Handler on Duty: Johannes Ullrich
Threat Level: green
Podcast Detail
SANS Stormcast Wednesday, October 22nd, 2025: NTP Pool; Xubuntu Compromise; Squid Vulnerability; Lanscope Vuln;
If you are not able to play the podcast using the player below: Use this direct link to the audio file: https://traffic.libsyn.com/securitypodcast/9666.mp3
My Next Class
Application Security: Securing Web Apps, APIs, and Microservices | Dallas | Dec 1st - Dec 6th 2025 |
Application Security: Securing Web Apps, APIs, and Microservices | Orlando | Mar 29th - Apr 3rd 2026 |
What time is it? Accuracy of pool.ntp.org.
How accurate and reliable is pool.ntp.org? Turns out it is very good!
https://isc.sans.edu/diary/What%20time%20is%20it%3F%20Accuracy%20of%20pool.ntp.org./32390
Xubuntu Compromise
The Xubuntu website was compromised last weekend and served malware
https://floss.social/@bluesabre/115401767635718361
Squid Proxy Vulnerability
The Squid team fixed an information disclosure vulnerabilty that may leak authentication credentials.
https://github.com/squid-cache/squid/security/advisories/GHSA-c8cc-phh7-xmxr
Lanscope Endpoint Manager Vulnerablity
https://jvn.jp/en/jp/JVN86318557/index.html
Application Security: Securing Web Apps, APIs, and Microservices | Dallas | Dec 1st - Dec 6th 2025 |
Application Security: Securing Web Apps, APIs, and Microservices | Orlando | Mar 29th - Apr 3rd 2026 |
Network Monitoring and Threat Detection In-Depth | Amsterdam | Apr 20th - Apr 25th 2026 |
Application Security: Securing Web Apps, APIs, and Microservices | San Diego | May 11th - May 16th 2026 |
Network Monitoring and Threat Detection In-Depth | Riyadh | Jun 20th - Jun 25th 2026 |
Podcast Transcript
Hello and welcome to the Wednesday, October 22, 2025 edition of the SANS Internet Storm Center's Stormcast. My name is Johannes Ullrich, recording today from Jacksonville, Florida. And this episode is brought to you by the SANS.edu Graduate Certificate Program in Incident Response. I mentioned yesterday talking about the compromise of the Chinese standard time servers that we have been collecting data about pool.ntp.org. These are the NDP servers that are basically an open project. Everybody can contribute their time server to it to make it easy to synchronize your time. And a lot of systems, in particular Linux systems, tend to use these NDP servers as their default. Now, took some time today to go over the data that we have collected just to see how accurate those time servers are. And it turns out they are very accurate. I had to use a double logarithmic scale to really show anything here but a real tight spike. Most of these, and we're talking here about more than 90% of the servers have an accuracy of less than 10 milliseconds. And pretty much all of them are less than 100 milliseconds. So, that certainly is sufficient for most sort of small business, home networks and the like to synchronize your time. And again, remember there are two parts to it. This is the synchronization to the external time standard. And then, of course, you also have your internal synchronization. The other thing here is you can contribute your own time server if you want to. But they put a little bit of warning out here that once you commit to it, you should better stick to it. Because, well, people will keep querying your time server even if they can't reach it. Also added a link to the feed that we have. Basically, you can look at the data yourself and check it out and see if you see any oddities or such in this data. But also having a list of NDP servers, these public NDP servers, can be handy because you will see some of your systems reaching out to these IP addresses. And I've seen in the past where firewalls sort of blocked the responses if there aren't really all that great in handling UDP statelessness. And that can sometimes cause some false positives. So easy then to discriminate against these false positives if you have this list of NDP servers, Andy. This weekend on Sunday, the ex-Ubuntu website was compromised. And download links did point to malicious software. This malware was, well, as often we were really lucky here. It was relatively basic malware. Apparently, some kind of crypto coin jacker that copies crypto coin addresses from the clipboard. It makes itself persistent via registry entry. So nothing really all that special. And antivirus often did alert on this malware. So by now, I would think that antivirus pretty much has taken care of it. And there wasn't really sort of, as far as I've seen, anything more malicious or more sophisticated behind this malware. There's no official statement yet from the Xubuntu website. They just disabled the download links for now. There is, however, a statement from Sean Davis, who is one of the maintainers here of Xubuntu, stating that they suspect or know that it's some kind of WordPress compromise. And they're sort of waiting a little bit on Canonical, the company behind Ubuntu, to resolve this issue. The main Ubuntu site download was not affected by this. So this was just Xubuntu. I have no real idea how popular that is compared to the official Ubuntu distribution. So whether or not people prefer Xubuntu, I doubt it. I think it's probably at least an order of magnitude or so less downloads than Ubuntu itself. Either way, if you downloaded Xubuntu this weekend and the entire compromise stretched for about 12 hours, I believe, on Sunday, you should double check that you downloaded the right thing. And then, you know, if you sort of manage a larger network, maybe go a little bit hunting and see if anybody downloaded this malware this weekend. Then let's talk about a couple of vulnerabilities. First of all, SQUID, the proxy web server. Well, it suffers from an information disclosure and error handling. No idea why this was assigned a CSS score of 10. Again, that certainly sounds inflated. Not a lot of details here. But what usually happens in these kind of vulnerabilities is that if you configure the default error messages, if the user triggers an error, things like headers, which may include authentication cookies and such, are being echoed back as part of the body of the page, which then, of course, can be accessed because now you're no longer restricted by things like HTTP only or other properties that may prevent JavaScript access and such to the value of these cookies. So, yes, you want to address this. It's not a 10. I would say probably something like a 7 or such, depending on how you exactly are able to trigger this vulnerability. And then we have a vulnerability notice for Lanscope. Endpoint Manager affects the on-premise solution. A single malformed network packet may lead to arbitrary code execution due to this vulnerability. And apparently this vulnerability is already being exploited. So, something that you must patch quickly. And well, that's it for today. So, thanks for listening. Thanks for liking and recommending this podcast. And talk to you again tomorrow. Bye.