AD, Group Policy, DNS, PowerShell, event IDs
Practise these as flashcards. One or two lines each; the ones you miss come back first.
Start freeA server running Active Directory Domain Services. It authenticates users and computers, holds the directory, applies policy, and replicates with other DCs. You run at least two for redundancy.
An OU is a container for organizing objects and linking GPOs. A security group is for granting permissions. GPOs link to OUs, not groups.
Run gpresult /r to see what applied. Check the GPO is linked to the right OU and the object is actually in that OU. Check security filtering. Run gpupdate /force. Check DNS, because clients find DCs through DNS.
netstat -ano | findstr :443 gives the PID; tasklist /fi "pid eq <pid>" or Get-Process -Id <pid> names it. PowerShell: Get-NetTCPConnection -LocalPort 443 | Select OwningProcess.
NTFS permissions are more restrictive. Effective access is the most restrictive of share and NTFS. Check the Effective Access tab on the folder.
Kerberos is ticket-based, the default for domain logins, and needs working DNS and clocks within 5 minutes. NTLM is the older challenge/response fallback, used when Kerberos can't work, like connecting by IP address.
Kerberos fails after 5 minutes of skew, so logins and service authentication break. Fix with w32tm /resync, and make sure the PDC Emulator syncs to a reliable external time source.
Check the System log: 1074 shows a planned restart and who or what requested it (often Windows Update or SCCM). 6008 or 41 means an unexpected shutdown or crash, so look for a crash dump next.
It didn't respond in time. Check its dependencies (sc qc name), the service account and password, and the Application log for the real error underneath.
Remote Desktop enabled, TermService running, firewall allows 3389, user is in Remote Desktop Users, NLA settings match, and on an RD host the licensing grace period hasn't run out.
Task Manager and Resource Monitor for the top CPU, memory, disk and network consumer. perfmon counters: % Processor Time, Available MBytes, Avg. Disk sec/Read and Write (over 20 ms is bad), Disk Queue Length. Event logs for errors. Then the hypervisor side: CPU ready, ballooning, datastore latency.
The app pool is probably stopped. Check its state with Get-IISAppPool. If it keeps stopping, rapid-fail protection is tripping on repeated crashes: read the Application event log. A classic cause is the app pool service account's password expiring or the account getting locked.