Hi Reinier, On Tue, 22 Sep 2009, Reinier Battenberg wrote:
> I have apcupsd running as a deamon and the system will in worst cases > shutdown when there is X amount of time left in the batteries. The BIOS > is configured to start up when power returns. Problem is: when power > returns while the system is shutting down, the system seems to stay off. > The BIOS never gets an event, so just shuts down the computer. Shutting > down takes about 3 minutes, which is quite a lot. So, first question, > how do you tell a shutting down machine that power is back, and it is > not shutting down anymore, but rebooting? You need to use a UPS daemon that can tell the UPS to initiate a timed (delayed) shutdown, and that does this automatically when it initiates a software shutdown. This will cause the UPS to force a power cycle of the server, bringing it back online. NUT does this, for example. > Secondly, we dont want these reboots. I can't buy an inverter with > proper batteries, so I bought a second UPS ( RS 1500) , and placed it in > series with the first one. Its a bit cheaper, so i placed it before the > SmartUPS, and keep that one connected to the server over USB (and AC). > > This new UPS is not doing its job. If power goes off, it does switch to > batteries, but the Smart UPS that it is feeding in to, immediatly > switches to batteries, too. Chaining UPSes is generally considered a bad idea. APC recommends against it. Most likely, the power quality emitted by the first UPS is not enough to satisfy the second that the mains power is OK, so the second switches to battery too. > I ran a few tests: > > With a disconnected RS1500 (no AC in) > - Connect the RS1500 to a moniter, and the monitor works. > - Connect the RS1500 to the second power-unit of the server. The led light on > the server goes on, but on the server-side the power cable makes a funny > sizzling noise. Nice :) Probably the server's PSU has little or no PFC and so its current draw is not proportional to voltage, causing a huge inductor in the UPS to vibrate violently (and eventually self-destruct or tear itself off the motherboard). > - Connect the RS1500 to the Smart-UPS, the smart-UPS does not charge from the > RS1500 batteries, and you do hear the sizzling sound on the Smart-UPS side. > Also, the Smart-UPS indicates it has lost its power-source and it is running > from batteries. > - The orange LED is lit on the RS1500, which means, we are on battery now. > There is no indication there is an overload. That's because the Smart-UPS is not pulling any power from it. > Plug the RS1500 in the grid > - Light goes green, and the sizzling sound disappears. > > Question: is my RS1500 actually helping my Smart-UPS to feed the server with > power? (and do i just need to add more batteries?) Or do i have a problem? No and yes. My advice would be to rip the batteries out of the RS 1500 and chain them in parallel with the batteries of the Smart-UPS, by hanging wires out of the side of the UPS case. Cheers, Chris. -- Aptivate | http://www.aptivate.org | Phone: +44 1223 760887 The Humanitarian Centre, Fenner's, Gresham Road, Cambridge CB1 2ES Aptivate is a not-for-profit company registered in England and Wales with company number 04980791. _______________________________________________ LUG mailing list [email protected] http://kym.net/mailman/listinfo/lug %LUG is generously hosted by INFOCOM http://www.infocom.co.ug/ The above comments and data are owned by whoever posted them (including attachments if any). The List's Host is not responsible for them in any way. ---------------------------------------
