• 0 Posts
  • 24 Comments
Joined 1 year ago
cake
Cake day: June 15th, 2023

help-circle




  • I would never use their firewalls/gateways, but their switches are pretty good for the price and their APs are decent (although tbh after 3 generations my next AP will likely be an enterprise Aruba).

    That said, I still use Unifi in docker, everything is up to date, and nothing is requiring a sign-in to the cloud. Am I missing something? If it’s just the firewalls, then I’m not surprised since I’ve never been remotely tempted to use them, but it sure isn’t all of their devices.







  • If you, Traefik, and your origin server are on the same network, then it’s going to be one hop regardless of whether you’re hitting the Traefik proxy or the origin server. If Traefik is serving up the origin server’s cert and not the LE cert, then Traefik is misconfigured to pass through instead of proxy, but I’m still not sure that’s the case as it’s almost harder to configure it that way than the correct way as a proxy.

    What IP:port is your origin server listening on, what IP:port is Traefik listening on, and how is Traefik configured to reach the origin server?



  • Yep totally. The documentation is downright wrong so much more today than it used to be. It’s all written like they pawned it off on a junior engineer, who then threw shit at the wall until they got it working, then that process becomes the official documentation.

    And don’t get me started on Copilot hallucinating Powershell cmdlets.

    With support it’s become kind of a game to see how quick you can get to T2. My tactic is to passive aggressively point out how their first response shows a complete lack of understanding of the topic, then directly request escalation.


  • The reality is they probably don’t know the full scope or root cause and are going off of limited reporting coming from their beta channels.

    But they likely determined the impact was low enough that they could still ship the update while they investigate further.

    There are similar known issues reported in the update KBs all the time that sound much worse to me as an admin but are as equally low impact in the end. But they’re not as easy for the layperson to latch onto like these low-effort “VPN no worky” articles.

    Regardless, none of this absolves IT of the responsibility of testing patches.


  • Exactly. Everybody on Lemmy a couple days ago was acting like the sky was falling when all we had were these one-paragraph FUD articles quoting Microsoft’s own KB article. Most people commenting have no clue that “VPN” is a broad term covering at least a dozen different possible protocols and acted like Microsoft was intentionally breaking all VPNs.

    The only thing I found was a reddit thread talking about how some VPNs using TPM-backed certs were broken. I, for one, am using an IPsec VPN with certs stored in TPM on one of the affected versions of Windows 11 and have had no problems. Nor have I had any issues with SSL or Wireguard-based VPNs, so it does just seem to be a fringe case they’re warning about.

    So Microsoft is just giving a heads-up that IT should probably include VPN testing in their patch cycle test rings and all the anti-MS people are losing their shit.





  • Third. The first thing I mention when one of my clients asks anything about PCI is to offload as much card processing onto third parties as possible.

    And if you have nothing in place yet, then 100% offloaded should be possible (with the possible exception of secure payment terminals if you need to process physical cards).

    That said, it is still possible to use your own hosted WordPress storefront and offload the payment processing via tokenization or redirection. But a turnkey solution like Shopify might be better if you lack the experience.