.db (1997)
Problem (1997)