tech failures
Over the last two weeks we’ve had some tech failures in the computing area here at Starlight Cascade.
First we investigated a very loud noise coming from the observatory one evening. It was the allsky1 desktop computer system fan running faster and louder than we’ve ever heard it before. Pulling the power plug and restarting it fixed that problem, but the system would not come back up and appear on the network. We don’t have a monitor attached to it so could not diagnose it until it was disconnected and brought back into the house.
Turns out the old Dell machine had a dead motherboard BIOS battery and required human intervention to restart after a long power outage. We didn’t seem to have any of those about so left it plugged in for a time, then unplugged for 5 minutes to see if it would come back up again without problems. It did, so out it went back to the observatory, plugged back in, restarted ok and all is well.
We missed out on one nights worth of imaging, but it was cloud and rain so that was OK.
It was critical that the system gets back up and running as:
(from http://www.amsmeteors.org/meteor-showers/meteor-shower-calendar/)
Geminids Active from December 4th to 16th 2013 (Peak night Dec 13-14)
The Geminids are usually the strongest meteor shower of the year and meteor enthusiasts are certain to circle December 13 and 14 on their calendars. This is the one major shower that provides good activity prior to midnight as the constellation of Gemini is well placed from 10pm onward. The Geminids are often bright and intensely colored. Due to their medium-slow velocity, persistent trains are not usually seen. These meteors are also seen in the southern hemisphere, but only during the middle of the night and at a reduced rate.
Radiant: 07:28 +32.2° – ZHR: – Velocity: 22 miles/sec (medium – 35km/sec) –
Parent Object: 3200 Phaethon (asteroid)
The Radiojove system runs on a 4 year old Toshiba Netbook. It’s wall adapter power supply has had repeatable issues of failing after a power failure. So a new $20 power supply arrived last week and was installed and all seemed well. Then yesterday the system drops off the network. arrg.
Investigating in -7C temperatures showed it in a continual reboot loop. A few minutes or power resets got it back into the operating system of Windows XP and on the network, but still acting very slow. It was not until a few hours later that we could remote desktop into it and that showed a lot of hard drive errors. It shouldn’t happen on a system only a few years old, but then again it has been exposed to severe temperature conditions over its lifetime outside. It has a 250GB sata drive that about 60GB is in use. Maybe it is time to think about replacing that drive with a Solid State Disk drive? a 64GB drive is about $100. Should keep it running for a few more years.
So we missed maybe 12 hours of data from this system but things seem to be going well for now.
Later this weekend we’ll take it offline, run some hard drive tests and checks, clean out more of the cruft on the drive, defragment it and see if we can shoehorn the remainder onto a 64GB SSD.
So far the jury says that SSD should last longer than spinning drives until normal use.
We have another 120GB SSD in service now for about a month and it is running well. Faster boots when necessary, slightly faster starting programs, pretty normal operation in general. We suspect that 2GB on a windows 7 machine is a little on the low side as we are getting memory swaps to the drive (now faster with the SSD than the old spinning drive, but still slower than not doing it at all).
Looks like a memory upgrade is in the works for that system. 2x2GB SODIMM DDR2 memory runs about $80.