IP7000 random watchdog resets

Paul Murphy

Re: IP7000 random watchdog resets

I can confirm that our systems have been stable since we reset the NTP resync to 86400.    With 4 units, the last crash was on 29th April, where previously this was happening two or three times per day in some cases.  Our reset interval was 3600 (1 hour)when the problems were happening, and coincidentally active calls which lasted more than 1 hour almost always resulted in a crash around that time.


As an aside, the reason for NTP on any system is that accurate clocks are expensive.  Clocks drift, and we keep them on track by synchronising regularly to a reliable source, so a sync interval of 1 day will only be useful if the system can keep the time accurately over that period.  I've yet to find anything costing under $500 which can reliably do this, so it is normal practice to drop the time to a few hours.  As we can't query the unit's time remotely or view it on the web console, I have no idea how accurate the IP7000 is - SNMP support would be useful on these systems for so many things:  uptime, call count, call duration, network stats, etc...



Message 11 of 11