Christmas cheer for KCL staffers with gift of extra holiday after IT disaster 'Tis the season to be jolly at King's College London, where staff have been offered extra holiday in recognition of their response to the university's IT disaster. Not too jolly, of course. It won't be possible to take pay in lieu of the extra two days, although staff aren't sniffing at the offer. An email sent from KCL's …
If he hadn't said this, no doubt there would have been plenty of internet professional offence-takers who would have filled the tweetisphere about "first world problems", "academic ivory towers", "join the real world", "public sector spongers" and "don't know how lucky they are".
This post has been deleted by its author
so what with long school holidays etc that'll be a 1/2 day in the real world.
It never ceases to amaze me how many people think that staff at schools, colleges and universities only work during term time. If ever a test is introduced to determine if people are capable of functioning in the modern world (allowed to vote, allowed to drive on public roads, allowed to reproduce, etc), a question to this effect should form part of the test.
Quite a few aunts and uncles who are teachers. Summertime - that'll be 6 weeks plus down in New Zealand or travelling around Australia. Easter/Winter term breaks - a quick 2 weeks somewhere sunny or snowy.
Never heard of any of them working during the holidays. And yes, some of them work in universities so it's not just primary/secondary school teachers I'm talking about.
So, from my own personal experience, teachers don't work in the holidays. Your experiences may differ.
@yao-fm
With the greatest of respect to your relatives, they don't sound very dilligent teachers. My parents were both secondary school teachers, and at least two weeks of their 6-week holiday were taken up with lesson and timetable planning for the new term,
I remember them with a great roll of paper spread out across the floor trying to fit all the lessons into a week, and all the pupils into the lessons without any clashes.
(this was in the eighties, nowadays I expect they'd use a spreadsheet).
Teachers doing timetable planning? It must have been a very small school (with only two teachers). In most schools timetables are drawn up by the admin staff and handed to the teachers. A second's thought would tell you that it would be impossible for each teacher to draw up their own timetable! Mr. Figgs could plan to have IIIC on period 3 every Tuesday, and I might timetable a double physics lesson for that form at the same time.
Teachers tend not to work in Unis..
While it may seem petty, Lecturers, while they do a similar job, tend to have higher qualifications.
That said, the academic staff of a Uni tend to be a fraction of the full staff. There are researchers, support staff and admin staff, all of whom work with the same holiday restrictions that people work to in the private sector. I.e getting so many days off per year, with the number going up for each year of service. Speaking as a Uni tech support bod, the holidays are often our busiest times, because we tend to schedule any mass upgrades to happen during the holidays, to minimise disruption to students and lectures.
Yes, the academic staff do tend to sod off during holidays (and object vociferously when asked to come in). However, even they are expected to work during the holidays, where ever they happen to be. The researchers tend to work in subjects they are interested in, so it's not uncommon for some researchers to work 10-12 hours a day, 5 days a week, and carry on that pattern throughout the holidays.
"
It never ceases to amaze me how many people think that staff at schools, colleges and universities only work during term time.
"
That's because it's usually perfectly true. I used to be a high school teacher. Preparation for the following term took maybe 4 days. Tops. The rest was pure holiday. That's if I didn't manage to get it done whilst invigilating exams during the last fortnight of the previous term. Maybe a few new teachers go overboard with preparation, but after a couple of years you realise that it doesn't make the slightest difference to the kids, and you can re-hash last year's preparation anyway.
Personal backups are a data protection nightmare, depending on what you're handling.
If KCL has any kinds of dealings with the NHS, outside companies, lists of people in any way, shape or form, a personal backup is a Data Protection breach waiting to happen, with huge fine attached.
That said, personal research, PhD thesis, lecture notes for lecturers, etc.? Why don't you ALREADY have it backed up?
In hindsight, would a cheaper backup system (i5 workstation with a single external USB hard drive which gets swapped out with other externals as the week progresses) have been more sufficient than a bells-and-whistles backup system that tend to bork horribly just by looking at it the wrong way? Should one external fail to be large enough, two (or more HDD's - one for each department) should be more than sufficient, yes?
Depending on the nature of failure, it is sometimes simpler and easier recovering data from a single stuffed HDD than trying to recover it from a RAIDed array, or so I'm led to believe, but I may be wrong.
What would YOU recommend as a low-cost, solid and reliable backup system that data can be recoverable should somebody do an accidental rm -rf *?
"In hindsight, would a cheaper backup system (i5 workstation with a single external USB hard drive which gets swapped out with other externals as the week progresses) have been more sufficient than a bells-and-whistles backup system that tend to bork horribly just by looking at it the wrong way?"
The thing is, the only reason they would be attempting to recover data from the borked RAID array is if they were treating it as their backup system (repeat it with me: "RAID is not NOT backup - never treat it as such") and don't have a regular backup regimen to tape or other media.
My understanding is that the system failed during a critical upgrade. It seems unlikely that a mirrored array would have both drives just die - it seems more likely that that someone failed to replace an already degraded drive, and then casually kicked off the upgrade with only one drive in the mirror working.
If so, a classic example of accidents tending to be the culmination of a series of errors, rather than one isolated issue.
Its actually quite likely that a mirror can break at both sides within a short time frame, as both disks have been written to in exactly the same manner, the disks may have been close on the production line as well. You really need multi party raid, so that multiple drive failures do not (or are highly unlikely) to cause problems. This is becoming a huge problem with large disk,s as rebuilding a raid array with 6TB,10TB or more takes a long time. in our old SAN we had 3 failure in a week out of 120 disks, luckily they are smallish drives so rebuild times are low. Our new one uses 3 drive parity, so you can have 3 simultaneous drive failures and still be okay.
"low-cost, solid and reliable backup system"
That's one of those "pick any two" things, isn't it? (It's also lacking "secure/compliant" and "sufficiently fast" at the very least.)
As usual, answers vary wildly, and some departments tend to have equipment that IT here jokingly refers to "press a button, and several TB fall out the other side", which have very different needs from 3 G of Dean's mailbox.
One of the tricks I like for dealing with beancounters is showing them real World scenarios where cost-cutting led to expensive disasters. I'm guessing this was one of those events where someone decided to buy a single-node, single-shelf 3PAR instead of an MSA or some P4000s (whaddya bet the salesgrunt was after the extra hp commission being offered for flogging 3PARs?), but it would be nice to know the design and operational details so the effects of the cost-cutting can be highlighted. I generally find that about 90% of IT disasters are actually beancounter-induced.
As an aside, I remember building a fileserver out of spare PC parts for a professor (many, many years ago) to get round a centralized backup policy he had no faith in. It seems little has changed in academia.
"You don't happen to have that list of disasters do you?...." Sorry, the majority of them are covered by NDAs. Maybe I'll retire to somewhere without an extradition treaty with Europe or the US and release them in a book (without names changed to protect the guilty).
In the meantime, a good source for scaring beancounters is the excellent Catalogue of Catastrophe. As probably a World record of expensive benacounter-induced failure ($43bn to date and counting), it looks like cost-cutting was the root of the VW emissions "cheat device" - ".....Reports indicate that the technology needed to comply with the government regulations was available, but someone in the project that designed the affected engine decided that the cost of the necessary components was too high....."
#4 King's College London Professor Edward Byrne £458,000 (£324,000) i.e. a 29% increase
£458,000-£324,000=£134,000
£134,000/£324,000=0.4136 => 41.36%
chk: £324,000(1.4136*)=£458,000
*assuming you don't drop the unrounded result from your trusted HP-15C.
Or is there a New Brexit-Math that hasn't made it to left side of the pond yet? We might need some of that here in Trumplandia come 20Jan2017.
It was quoted from two different tables in a report by the UCU. I didn't check their calculations but I suspect it's to do with a relocation grant that was awarded as a one off rather than the regular pay element. One table is for all element emoluments, the other just emoluments.
Two days. Whoopee fuckie doo. While the vice chancellor of KCL trousers £458,000.
http://www.dailymail.co.uk/news/article-3441526/It-s-no-wonder-universities-want-curb-freedom-law-Dozens-vice-chancellors-pay-packets-soar-10-past-year-spend-thousands-flights-hotels.html
Nice to know you're appreciated, IT staff.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
