Glad to be of help. kill -3 is one of those gems you should never forget as it can save your life. "Kill to save your life": nice slogan... :D
Martijn On 3/19/08, Jeremy Levy <[EMAIL PROTECTED]> wrote: > Thanks Igor / Martijn for your help. > > As you mentioned Wicket had nothing to do with it. I had a piece of code > that was executed when a user clicked a certain link and was blocking by > accident, caught in a loop peging the CPU. Your suggestions of kill -3 to > get the thread dump over a couple of samples helped me to narrow down the > issue and fix it. > > Thanks very much. > > > Jeremy > > > On Tue, Mar 18, 2008 at 4:26 PM, Martijn Dashorst < > [EMAIL PROTECTED]> wrote: > > > I suggest getting a thread dump at the time of the exception, to see > > which threads are stuck. You can do that by sending a kill -3 to the > > java process running your webapp. The stack traces should be in your > > console output captured by your server. > > > > Martijn > > > > On 3/18/08, Igor Vaynberg <[EMAIL PROTECTED]> wrote: > > > i think the only way this can happen is that you have something that > > > really takes longer then 1 minute to execute. i dont see how wicket > > > can take over your cpu....so may be it is that runaway process that > > > does that... > > > > > > i think to debug it you might want to attach a profiler to the server > > > and wait for it to happen, take a cpu timing profile and see what was > > > going on during the spike... > > > > > > does the message tell you who it is locked by? which page or > > > otherwise? if not that is possibly something we can improve also... > > > > > > -igor > > > > > > > > > On Tue, Mar 18, 2008 at 1:06 PM, Jeremy Levy <[EMAIL PROTECTED]> wrote: > > > > I am seeing the "After 1 minute the Pagemap null is still locked" > > error in > > > > production a couple of times a day for about the last week or so. > > We put in > > > > a major update to our site about 3 weeks ago, but only recently > > started > > > > seeing this error. I suspect they are related. > > > > > > > > From doing some research on the mailinglist it seems that this comes > > up when > > > > a user requests a new page while there is an open request still > > processing. > > > > > > > > http://www.nabble.com/pagemap-locking-to11350774.html#a11350774 > > > > > > > > At the same time (or a few seconds before) the CPU is pegged at 100% > > and > > > > stays that way. Overtime the CPU usage progressively gets more and > > more > > > > backed up (I assume because this issue keeps happening and compounds > > the > > > > load). > > > > > > > > The longest page request on the site is a panel which has 4 child > > lazy > > > > loading panels but none of them take longer then a max of 4 seconds, > > which > > > > is pretty strictly enforced. > > > > > > > > The CPU getting pegged is a huge problem because it is forcing us to > > restart > > > > the server every some often. > > > > > > > > For now we have not been able to reproduce this error in a > > development > > > > environment, any suggestions, or troubleshooting ideas? > > > > > > > > Jeremy > > > > > > > > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > > > > -- > > Buy Wicket in Action: http://manning.com/dashorst > > Apache Wicket 1.3.2 is released > > Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.2 > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > -- Buy Wicket in Action: http://manning.com/dashorst Apache Wicket 1.3.2 is released Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.2 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
