Common Errors – Error Tutorials

#FC-001 CRITICAL_PROCESS_DIED [STATUS: SEVERE]

PLATFORMWindows 10 / 11 TYPEBSOD / Kernel Stop Error STOP_CODE0x000000EF

How and Why CRITICAL_PROCESS_DIED Happens

The CRITICAL_PROCESS_DIED Blue Screen of Death occurs when Windows detects that a process essential to the operating system's integrity has stopped unexpectedly or entered an invalid state. The Windows kernel monitors a set of critical system processes — including csrss.exe, wininit.exe, winlogon.exe, and smss.exe — that must remain running for the OS to function. When any of these terminates without authorisation, the kernel issues stop code 0x000000EF and halts the system to prevent data corruption.

The most common root causes are: a failing or incompatible device driver writing bad data into kernel memory; corrupted Windows system files caused by a failed update or abrupt shutdown; faulty RAM producing read errors mid-execution; an overheating CPU throttling unpredictably and causing process timeouts; or malware that terminates protected system processes as part of its attack chain.

Possible Fixes

Fix 1: Run the System File Checker and DISM

Corrupted system files are the leading cause. Open Command Prompt as Administrator and run the following commands in order. First, type sfc /scannow and press Enter — this scans all protected system files and replaces corrupted ones from a cached copy. Wait for the scan to complete fully; it takes 10–15 minutes. If it reports corrupted files it could not fix, run DISM /Online /Cleanup-Image /RestoreHealth next. This fetches fresh component files directly from Windows Update servers and rebuilds the local store. Reboot after both commands complete, then check if the BSOD recurs.

Fix 2: Update or Roll Back Device Drivers

A newly installed or auto-updated driver is frequently the trigger. Open Device Manager (right-click the Start button) and look for any device marked with a yellow warning icon. Right-click it and select Update driver → Search automatically. If the crash started after a recent Windows Update, rolling back the display or network adapter driver often resolves it immediately: right-click the device, choose Properties → Driver tab → Roll Back Driver. If no roll-back is available, uninstall the driver entirely, reboot, and let Windows reinstall the generic version. GPU drivers from NVIDIA and AMD should be downloaded directly from the manufacturer's site rather than via Windows Update.

Fix 3: Test RAM with Windows Memory Diagnostic

Faulty memory modules cause kernel processes to crash mid-read. Press Win + R, type mdsched.exe, and hit Enter. Choose Restart now and check for problems. The tool runs before Windows loads, performing a thorough pass across all installed RAM. If errors are detected, the report appears on next login under Event Viewer (Windows Logs → System → MemoryDiagnostics-Results). A confirmed RAM error means the faulty stick needs physical replacement. Test sticks individually if you have multiple DIMMs to isolate the defective one.

How to Prevent CRITICAL_PROCESS_DIED in the Future

Keep Windows Update set to install updates automatically so driver and security patches are applied before vulnerabilities can be exploited by malware. Run a reputable antivirus scan monthly to catch any process-targeting threats early. Monitor CPU and GPU temperatures using a tool like HWMonitor — sustained temperatures above 90°C indicate a cooling issue that will eventually cause process failures. Avoid force-shutting the computer during updates. Before installing third-party drivers, verify they are WHQL-signed. If the system is over three years old, consider running a memory test annually as RAM degrades over time. Finally, keep a System Restore point active so any bad driver update can be reversed within minutes.

#FC-002 0x80070005 — ACCESS DENIED [STATUS: SEVERE]

PLATFORMWindows 10 / 11 / Windows Update TYPEAccess Control / Permissions WIN32_MSGERROR_ACCESS_DENIED

How and Why Error 0x80070005 Happens

Error 0x80070005 maps directly to the Win32 status code ERROR_ACCESS_DENIED. It fires whenever a process or service attempts to read, write, modify, or delete a resource — a file, registry key, directory, or network object — without holding the required security permissions. Windows enforces access through Access Control Lists (ACLs) attached to every object, and when the token presented by the requesting process does not match the required privilege level, the kernel returns this error code rather than allowing the operation.

The most frequent contexts are: Windows Update failing because the Update service cannot write to a system folder due to a locked or corrupted ACL; a third-party application attempting to write to Program Files or the registry without elevation; antivirus software blocking access to files it has quarantined; or a user account operating without Administrator rights trying to perform a privileged action. Corrupted user profile permissions and inheritance breaks in the SYSTEM account's folder rights are also common triggers.

Possible Fixes

Fix 1: Run the Affected Application as Administrator

The simplest resolution for application-level errors. Right-click the program's shortcut or executable and select Run as administrator. If this resolves the issue and you need it permanent, right-click the shortcut, go to Properties → Shortcut tab → Advanced, and tick Run as administrator. For Windows Update specifically, open an elevated Command Prompt and run net stop wuauserv, then net start wuauserv to restart the service under the correct security context. If the error is tied to a specific folder, verify that the SYSTEM and Administrators accounts have Full Control in the folder's Properties → Security tab.

Fix 2: Reset Windows Update Permissions and Cache

When 0x80070005 appears specifically during Windows Update, the Software Distribution folder is often the culprit. Open an elevated Command Prompt and run these commands in sequence: net stop wuauserv → net stop cryptsvc → ren C:\Windows\SoftwareDistribution SoftwareDistribution.old → net start cryptsvc → net start wuauserv. This forces Windows Update to rebuild its cache folder from scratch with correct permissions. Then run Windows Update again. If the error persists, also run net stop bits and net start bits to reset the Background Intelligent Transfer Service.

Fix 3: Fix Folder Permissions via the Security Tab

If a specific folder is blocking access, navigate to it in File Explorer, right-click, and select Properties → Security → Advanced. Click Change next to the Owner field and set it to your Administrator account. Tick Replace owner on subcontainers and objects and apply. Then return to the Security tab, click Edit, and ensure the SYSTEM account and your user account both have Full Control. Click Apply, confirm any prompts, and retry the operation. For registry key permission issues, the same process applies inside regedit.exe via right-click → Permissions.

How to Prevent 0x80070005 in the Future

Avoid manually editing ACLs on system folders unless you have a specific documented reason — incorrect permission changes are the primary cause of this error appearing on previously healthy systems. When installing software, always use official installers that request elevation properly rather than manually copying files into protected directories. Keep a standard user account for daily use and a separate Administrator account for system tasks, reducing the surface area for permission conflicts. If running automated scripts or scheduled tasks that touch system resources, ensure the task's Run As account has explicit permission grants rather than relying on inherited access. Periodically run sfc /scannow to catch ACL corruption before it causes visible errors.

#FC-003 502 BAD GATEWAY [STATUS: HIGH]

PLATFORMBrowser / Web Server / Proxy TYPEHTTP Server-Side Error RFCRFC 9110 — HTTP Semantics

How and Why a 502 Bad Gateway Happens

A 502 Bad Gateway error is an HTTP response status code indicating that a server acting as a gateway or proxy received an invalid, incomplete, or no response from an upstream server it needed to fulfil the request. The server returning the 502 is not the origin server — it is an intermediary such as an Nginx reverse proxy, Cloudflare, a load balancer, or a CDN edge node that sits between the user's browser and the actual application server.

Common causes include: the upstream application server (e.g., a PHP-FPM process or Node.js instance) has crashed or restarted; the origin server is overloaded and timing out before responding; a firewall rule is blocking the connection between the proxy and the origin; a deployment introduced a configuration error causing the upstream to return malformed headers; or DNS misconfiguration is routing the proxy's upstream requests to the wrong IP address. From the browser's perspective, this is always a server-side problem — nothing the user can fix directly on their machine.

Possible Fixes

Fix 1: Reload and Check for Upstream Outages

First, confirm the problem is not transient. Do a hard reload with Ctrl + Shift + R (Windows) or Cmd + Shift + R (Mac) to bypass any cached error page. Open the site in a different browser or an incognito window to rule out extension interference. Check the service's official status page or search Twitter/X for the domain name to see if others are reporting the same issue. Tools like downforeveryoneorjustme.com and isitdownrightnow.com provide a quick independent check. If it's a widespread outage, the fix is waiting for the hosting provider or CDN to restore service — nothing on your end will resolve it.

Fix 2: Flush DNS and Clear Browser Cache

If the 502 is isolated to your machine, stale DNS records or a corrupted browser cache can be the cause. In Chrome, navigate to chrome://net-internals/#dns and click Clear host cache. Then go to chrome://settings/clearBrowserData and clear cached images and files. In Windows, open an elevated Command Prompt and run ipconfig /flushdns followed by ipconfig /release and ipconfig /renew to reset your IP lease. Also try temporarily disabling any VPN or proxy service you have running — these add an extra routing hop that can produce 502s if the VPN server itself is having upstream issues.

Fix 3: Server-Side — Restart the Application Process

For developers or server administrators seeing 502s on their own infrastructure: the most common cause is the upstream application process dying silently. SSH into the server and check the application's process status — systemctl status php8.2-fpm or pm2 list for Node.js apps. If the process is stopped or in an error state, restart it with systemctl restart php8.2-fpm or pm2 restart all. Check application logs at /var/log/nginx/error.log or your app's log directory for the root crash reason. Also verify that the upstream timeout values in your Nginx config (proxy_read_timeout, fastcgi_read_timeout) are set high enough for your application's response time.

How to Prevent 502 Bad Gateway in the Future

For server operators, implement a process manager like PM2 (Node.js) or Supervisor (PHP/Python) with auto-restart enabled so crashed upstream processes recover without manual intervention. Configure health checks in your load balancer or reverse proxy so traffic is automatically rerouted away from unhealthy upstream nodes. Set up uptime monitoring via UptimeRobot or Better Uptime to receive immediate alerts when a 502 is detected. Review and raise timeout values to match your slowest legitimate request. For end users, keeping browsers updated and avoiding excessive extensions reduces the chance of client-side cache corruption producing false 502 reports.

#FC-004 ERR_CONNECTION_TIMED_OUT [STATUS: HIGH]

PLATFORMChrome / Edge / Browser TYPENetwork / TCP / DNS Failure NET_ERRERR_CONNECTION_TIMED_OUT

How and Why ERR_CONNECTION_TIMED_OUT Happens

ERR_CONNECTION_TIMED_OUT appears in Chromium-based browsers when the browser initiates a TCP connection to a remote server but receives no acknowledgement within the timeout window (typically 30 seconds). The browser successfully resolved the domain's IP address via DNS, sent a SYN packet to open the connection, but never received the SYN-ACK response that confirms the remote host is alive and accepting connections.

This differs from DNS_PROBE_FINISHED_NXDOMAIN (where the domain doesn't resolve at all) and from a connection refused error (where the server actively rejects the connection). Here the packets are simply disappearing — either dropped by a firewall, blocked by the user's router or ISP, intercepted by a security tool, or the destination server is genuinely unreachable due to being overloaded, offline, or misconfigured. It can also be caused by an incorrect system clock invalidating TLS handshakes, or a hosts file entry pointing the domain to a wrong or unreachable IP address.

Possible Fixes

Fix 1: Flush DNS and Reset Network Stack

Open Command Prompt as Administrator and run the following commands one by one, pressing Enter after each: ipconfig /flushdns clears cached DNS entries that may be pointing to an outdated IP. ipconfig /release and ipconfig /renew request a fresh IP lease from your router. netsh winsock reset resets the Windows socket layer, which can become corrupted by VPN software or malware. netsh int ip reset resets the TCP/IP stack to default. Reboot after running all four commands. This sequence resolves the majority of ERR_CONNECTION_TIMED_OUT cases that are caused by local network configuration drift rather than a server-side problem.

Fix 2: Change DNS Servers

Your ISP's default DNS servers can be slow or temporarily unresponsive, causing timeouts on resolution. Switch to a public DNS provider. Open Control Panel → Network and Internet → Network Connections, right-click your active adapter, select Properties → Internet Protocol Version 4 (TCP/IPv4) → Properties. Select Use the following DNS server addresses and enter 8.8.8.8 (Google Primary) and 8.8.4.4 (Google Secondary), or alternatively 1.1.1.1 and 1.0.0.1 (Cloudflare). Click OK, close the dialogs, then flush DNS again with ipconfig /flushdns in an elevated Command Prompt and retry the page.

Fix 3: Disable VPN, Proxy, and Check Windows Firewall

VPNs and proxy services route traffic through an intermediate server — if that server is overloaded or misconfigured, all connections through it will time out. Temporarily disable any active VPN or proxy and test the connection directly. In Chrome, go to Settings → System → Open your computer's proxy settings and ensure Use a proxy server is toggled off. Check Windows Defender Firewall → Advanced Settings → Outbound Rules for any rule blocking the browser or the destination port (80 or 443). Third-party antivirus products with web filtering modules (Kaspersky, Bitdefender, ESET) should also be temporarily paused to determine if they are the blocking layer. Re-enable them after testing.

How to Prevent ERR_CONNECTION_TIMED_OUT in the Future

Switch to a reliable DNS provider permanently using the steps in Fix 2 — public resolvers like Cloudflare (1.1.1.1) and Google (8.8.8.8) have significantly better uptime than most ISP-provided servers. Keep your router firmware updated, as outdated firmware is a common source of NAT table corruption that causes intermittent timeouts. If you use a VPN, choose a provider with a large server pool so you can easily switch nodes when one is congested. Review your Windows Firewall rules after installing new security software, as aggressive scanners occasionally add overly broad outbound blocking rules. Periodically clear your browser cache and DNS cache as part of routine maintenance to prevent stale entries accumulating over time.