>> There's a slim possibility that the timestamp could be useful in the >> upsc output, especially with drivers that have to cope with buggy >> hardware that does not always respond in the same time interval. > That's exactly the case I was thinking about; I'm coping with it now.
You'd beat the server to death by adding this, since it will receive an update for every poll of the UPS data, instead of just the changes (like it is now). In servers that are monitoring many drivers, that means a lot of unneccesary overhead. I'm clearly not in favour of such a change. You need logging in the driver, not in upsc here. Communication problems with the UPS hardware will already surface when someone is trying out a driver for a UPS, it makes no sense to require upsd running in that case. You'll want more information too than just a timestamp, since it makes a lot of difference if the communication is non-existant or is incomplete. I admit that many drivers are lacking in this aspect, we definitly need to improve on this. Not only in the amount of debug information given, but also what information should be available at specific debug levels. In general, at the highest level the full communication between UPS and driver should be made available. Best regards, Arjen _______________________________________________ Nut-upsuser mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
