NHS staff trigger Google cyber-defences | Hacker News Reader

NHS staff trigger Google cyber-defences

theregister.co.uk

This is a stupid headline. They're tripping the request limit from a single ip, which makes you fill out a captcha. We've been hit with that a few times where I work.

It's The Register, daft headlines is their style. They imitate the cheap, mass-market newspapers in the UK like The Sun or The Mirror.

The NHS is one of the world's largest employers. If all that traffic is behind a small range of IPs (one wouldn't be enough) I'm not surprised there are issues. It does seem an odd network design though.

> If all that traffic is behind a small range of IPs (one wouldn't be enough) I'm not surprised there are issues. It does seem an odd network design though.

"Plonk all our desktops behind a router, NAT the whole thing" isn't that unusual as a network design.

One would hope that the government in particular would be moving to IPv6 as the socially responsible thing to do.

It's a bit unusual for an organization with thousands of sites, spread across an entire country. I haven't checked, but it probably has a huge IPv4 allocation too.

I'd expect something more like the university networks.

This, plus the dodgy headlines about Google employees staging a "walkout" a few days ago. The state of news media is depressing.

Or perhaps there is active malware within the NHS network? Why is the immediate assumption that the amount of traffic is not actually an indicator something is wrong?

When I've managed office networks and we've seen the captchas, every time it was traced to actual infections of a few machines. Re-imaging them (and educating users) fixed the problem.

I wonder if it will turn out that some random computer in the NHS was infected with a SEO scraping bot.

But no, that is impossible. The NHS is clearly much, much better at running IT than Google is.

Hell, I sometimes forget which end of the Ethernet cable plugs in where.

Seems like quite the overblown headline, but it reminded me of the recent self-inflicted email DoS at the NHS: https://www.theguardian.com/society/2016/nov/14/186m-needles...

I suppose articles like this are a worthy reminder to remember the extreme cases in our engineering

But this (the e-mail DoS) isn't even an extreme case. When you have on the order of a million accounts, especially when you have large distribution lists, you need to consider this as an expected case (and the best practice solution is pretty well understood: limit send-rights to large distribution lists to the people who need it).

It's depressing how little effort Microsoft invests in trying to fix the obvious brokenness[1] of email (including reply-all) and (especially) calendaring. It's a particular contrast to the often controversially heavy handed actions by Google to "outlaw" insecure internet practises.

1: The brokenness is obvious, the solutions admittedly less so.

I work in the NHS and regularly get the "I'm not a robot" captcha. I know what it is and why it's there when I see it, but colleagues often complain that their "Google" has a "virus".

A lot of NHS traffic goes through a few large networks like THIS [https://www.this.nhs.uk/home/] which don't have especially large IP allocations.

I hope I'm pointing out the elephant in the room here: how many of their PCs are infected ?

This!

I always wonder if in an increasingly automated world, these sort of mishaps will become more common, and take longer to elevate to a human level where it can be fixed.

Definitely been in this camp before at previous places I've been. It's particularly crazy that this is happening with NHS, given Google's very public partnership with them to collect and process their medical data through DeepMind.

The NHS is not a single entity, it is a huge, loose collection of different bodies in a very complex arrangement. The bodies that Google is involved with are likely entirely separate from the ones providing the affected IT service.

Clickbait and highly misleading headline.

Reply on news.ycombinator.com