S80_UK

I'm travelling to the US next week and already looked up store locations - but they're too far off the current travel plan to be a viable option.

To be honest, I doubt anyone else can figure it out either unless you tell us what motherboard and/or chipset you have and what the drives are. At the end of the day, 1.5Gb/s is still faster than the drive can read or write the data to the disk surface and faster than a 1Gb/s network, so it's often not a big deal.

Hi PeterB I am in the exact same position but with a Windows 7 machine slaved from my unRAID box which is connected to my APC Back-UPS 800. unRAID detects the UPS and will power down cleanly. The Win 7 machine is running apcupsd as a service and correctly detects apcupsd on unRAID via the network and reports the UPS status. It also pops up balloons when then the mains / UPS status changes, but I cannot figure out what needs to change in the configuration to force the Win 7 box to shut down. I have tried changing some timings but to no effect. The Win 7 box sometimes runs unattended and so a clean shut down would be nice. I am sure it's possible - I am just missing a couple of steps. If you find anything please share - I will do likewise. Les.

Is the WHS also getting a 1Gb/s link? I once had a Netgear gigabit switch that would not play fair with one particular motherboard and would drop back to 100Mb/s when the PC was rebooted. If I power cycled the switch it would come good again. Not saying your switch or router is at fault, but like others I also think this is most likely a network issue.

Or save as "//tower/flash/config/Plus.key" if that was its name. Do not change the name of the key from that of the file as attached to the email. And if you are using Windows, the slashes will be backslashes - so in Explorer you would use the path "\\tower\flash\config\Pro.key" for example.

Wow! Here we start paying VAT, our sales tax (at 20%) above £18 (AUD 28) and £40 (AUD 62) for gifts. Import duty comes in at £135 and up. We can at least by some Supermicro boards from a reliable UK reseller.

Is the drive that's spinning up also containing the data being streamed? If so, then the split-level settings are there to allow the data to be placed all on one disc as needed to prevent that problem. If the drive is spinning up because it contains some other data that is needed, perhaps by another user, and if that is then interrupting the streaming, that would suggest a lower level problem within unRAID or its settings. Certainly with my server I can cause a drive to spin up (for audio streaming) while I am serving video to XBMC somewhere else and the video streaming is not held up during the spinup time of the audio drive. I would therefore think that it is the split-level settings that you need to look at. Of course, if you make changes then data already on the server will not be moved - you'd need to manage that separately.

Here in the UK import duty, VAT and the admin fees negate most advantages. And the shipping that Superbiiz charge would more than wipe out any remaining benefit. I didn't think it was possible to charge so much for shipping simple things like graphics cards. I'll be sticking with local suppliers for the most part.

As I understand it, with the high-water setting, the third disk will start to be used when the second has passed the 50% mark, unless you set the thresholds any differently.

I am not aware of a confirmed list. I have Back-UPS 800 which works very well (detects mains fail and restore and will shut down cleanly and restart for longer outages). Status report looks like this... UPS Status (from /sbin/apcaccess status) APC : 001,045,1145 DATE : Sat Apr 09 09:15:56 UTC 2011 HOSTNAME : Tower RELEASE : 3.14.3 VERSION : 3.14.3 (20 January 2008) slackware UPSNAME : Tower CABLE : Custom Cable Smart MODEL : Back-UPS BR 800 UPSMODE : Stand Alone STARTTIME: Sat Mar 26 23:48:13 UTC 2011 STATUS : ONLINE LINEV : 246.0 Volts LOADPCT : 9.0 Percent Load Capacity BCHARGE : 100.0 Percent TIMELEFT : 66.5 Minutes MBATTCHG : 10 Percent MINTIMEL : 5 Minutes MAXTIME : 300 Seconds OUTPUTV : 230.0 Volts SENSE : High DWAKE : 000 Seconds DSHUTD : 000 Seconds LOTRANS : 194.0 Volts HITRANS : 264.0 Volts RETPCT : 000.0 Percent ITEMP : 29.2 C Internal ALARMDEL : Always BATTV : 27.1 Volts LINEFREQ : 50.0 Hz LASTXFER : Low line voltage NUMXFERS : 3 XONBATT : Sat Apr 09 01:23:20 UTC 2011 TONBATT : 0 seconds CUMONBATT: 110 seconds XOFFBATT : Sat Apr 09 01:23:28 UTC 2011 SELFTEST : NO STATFLAG : 0x07000008 Status Flag SERIALNO : 3B0951X24834 BATTDATE : 2001-09-25 NOMOUTV : 230 Volts NOMINV : 230 Volts NOMBATTV : 24.0 Volts NOMPOWER : 540 Watts FIRMWARE : 9.o5 .I USB FW:o5 APCMODEL : Back-UPS BR 800 END APC : Sat Apr 09 09:16:48 UTC 2011

Yeah - just saw that and edited my post as you entered yours. Thanks for the correction.

One fact, apparently not well known, is that APC do not use standard pinning on their connectors, so a regular null-modem cable will NOT work. I cannot check it myself (my UPS has USB and serial on a 10 way RJ style connector) but the pinout is given here... http://lists.alioth.debian.org/pipermail/nut-upsuser/2005-August/000118.html http://www.everythingtech.net/2009/04/apc-smartups-serial-cable-pinout/ Edit: Not sure if this helps having just re-read the original post and link

Me too. I would be interested enough to put together a second box with a basic unRAID just to try this stuff out. (I don't want to touch my regular server - it's not experimental.)

My solution to that dilemma has been a DOS bootable USB stick. Not too hard to do (I googled it at the time). There's something quite cool about a DOS boot device that's as fast as a stick and can hold a gig or so of files. Of course, I still have a box of floppies in reserve.

.. and even then, the spec sheet doesn't state whether that value relates to the optional 'low start' setting, or the standard, default, setting. From what I have read, this would be standard - as I read it, the low-start setting lowers current down to about one amp peak.

That's cool. I need to sharpen my Photoshop pencil...

Personally, I would rather not rely on someone else's measurements - they may be right some of the time, but we have no idea of the test conditions. WDC give there figures here - http://www.wdc.com/wdproducts/library/SpecSheet/ENG/2879-701229.pdf These are quite a lot higher. They may not always be seen in use, but these would be a safer bet if budgeting for a power supply.

That's true. I had forgotten that unRAID may cause all drives to spin up if it is being shut down.

I came; I saw; I read; I understood; I decided I wanted remote access without another machine running... Still looking for the best way to do this. And I'm not enough of a Linux guru to do it all myself at this point.

Mine doesn't give the peak figure. This is what I can get... UPS Status (from /sbin/apcaccess status) APC : 001,044,1087 DATE : Mon Apr 04 20:48:57 UTC 2011 HOSTNAME : Tower RELEASE : 3.14.3 VERSION : 3.14.3 (20 January 2008) slackware UPSNAME : Tower CABLE : Custom Cable Smart MODEL : Back-UPS BR 800 UPSMODE : Stand Alone STARTTIME: Sat Mar 26 23:48:13 UTC 2011 STATUS : ONLINE LINEV : 244.0 Volts LOADPCT : 8.0 Percent Load Capacity BCHARGE : 100.0 Percent TIMELEFT : 68.9 Minutes MBATTCHG : 10 Percent MINTIMEL : 5 Minutes MAXTIME : 300 Seconds OUTPUTV : 230.0 Volts SENSE : High DWAKE : 000 Seconds DSHUTD : 000 Seconds LOTRANS : 194.0 Volts HITRANS : 264.0 Volts RETPCT : 000.0 Percent ITEMP : 29.2 C Internal ALARMDEL : Always BATTV : 26.9 Volts LINEFREQ : 50.0 Hz LASTXFER : No transfers since turnon NUMXFERS : 0 TONBATT : 0 seconds CUMONBATT: 0 seconds XOFFBATT : N/A SELFTEST : NO STATFLAG : 0x07000008 Status Flag SERIALNO : 3B0951X24834 BATTDATE : 2001-09-25 NOMOUTV : 230 Volts NOMINV : 230 Volts NOMBATTV : 24.0 Volts NOMPOWER : 540 Watts FIRMWARE : 9.o5 .I USB FW:o5 APCMODEL : Back-UPS BR 800 END APC : Mon Apr 04 20:49:21 UTC 2011 So I can spin up all the drives and get a figure, but it's not the peak start up power.

It's worth adding that some motherboard BIOS settings allow delayed or staggered spin up of hard drives when booting. Also I have seen some drives that have a link option that delays or inhibits spin up when power is first applied. In either case they will still spin up normally in response to system commands. These measures can significantly lower the surge requirements when a large number of drives are used. Also worth noting that the popular WD green drives do quite well - WD20EARS has a 1.55 amp requirement. WD also have a software tool to allow a slower spin up to be configured in a drive's BIOS in order to reduce the current further. data here... http://www.wdc.com/wdproducts/library/SpecSheet/ENG/2879-701229.pdf spin up control utility here (I have not tried this myself yet) http://support.wdc.com/product/download.asp?groupid=609&sid=114〈=en some

There is a slight risk that the read amplifiers (which take the raw signal from the heads) would have been damaged. They are generally mounted with the head assembly (inside the drive enclosure) in order that they are as close as possible to the heads themselves and therefore they would not be fixed by changing the main board. However, it's all speculation at this point - they could just as easily have been protected by the other electronics failing first. B1G (the OP) needs to decide which options are viable relative to the value of the data.

I find that puzzling! The black wires are zero volts, or ground, and the connections are usually commoned at both ends - I don't understand why crossing the two black wires should have caused any problem. I have to suspect that something else happened. Swapping one of the black wires with one of the coloured wires ... now, that could easily cause a major problem. In any case, the controller board may include fuses on the power lines - try following the tracks and see where they go. As I read it the swap (or not) is between a Molex 4-way disk drive connector (yellow-black-black-red) and a SATA power connector (should be [optional orange]-black-red-black-yellow). I am guessing one block of SATAs may have been wired in the same sequence as the Molex connector.

Eeek! Most likely the electronics are fried. The suggestion of swapping the electronics with that of a donor drive is a good one, but you clearly need to match the model and version of the drive as closely as you can - and you need to be comfortable with the work involved - some drives bury some of the connections under plastic resin to ensure mechanical security.

More than that if the drives are spinning at the time, apparently... For a disk of mass, M, and radius, R, the moment of inertia, I, is I=MR^2/2. If the disk is spinning with angular speed, ?, its rotational kinetic energy is K.E. = I?^2/2. The angular speed of a disk is related to the speed (VD) of its outer edge by, ? =VD/R. Combing the above equations, the kinetic energy of the disk can be written as, K.E. = MV^2/4. The kinetic energy of a falling mass (m) with speed, v, is, K.E. = mv^2/2 The total kinetic energy of the disk and falling mass is the sum of the individual kinetic energies. The statement of the conservation of energy now reads, mgh = MV^2/4 + mv^2/2 ...with condolences to the OP and his wife.