Back on August 12th 2025, I’d suspected Colt Technology Services had a ransomware incident running. For those who haven’t heard, Colt are quite important in the UK and Europe:At the time they were referring it to as a “technical issue” on their status page. This continued for just over two days.Here’s an archive of their status page — they text about technical issue has since disappeared:They were trying to respond to an unfolding ransomware/extortion incident without telling customers what was happening.As always with these cyber incidents involving unpredictable external parties (i.e. e-crime groups), transparency is the best option as everything begins to unfold quickly otherwise — I’ve seen this time and time again at organisations. Quite often internal forces — e.g. legal departments — will push incidents in the wrong direction from actually protecting the overall business, as they’re operating in isolated thought around their own department responsibilities.You have to place protecting customers first in every situation, as the business is essentially experiencing both a technical heart attack and an attack on trust by the threat actor. Perception is reality. It’s best to control the narrative as much as possible, and the best way to do that is to do generally the opposite of everything people who’ve never dealt with a ransomware incident will tell you to do.After a few days, they still weren’t letting customers know it was a cyber incident — so I started a thread online to try to coax Colt out of the dark:https://cyberplace.social/@GossiTheDog/115026580893996072From telemetry, I could see an operator from Warlock ransomware group hitting sharehelp.colt.net, a SharePoint server that was now offline. It was also clear they’d done data exfiltration.After tweeting this, LeakIX — an internet scanning service — contacted me to let know Colt had a popped SharePoint box — also the same IP as sharehelp.colt.net — which had a webshell on it. They’d let Colt know on July 22nd:Email thread from LeakIXIt looks like this was the entry point. The box was initially vulnerable to CVE-2025–53770, however it’d been patched fairly quickly from version fingerprint scan data — but it already had a webshell on it by that point, i.e. the backdoor was placed for later.You may recognise these TTPs if you pay attention to Warlock — aka STORM-2603 — as Microsoft had reported this vuln being used for ransomware deployment by the group, with the same webshell filename (spinstall0.aspx):After these delightful toots, Colt pivoted to the cyber incident messaging later that day:From this point, their comms improved considerably.Although the comms refer to “an internal system”, it’s pretty clear given the scale out of the systems still offline that it’s a big one not confined to a system.By August 15th I’d noticed there was a forum post on RAMP, a Russian Tor site, from Warlock advertising data for sale:In this, it included a file listing of approximate 400k documents, including Colt customer documentation and performance reviews of staff and such.That list is available here, is orgs want to understand their level of risk:https://www.klos.com/~john/colt_filename_tree.txtColt have now setup a dedicate breach website — it’s not linked on their website for some reason and the HTML code includes noindex, which stops search engines listing it:Cyber Incident | Colt Technology ServicesThey’ve also admitted the theft of customer data on said site.Some learningsThese incidents are always a bit of a car crash — I’ve been there several times — and as Mike Tyson once said, “Everybody has a plan until they get punched in the face”.Ransomware and extortion incidents are those situations as the incident environment becomes fluid, due to the usual pillars of business becoming unstable — for example, IT systems and access to documents. Staff also enter an extreme amount of stress. Watch for that, rotate people on shifts, you’re going to be recovering the company for a while, it’s not a sprint.You have to be agile and you have to be fast, and think outside the box.For me, I think transparency is the centre point to aim people at. For example, people should be asking ‘if I’m a customer, would I want to know? Should my colleagues know what is happening?’You can, and should, aim to suck the life out of the threat actors extortion by pre-empting them. Yes, they’re going to go public to try to get customers to complain. So tell the customers first. This one was pretty slow in terms of external response — I’m sure lots was happening internally — which allowed the threat actor to go a bit nuts.Don’t pay the threat actor. These guys are running around with millions of dollar attack budgets — far more than your cyber defence budget — because orgs keep paying.I think running SharePoint on prem in 2025 is hugely risky. There’s not that many on prem boxes left online — for example, they’re dwarfed by on prem Citrix Netscalers nowadays by a factor of several times (not that it’s a good idea to run Citrix Netscaler on prem in 2025, either) if you look on Shodan.Threat actors will find vulnerabilities in that product, as it’s a dwindling number of high value orgs running it. It’s time to say goodbye to SharePoint.Orgs probably want to look at attack surface management as clearly when those SharePoint vulns hit as a zero day in July, things were going to go wrong for some orgs. We all need to know where our web facing systems are, know a way to look for known webshells (the filename in this one was a known one, hence LeakIX being able to scan for it) and respond accordingly.But, you know, hindsight is a wonderful thing, cyber incidents like this are truly terrible, and there’s no perfect way to respond. I think Colt get serious credit for — per their cyber incident website — having a properly segmented network for customer systems, which has limited the impact considerably.I definitely think everybody should try to learn from incidents like this, though, and really think about: do we know what to do really? Ransomware and extortion incidents aren’t a IT disaster recovery scenario. It’s somebody deliberately starting a fire. Every org I’ve come across has an IT disaster recovery plan, but it never survives somebody burning down the building on purpose.One final thing. The UK government needs to continue to treat ransomware seriously. There are plans afoot to stop critical infrastructure paying ransomware, and all incidents to be notified centrally, here in the UK. These plans need to continue at pace, and get implemented. We’ve got to break this economic cycle.The longer these ransomware groups keep getting funded, the more incidents there will be, the more vulnerabilities purchased etc — sooner or later you’re going to get a Colt style incident at another provider who hasn’t segmented their network properly. And that’s going to be a highly significant incident. This wasn’t.The UK government should get out in front of that situation and really be a world leader by quite literally leading the world on how to defuse these ransomware groups through thoughtful legislation and central governance.UpdatesYou can follow me for updates on Mastodon if you’re really bored.https://cyberplace.social/invite/BeKU6RCGColt Technology Services gets ransomware’d via SharePoint initial access— some learning points was originally published in DoublePulsar on Medium, where people are continuing the conversation by highlighting and responding to this story.