On Thursday 12 June 2008, Carsten Bock wrote: > this is odd. The timer is inserted from set_timer (modules/tm/timer.c) > like this: > > insert_timer_unsafe( list, new_tl, > timeout + > ((timer_id2type[list_id]==UTIME_TYPE)?get_uticks():get_ticks())); > > get_ticks() / get_uticks() from "../../timer.h" is called (both in 1.3 > branch and in trunk). From that point of view, i think the "1841815" > value makes sense, since the server has been running for quite a while. > Or am i wrong here?
Hi Carsten, you are correct, the value was smaller during my tests because this was a new process. I misunderstood the debug output. > The curious thing here is, that the value of "1841815" is not created > from my module-code, but from the TM-API (t_request). > The "t_request"-call is created from a separate timer-process created by > my module (register_timer_process()); the timer triggered here is from > the "regular" timer-process (register_timer()). > Could this lead into trouble? Could be possible, but i've not used a scenario like you described so far. Does this error happens for every t_request call from this separate timer process? Henning _______________________________________________ Devel mailing list Devel@lists.openser.org http://lists.openser.org/cgi-bin/mailman/listinfo/devel
