EXOIT
- The Wi-Fi outage that NPS never saw, part 1
Clients dropped off corporate Wi-Fi after a certificate renewal, and NPS logged nothing for 14 days. For most clients the fault was in the wireless profiles.
- The Wi-Fi outage that NPS never saw, part 2: a binding, a wall and a proxy from 1996
Three faults in series behind one Wi-Fi outage: a dead NPS certificate binding, a blocked CRL and a proxy that only speaks HTTP/1.0.
- The error code in the portal is the fourth link in the chain
Azure Update Manager said 0x80240007, a corrupt datastore. The real cause was a WinHTTP proxy nobody remembered, three layers further down.
- The sensor that was stuck migrating had already arrived
A Defender for Identity sensor sat in Migrating for 48 hours. The v3 sensor was already healthy, hidden behind a case-insensitive portal view.
- Last logon is not usage: the SSO outage I caused
I disabled 19 inactive service accounts and broke SAP single sign-on. The accounts were in use; lastLogon just can't see it.
- The retention policy that ran successfully and did nothing
A move-to-archive policy ran without errors for a week and moved nothing, because the mailbox had no archive to move to.
- The door is open
What this place is for, and why it starts with a grid of dots.