- May 23, 2007
- 95
- 7
- Home Country
- Germany
Hello all,
I'm using MP now for almost 3 years and went through many funny problems. A good addition IMO was the watchdog and I propose to extend it's functionality. Working in the automotive industry I see that the SW working in our cars most of all has to be reliable and in future also safe. The HTPC is also a system which has to work very reliably. However, 'normal' PC's are often not that reliable because either the HW or the SW or both with the time run into tricky configurations where they get stuck under seldom circumstances.
Rationale:
Besides of creating stable SW a watchdog can easily improve the reliability of the system 'HTPC'. The user shall get diagnostic information if the watchdog was active.
Current situation:
Currently the MP-Watchdog monitors the existence of the process of MP. As soon as it gets killed, the watchdog is activated, collects information and restarts MP.
Proposal:
Extend the MP-Watchdog to monitor also:
Benefits:
Kind regards from Stuttgart
I'm using MP now for almost 3 years and went through many funny problems. A good addition IMO was the watchdog and I propose to extend it's functionality. Working in the automotive industry I see that the SW working in our cars most of all has to be reliable and in future also safe. The HTPC is also a system which has to work very reliably. However, 'normal' PC's are often not that reliable because either the HW or the SW or both with the time run into tricky configurations where they get stuck under seldom circumstances.
Rationale:
Besides of creating stable SW a watchdog can easily improve the reliability of the system 'HTPC'. The user shall get diagnostic information if the watchdog was active.
Current situation:
Currently the MP-Watchdog monitors the existence of the process of MP. As soon as it gets killed, the watchdog is activated, collects information and restarts MP.
Proposal:
Extend the MP-Watchdog to monitor also:
- unresponsiveness of process of MP
- unresponsiveness or killing of process of TV Service
- unresponsiveness or killing of process of MySQL Service
- Status Information on some screen
- other usual notification mechanisms for server monitoring
- kill/restart service/application
- reboot PC
- shut PC down
Benefits:
- 'Hangs' due to SW and partially due to HW faults could be pragmatically resolved
- The reliability of the system would increase even in the diversified landscape of HTPC implementations
- One could go to vacation and leave the system safely behind
Kind regards from Stuttgart