Hi Shuveb, On Mon, May 26, 2008 at 2:46 PM, Shuveb Hussain >> > > First the non-technical babble: > You application crashes once in a while? An ominous sign indeed. Why don't > you fix this first? Watchdog timers are only when your application crashes > unexpectedly.
Sure. I have found the issue and I have fixed it. It was a Heisenbug. http://en.wikipedia.org/wiki/Heisenbug#Heisenbugs It was due to enormous amount of unlinks without closing the fds associated with the file. > Technical: > Are you able to verify that the watchdog daemon is running? If /dev/watchdog > present and accessible? /dev/watchdog is present. but i dont see any watchdog process running in the system. Can you manually verify that the system reboots if > the watchdog's timer is not refreshed? Let us initially rule-out issues in > the watchdog driver in the kernel. Sorry I am forced to ask you back how to do this.How to refresh the timer from our application. I am using busybox 1.6.1 One more doubt.. whether the device shown in /dev/watchdog is a softdog or hardware watchdog. Any pointers towards these documentation is most welcome. I am not able to locate them. > All the best, > -- > Thanks Deepak _______________________________________________ To unsubscribe, email [EMAIL PROTECTED] with "unsubscribe <password> <address>" in the subject or body of the message. http://www.ae.iitm.ac.in/mailman/listinfo/ilugc
