The Register carries the funniest, most topical IT story of the year: ‘Facebook’s first data center ran into problems of a distinctly ironic nature when a literal cloud formed in the IT room and started to rain on servers. Though Facebook has previously hinted at this via references to a ‘humidity event’ within its first data center in Prineville, Oregon, the social network’s infrastructure king Jay Parikh told The Reg on Thursday that, for a few minutes in Summer, 2011, Facebook’s data center contained two clouds: one powered the social network, the other poured water on it.
Someone had better explain to Facebook that having an actual cloud in the data centre isn’t really what cloud computing is all about!
That said, this has happened before. Boeing’s Everett factory (initially constructed to assemble the 747, and now assembles all of their widebody jets) used to have clouds forming near the ceiling, before the installation of a specialised air circulation system sorted that little problem out. Evidently, some are destined to repeat history’s mistakes…
And back to real life: here in Cape Town, winter has come. (Apologies to the Starks.) It’s at this time of year that those of us living here realise that we’re on a tiny peninsula on the southern tip of Africa with nothing between us and the Antarctic.
While Tim (my housemate) posted on how his office park became a raging torrent, I came across photos on how the persistent winter rain caused a section of Lover’s Walk (just below the university) to collapse.
Dammit — I used to stay right around the corner from there…
The nerd community had a bit of news earlier today: a meteor streaking across the skies of Russia. Since everyone in Russia seems to have a dashcam (apparently it’s a car insurance requirement over there due to police/judicial corruption), we even got footage of it:
While the meteor didn’t cause any damage itself, the sonic blasts were significantly more destructive; current reports seem to indicate around 400 injured, mainly from broken/falling glass. Not surprising: the sonic wave arrived ~30 seconds after the meteor passed over, and a lot of people would have been at various windows having a look at what had just happened (and in their defence, if I saw a massive flare like that, with a beautiful contrail in its wake, I’d be out there as well trying to get photos for this blog!). It’s similar to tsunamis: most fatalities happen to people who chase the receding sea, unaware that the reason for the receding waters is that they’re about to un-recede in a rather spectacular and destructive way.
(You may want to turn your speakers/headclamps down for the sonic blast videos, unless you either enjoy extremely loud bangs or you’d like a crash course in Russian swear words…)
Of course, being Russia and therefore having Russian temperatues, blown-out windows are a serious problem in winter, so hopefully the emergency services there are more jacked up than ours.
The big question here is: is this related to the asteroid 2012 DA14? For those of you living under a rock, it’s an asteroid with an estimated diameter of 50 meters, an estimated mass of 190,000 metric tons — and it will be passing approximately 27,000 kilometres from Earth’s surface in a few hours (19:25 UTC, 21:25 South African time). Over on Bad Astronomy, Phil Plait seems to think that it’s totally unrelated:
For one thing, this occurred about 16 hours before DA14 passes. At 8 kilometers per second that’s nearly half a million kilometers away from DA14. That puts it on a totally different orbit. For another, from the lighting, time of day, and videos showing the rising Sun, it looks like this was moving mostly east-to-west. I may be off, but that’s how it looks. DA14 is approaching Earth from the south, so any fragment of that rock would also appear to move south-to-north.
Not necessarily. Imagine a basketball in front of you. That is the Earth. Now draw an imaginary line from your nose to the left side of the basketball. Your line is going east to west. Now draw another line to the right of the basketball. Your line is now going west to east. Same point of origin. Same basic direction of movement. Different perceived trajectory for those living on the basketball.
Information on the meteor is still rather sketchy, although we may get a better picture in the next few days — it appears that the meteor broke up in the atmosphere and may have rained down some chunks on the ground. If true, and if and when those chunks are found, we may learn more.
For now, I’m going to have to end off with this image by EUMETSAT, showing the vapour trail:
If you’re ever taken the time to read the crap on the About Me page, you’ll know that I co-founded the FRAG LAN back in 2007. Sadly, since I left Durban, it appears that the event has slowly gone to the dogs (as evidenced by an article on Do Gaming).
Of course, I don’t want to see something that I started disappear like that, so I popped on to their forums (after a good deal of time and effort trying to remember what my password was!) to offer some advice. I’m not sure how long the post will stay there though, so I’ve reproduced the contents below:
Right.
First off — long time, no see folks. To answer the most obvious question, yes, I’m still alive, Cape Town is treating me well, and Web Africa offering uncapped is definitely no longer “unsustainable” (lead developer on that project).
Now that the pleasantries and formalities are out of the way, on to business.
I haven’t kept in touch with what’s been going on with the LAN since my move to the non-ANC province, but a recent article on Do Gaming pretty much told me the entire story: in the interceding years, the LAN has pretty much been run into the ground and currently has an extremely uncertain future. Now, I’m not here to bash anyone in particular (well, too much!), mainly to offer my own input and suggestions on how to get out of the hole that the LAN has dug itself.
I can pretty much sum up the below simply as follows: get the LAN back to its roots.
The LAN pretty much started out as a small 5-man event in my parents’ double-garage in 2007, and we decided to explore expanding it further. A few months later, it had outgrown both our imaginings, and the resources of those involved; that’s when James came onboard and took it further, and eventually took it over. The only problem was that James seemed to be far more interested in the bottom line than gamer satisfaction, and I guess you all know the rest. (Even I struggled working with him at times, although to be fair, my opinion on things was always taken into serious consideration.)
However, the great thing about those early days was the camaraderie that existed between the event organisers and attendees at the time. The early goal was simply to provide a social event for people to just go and enjoy themselves for the weekend, and we managed that really well. It was easy for us because there was nothing else at the time — UberLAN had recently shut up shop, and Vendetta started at roughly the same time (and went for a little while until James managed to sabotage that effort). From the sounds of things, that’s all but disappeared, with the organiser goals and attendee expectations being totally at odds. That needs to change.
I don’t buy the opinion that “LAN gaming is dying”; despite the always-online requirements of this day and age (which I don’t necessarily approve of, but that’s a different story), there are still times and places for us folk to pack up our PCs, head off somewhere for a weekend, and kill players with the gauntlet while holding the BFG (yes, I was on the receiving end!). But it needs to be done right. If you need an example to model yourselves off of, one need look no further than Organised Chaos in this part of the world — the primary objective there is for the attendees to have a great time, everything else is secondary. (I have contacts within OC management, so if you’d like me to facilitate any kind of knowledge sharing, I’d be happy to help.)
I do think that Doug has the right idea; he’s trying to focus on becoming more gamer-centric while weeding out the seedier elements and getting the event back to what it should be. Thing is, he is going to need your help — if you guys want LAN gaming in Durban to survive, it’s up to the entire Durban gaming community to seize the initiative and get things back to where they should be. It’s going to be a long and hard process, but if everyone pulls in, it’s definitely possible.
Over to you guys.
Carpe Diem.
UPDATE: I’ve had a few folks contact me for clarification about the Vendetta sabotage part. My only response was to point them in the direction of the embrace, extend and extinguish Wikipedia article.
An announcement posted on the ZA NiC page states that no new .za.net or .za.org registrations will be accepted after October 31, 2012.
A (blatant) copy/paste explains their reasoning:
The primary reason for starting this project in 1998 was to provide an affordable alternative to the domain name prices of the time. Back in the late ’90s a .com domain name could easly set you back as much as $100! Today the Internet is a very different place with a .com domain being available for as little as around $8. Even in the primary market we served (South Africa), co.za domains are just R50 per year and org.za domains are free for non-profit organisations.
Unfortunately just as the Internet has become a better place (cheaper!) it has also become a worse place in the last 14 years. Today the admin overhead of managing domain names registered purely for spam or scamming purposes has made running a free, community serving project such as ZA NiC just not fun anymore. And that is our secondary reason for ceasing new registrations.
Just because we will no longer be taking new registrations, it does NOT mean any existing domains will cease to function. Domains registered before 31/10/2012, with valid and responding nameservers, will continue to be served by us indefinitely. The modification system for keeping contact and domain name details up to date will also continue to work as it does today.
Despite the statement that existing domain names will be left alone, the announcement means that the future availability of this site’s domain name in the future is no longer guaranteed. As such… any suggestions for an alternative name, just in case? (Some research indicates that the .nom.za domain namespace is designed for personal names of South African entities, so something in there would be my first choice.)
So, Development was (for the most part) working away nicely on a Friday afternoon, when we noticed a plume of smoke outside the window. Of course, the thought of a field trip was irresistible, so we all climbed the fire escape to the building roof, and noticed this:
Yes, that’s a veld fire in the District Six area. Obviously, we don’t know what caused it, but as everything in Cape Town is extremely dry after some pretty intense heat that we’ve been having lately, it probably wouldn’t take much for a cigarette end carelessly flicked out of some motorist’s window to set that off.
Of course, this being Cape Town, the wind blew the fire across the road:
Now, for those unfamiliar with the area, just behind that other building in the office park lies the N2 freeway. Hate to think what visibility is like for those motorists right now.
And then of course, the smoke plume gets blown over towards the CBD:
Two things to consider: the aforementioned visibility problem on the N2 (which you can’t even see in the above shot!), and the fact that it’s now 15:30 and people are starting to bugger off from work.
Rush hour this afternoon is going to be FUN.
(At least I leave the office at 18:00 and drive AWAY from the fire…)
Disk manufacturing sites in Thailand — notably including the largest Western Digital plant — were shut down due to floods around Bangkok last week and are expected to remain shut for at least several more days. The end to flooding is not in sight, and Western Digital now says it could take five to eight months to bring its plants back online. Thailand is a major manufacturer of hard drives, and the shutdowns have reduced the industry’s output by 25 percent.
Western Digital, the largest hard disk manufacturer, makes more than 30 percent of all hard drives in the world. Its plants in Ayutthaya’s Bang Pa-In Industrial Estate and Pathum Thani’s Navanakorn Industrial Estate together produce about 60 percent the company’s disks. Both were shut down last Wednesday. (Western Digital also has a major plant in Malaysia that hasn’t been affected by the floods, so some production will likely shift to that plant.)
Fourth-ranked hard-disk manufacturer Toshiba makes more than 10 percent of the world’s hard disks, and half of its capacity is in Thailand. Toshiba’s plant has also been closed due to flooding.
Key disk component suppliers have also been hit. Nidec, which makes more than 70 percent of all hard drive motors, has temporarily suspended operations at all three of its plants in Thailand, affecting 30 percent of its production capacity. Hutchinson Technologies, which makes drive suspension assemblies, has also suspended operations due to power outages, although it says it will shift operations to its U.S. plant.
Seagate, the second-largest hard disk manufacturer, has two plants in Thailand, but neither is in the flooded parts of the country. Seagate notes that “the hard disk drive component supply chain is being disrupted and it is expected that certain component in the supply chain will be constrained.” Translation: Component prices are going up, at least for some parts.
Economics 101: if supply is constrained and demand remains constant, prices are going to shoot up. This has already happened now.
So with hard drive production being affected so heavily, what does that mean for us consumers? We run the possibility of seeing high prices and limited availability at one of the most important times of the year. Indeed, some retailers like Newegg and NCIX have already begun limiting the number of drives a customer can purchase (Newegg is currently at a limit of one drive per customer while NCIX has announced a limit of two) as they gear up for supply shortages and price increases throughout the channel. Speaking of price increases, we have seen a spike of 15% to 30% in the cost of some models over the last 72 hours.
This couldn’t have come at a worse time since Christmas shoppers will see higher prices on everything from notebooks to PVRs. Flash memory isn’t safe from this turbulence either so expect its prices to increase as consumers begin looking for alternatives to HDDs. So gear up folks because the days of cheap storage space are about to end….for a few months at least. Our hearts and prayers go out to the people suffering through this disaster.
And I’ve confirmed that this has hit our local suppliers as well. I priced a rig for a friend of mine in mid-September, and quoted a 1TB Western Digital HDD (from one of South Africa’s more reputable online stores) for R 864.12. That same HDD, from the same store, is currently R 1,127.46, and that price is not going to drop anytime soon.
To make our lives even worse, ASUS has confirmed that they only have HDD inventory until the end of November. This has the potential of affecting their supply of notebooks, desktops, etc. – which could result in those prices rising as well. We could well see more knock-on effects elsewhere as well.
Cape Town disappeared off the electricity grid at around 11:00 today. We’re powered by some fairly sizeable generators (when you’re in the ISP business, you can’t let a measly city-wide power outage break your stride), but it was interesting following the reports of the outages and the rumours surrounding it while it was all going on.
And I’m going to start by dispelling the most common one: Koeberg (our nuclear power plant about 40 km up the West Coast) did not do a Chernobyl/Fukushima on us. Although most of Cape Town went down (in particular, the CBD, Cape Flats, Southern Suburbs and South Peninsula went down), the northern and West Coast areas (Bellville, Durbanville, Table View, Bloubergstrand et al) stayed up. Trust me, if something ever happens at Koeberg, those guys are the ones who will be the first to know about it – one way or the other.
According to several reports on our local news sites, the issue was with the Eskom supply line from the Muldersvlei area. From what I’ve heard, there are two supply lines: one was down for maintenance, and while this was going on, the other one tripped. This isn’t first-hand information, and the South African press is slightly notorious for getting facts wrong, so I’d take this with a slight pinch of salt – that being said, it’s entirely plausible to me.
As I type this, power to the CBD has been restored, we’re off the generators, and other parts of the city are either back online or are being brought up. I should be fine for World of Warcraft raiding tonight then. (I hope!)
For those interested in the geeky side of the power failure: Andrew Trench put together a map of the power failure as it was mentioned on Twitter. It’s quite interesting viewing, and just shows one the impact that social networking has had on our lives (if the Arab revolutions earlier this year weren’t proof enough).
