Posts by Benno 71 publicly visible posts • joined 22 Oct 2013
MS have shifted the goal-posts after the game has started - The system requirements for "Windows 10" are now *not* the same as they once were. So what legal standing does one have now that their system, which clearly met the original system requirements, now does not? It's no longer a forced software upgrade (via GWX), it's a forced hardware upgrade.
Such is the side effect of the current versioning & release methodology.
In my part of the world, they're called 'time points', but the rules are somewhat the same (dare I say almost an international standard?!). A few minutes late is ok, but a minute early is most certainly not.
When things go wrong and vehicle needs to go off-route, or cut in/out due to breakdowns, accuracy can go a bit pear shaped. There can also be issues with prediction resolution at the start of trips (due to insufficient _relevant_ data from that vehicle at the time).
And the thousand other 'edge cases' - not including a full API failure!
Yeah, I run a public transport real-time tracking system as part of my day job - no it's not with TfL :P
IE? Oh, that unused browser on the air-gapped system?
Most of my 2k8r2 boxes are still running IE8. But what justification for downtime do you provide to _upgrade_ a non-removeable, unused product on an isolated system (when the only interactive use is viewing the console to keep an eye on boot progress)?
Also, when is running an n-x OS that's fully patched, not patched?
See the source article...
(lies, damn lies, and statistics)
If we are taking about dimwits that have exposed SMB (which was a bad idea in the 1990's FFS...), there are a few issues with the theory as it stands;
By default, Windows will only allow file share access when in private firewall mode.
Consumer grade modems don't have any port forwarding enabled (again, by default) - even that flawed facility, UPnP, didn't tend to dynamically allow it.
A large number of ISP's will automatically block such ports for their subscribers (unless you request them to be open).
Very, very, very badly configured VPN's or shockingly bad gateways seem to be the most likely vector. If you had, or still have a private network with internal DNS, SMB, NMB etc. exposed, then I suggest you change careers voluntarily, before you are lynched...
You can have multiple vCenter (grrr US spelling...) servers, but you'll pay for it.
I used to have a dedicated physical VCS v2.0 system (yes it was a while ago!), but the egg/basket equation is just nasty if the box completely dies and is offline for longer than planned due to spares availability, smoke etc...
If you have it on an ESXi cluster, you can vMotion it around during host maintenance. If you have an unplanned host outage, you can manually delete/re-add it into inventory on another host (or go through the CLI host-ownership rigmarole if it really goes south). The inventory delete/re-add is also the method required for cold migrations.
(shared storage assumed)
Right now I need to do the whole boot storage controller architecture change on our vCenter box, which means killing the host management agents to 'break' host connectivity with vCenter (the software), in order to change the vCenter (virtual hardware).
Fun times!
Be careful there - lots of kit will happily talk to AD (via RADIUS), but will send the credentials in plain text only (they simply don't support encrypted logons). I've got into the habit of having specific credentials for the RADIUS authentication that have no 'normal' AD user/admin privileges. You don't want your normal login at risk of compromise as a result of trying to improve security!
Not only that, quite often you're searching for ages just to find out how you need to configure the RADIUS server so the remote kit actually authenticates properly at all.
I had a MS staffer demo this to me a few months back - it didn't work...
It turned out to be yet another immature, over-hyped 'feature', that had to be vaguely excused at the time.
It didn't start off well when he asked who had a win phone, and no-one put their hand up.
Then there was the dongle that was required to make it work!
Quite some time ago, the number of legitimate help-phone use in a jurisdiction I was involved in was 3%.
Not saying that the 3% should go without, but the number of people that thought that help-phones were an information intercom, or just let their kids push the duress button was astounding.
I'm all in favour of getting rid of the freeway phones - _If_ there is a 'better' option.;
The strategy taken by VicRoads sounds ok to me.
Heck, even my AMEX card is contactless...
I've been using contact & contactless for as long as they've been available in Australia, no fraud yet!
(contactless is a $100 limit without PIN here in AUS)
Some terminals even allow contactless with PIN for values over $100.
Although some terminals don't like the AMEX, and I have to go chip 'n' PIN - bah, old-school :)
Exactly - I've had a G4 for about a year, and it's bloody marvellous. With some effort I've managed to use 50% of the battery life in a day, but the thing was almost constantly in use to do so (i.e calls and business work, not social crap). Generally I go home with ~%75% charge, and it will use about 2% overnight if I forget to plug it in. Camera is great, and I even managed to put a Defender case over the leather back. Bought it from Kogan unlocked and currently running Android 6.0 (Jan patch tho...)
Best smartphone I've used - hands down.
Back in the late '90's I had a staff member accidentally ran a ghost multicast onto an entire subnet! Thankfully WOL wasn't implemented - so a bunch didn't start, plus we managed to turn a heap of systems off before they ran the client (thankfully Pentium 233MMX's don't' boot that fast!)
It did take a few days to sort out the remaining carnage though!
Exactly - I have about 2000 Telstra mobile accounts at work, and because I couldn't use connectivity on my phone, tablet or laptop, I couldn't check how many were offline (during a peak customer period)...
Once I got home to fixed internet, I was able to tell that most things were offline for at least an hour, the saving grace is that a large number of customers at that time would have been on mobile too, so a significant % of them probably didn't know we were offline.
Free data on a Sunday?! What about all the loss of revenue for business where EFT transactions didn't go through and the customer walked?
What about the loss of SCADA to remote power/water/etc. monitoring systems? How much went or will still go wrong as a result of this outage?
Not saying that a compensation plan is easy to determine, but reliable communications is usually categorised as critical national infrastructure - a partial outage of which is a _big_ deal.
Do we need another standard, is there an issue with using the current 'almost CAT6' 28 & 30 AWG cables that are available?
e.g.
Telegärtner:
http://www.telegaertner.co.uk/telegaertner-uk-news/press-releases/30AWG-Reduced-Diameter-Cat-6-patch-leads_165.htm
Panduit:
http://www.panduit.com/en/landing-pages/small-diameter-patch-cords
(I use both of these in various fixed and mobile applications, they're both LSZH, so what's not to like?)
Just retiring some dc7700's here (a mix of XP & Win7) - admittedly we've had a few of those with faulty mainboards over the last decade or so!
(and we still have 1x d530s in use)
I'll echo the sentiment of "why upgrade?"
We're using HP desktops at work. I recently took delivery of a bunch of fully-optioned 800 EliteDesk G1's (G2 is out now). The boxes they're replacing are 5 years old, and still perfectly functional. Being retired due to age, not capability...
Sorry guys, Windows 10 isn't yours. It's being leant to you for free (for now). At what stage did you think you had the right to prevent this 'telemetry' stuff from happening?
This is one of the things that annoys me so much about W10 - if the product had been managed 'better' people would be falling over themselves to use it (it certainly seems to perform well - from what I've read).
C'est la vie...
And another hot tip (learned the hard way...) use the 'startup delay' feature of your servers BIOS. It will give you a staggered startup of the boxen - reducing that dreaded inrush!
(We had 2x 2kVA UPS's running at about 30% load under _normal_ circumstances. But when everything wants to start up at once, one UPS overloaded and went offline and the other was stuck on bypass at 150% load. That's the sort of thing you can't fix remotely...)
When the storage requirements aren't clearly explained, or just described as 'space for the install, not all the temporary crap I do before I've finished'.
I recall a few Exchange upgrades (admittedly _many_ years ago!), that needed some hasty data relocations to be performed while the installation was running, lest it ran out of space!
10 years old, on it's second battery.
A fine example of the level of quality that is possible if the manufacturer is interested.
Yes, is was a $5K AUD laptop, but it's still in the field - every working day (supporting a system that needs an XP box).
Personally I have an Asus N61J that's over 6 years old - SSD and new battery fitted. The onboard DVD-RAM drive is dead, but I think she's got a few years left yet.
IMHO - Commodity PC's are just that, decent servers are far superior, and some shit just lasts longer than other shit.
I'm sure I can dig up a functioning Apple Newton from around here somewhere...
I've just resurrected one of these with my father-in-law (actually a 'Break' or estate/wagon), and it's a fun little car to drive. It will happily motor along at 100km/h with very little road or engine noise (even though it's spinning at nearly 5000 rpm), and keep up with modern vehicles in traffic. Don't let the revs drop too much tough, or you'll be rapidly seeking a lower gear!
Amen to that - have a look on the PCI web site at the number of EFT terminals that have '2G' backhaul and compare it to those that have '3G'. I'm not willing to take a guess as to how many 2G devices need to be replaced in Australia over the next 18 months, but it could be millions.
Yes, I'm working on replacing a 'small' number of these in vending machines (~200) and it's quite a big job!
Shirley, things like the executable 'protection' above could be backported to 7 without too much dram? Oh that's right, 7 is old hat now.
Well MS, I too have used and deployed your Operating Systems on many thousands of machines over the last 30 years, and this 10 thing will go down as a bigger fiasco than the 'Windows Vista Ready' cockup.
I currently manage a government network of about 8000 nodes that is not internet connected - for many reasons. Windows 10 is just wrong on too many levels to be even considered for testing in this environment. Not only that, but with all the telemetry stuff going into 7, we're fighting to prevent the current fleet from wasting it's time trying to phone home...
Nice work - alienating the corporate customers
(Opinions are my own, but I suspect that they are consistent with that of many others)
The mind boggles on how poor their ICT systems must be for them to not know where over 1000 XP boxes are. While they may have a massive fleet (I expect it would be in the hundreds of thousands of systems), surely they can use one of the multitude of auditing methods out there to find those machines?
If they're on an AD, it's a no-brainer - even if they're not, it's still straightforward...
You make a valid point, but sometimes it just can't be done in the application space. FT can give an additional layer of comfort for such critical workloads.
I was planning on using FT for ay DHCP server - but it still has limitations, and requires _very_ fast networking to keep the VM's in lock-step.
(from memory, a dedicated 10gbe link - or faster)
I've just had a similar fight with Aust. customs. They wanted to charge me 5% duty on $2000AUD worth of goods from the USA that were meant to be duty fee under the FTA. I eventually won the dispute and only paid brokerage + GST.
But the fact that they tried it on anyway (when the nature of the goods and tarrif code clearly indicated the situation), is the bit that is annoying.
It _seems_ that that their SOP is 'charge unless we get caught out'.
Not looking forward to them being involved in collecting/processing the little stuff too.
(and I'm sure they're not chuffed about it either)
Hmmm, I seem to recall that 20 years ago, I was working with 3rd generation 32-bit x86 CPU's, admittedly they had an FDIV bug, but you get that...
As for other Architecture types, they'd been 32-bit for even longer (MIPS/Alpha/PPC/SPARC).
You sir, have misquoted a misquote!
:)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
