We are using lm-sensors in an embedded system, and have noticed that when errors occur, sensord exits. We occasionally encounter SMBus collisions which cause transient errors on SMBus reads by the sensor chip driver. We modified the most recent w83793 driver (which is much improved in dealing with SMBus issues) to return cached data for up to 30 seconds in the case of SMBus errors, and then to return EAGAIN on the sysfs file read if the SMBus errors persist. However sensord exits when it gets EAGAIN. We have patched sensord not to exit on errors, but instead to log them and continue. This seems to us to be the robust behavior for an important system monitoring daemon - if the error is transient it rides it out, and if it is permanent it complains to get attention. It's about a 2-line change in the main loop of sensord. I'd like to push these patches upstream, but wanted to first see if it is acceptable to change the behavior of the w83793 driver and sensord in this way. -Andy
