# Troubleshooting Mysterious Website Downtime: A Case Study on VPS, DNS, and Hosting Expiry Troubleshooting Website downtime can be a frustrating experience, especially when there’s no clear sign of what went wrong. Recently, we encountered an issue where two websites hosted on our [VPS](https://go.webdevstory.com/namecheap-vps) were behaving differently- one was accessible, while the other was completely unreachable. This led us on a deep troubleshooting journey to diagnose and fix the problem. ## Initial Troubleshooting Steps for Website Downtime * **Two websites on the same VPS were down, but** not in the same way. * **Other websites on the same VPS and the same cPanel account were working fine**, resolving correctly, and loading without issues. * **However, one of the downed websites returned an NXDOMAIN error**, indicating that its DNS records did not contain the domain name. * There was no DNS resolution, server response, or immediate explanation for why this happened. ## 🛠 Step 1: Identifying the Problem * We first ran `nslookup` and `dig` to check DNS resolution. * The **NXDOMAIN** error meant the domain was not resolving. * Checked the **WHOIS record** to confirm domain status and name servers. * Discovered the domain was using **parked nameservers** instead of pointing to our VPS. ## 🛠 Step 2: Ensuring the VPS is Running Before jumping into DNS-related issues, we verified that our VPS was running. If the VPS itself was down, that could explain why the sites weren’t loading. We ran tests on our server: * Checked system uptime and resource usage to ensure no crashes or outages. * Loaded other sites hosted on the same VPS to confirm they were working. * Confirmed that this was a site-specific issue rather than a [server-wide problem](https://www.webdevstory.com/server-troubleshooting/). Since other sites on the VPS were working perfectly fine, we ruled out a VPS failure and focused on investigating domain-specific problems. ## 🔧 Step 3: Investigating Potential Causes ![DNS propagation check showing missing A records globally](https://miro.medium.com/v2/resize:fit:1400/1*NN-mF1RXOC27IUO8SAeaWA.png align="left") ## Domain Expiry Check * Used [**WHOIS**](https://who.is/) to confirm the domain was active (not expired). * Verified the **registration details**. ## DNS Propagation Check * Used [whatsmydns.net](https://www.whatsmydns.net/) to check global DNS resolution. * Found that the **A record was missing everywhere**. ## Checking Web Hosting & Server Status * Verified that the **VPS was running**. * Checked / status to confirm the **web server was active**. * Ensured the **firewall wasn’t blocking access**. ## 🛠 Step 4: Different Causes for Each Website’s Issue ![Domain name pointing to parked nameservers instead of VPS](https://miro.medium.com/v2/resize:fit:1400/1*RCZleeIff7fLLL37oVnYuA.png align="left") While both websites had been moved to our hosting environment, they still had some time left on their previous hosting plans, and we leveraged their **cPanel zone editor** to host them from our VPS. After diagnosing the problem, we found that both websites had different underlying issues: * **One website’s DNS records were completely removed** when the previous hosting provider expired, making it impossible to resolve. * **The other website changed its nameservers**, pointing to a location different from our VPS. Although both websites faced similar downtime issues, the root causes were distinct and required different solutions. ## 🛠 Step 5: Fixing the DNS Issue ## Updating the Nameservers * Logged into the **domain registrar** and replaced the **parked nameservers** with the correct ones from our hosting provider. * Added an **A record** pointing to our **VPS IP**. ## Verifying Propagation * After updating, we **waited for DNS changes to propagate**. * Used CLI (`nslookup`, `dig`) and online tools to **track progress**. ## Clearing DNS Cache To speed up the local resolution, we cleared the DNS cache: ```bash sudo systemctl restart systemd-resolved # Linux ipconfig /flushdns # Windows ``` ## 🛠 Step 6: When Hosting Providers Delete DNS Records ## Why Does This Happen? * Some hosting companies **automatically remove all DNS zone records** when a hosting plan expires. * Even if the **domain is registered**, the system wipes out the DNS settings, making the website inaccessible. * Users assume that just pointing the domain to a new hosting provider will work, but without DNS records, nothing resolves. ## How to Fix It 1 — Contact the Previous Hosting Provider * If your previous hosting still controls the nameservers, ask them to **restore the DNS zone**. * You must manually set up DNS records if they refuse or delete everything permanently. ## 2 — Manually Add New DNS Records If the old provider deleted all records, you need to configure them manually: * **A Record** → Points the domain to the new hosting IP. * **CNAME Record** → For subdomains like `www`. * **MX Records** → If using email hosting. * **TXT Records** → For email authentication (SPF, DKIM). ## 🚀 Step 7: Confirming the Fix * After propagation was completed, we **reran tests**. * The domain is now **properly resolved**. * The website was **accessible again**. ## 🔑 Key Lessons from Troubleshooting Website Downtime * Always check **DNS records, WHOIS status, and propagation** when facing downtime. * VPS and DNS issues **are not always the same; identify** the root cause before applying fixes. * Hosting providers may **delete DNS records upon expiration**, so it’s crucial to back up DNS settings. * Monitor **nameservers and A records** after domain migrations to avoid misconfiguration. Note: Some links on this page might be affiliate links. If you make a purchase through these links, I may earn a small commission at no extra cost to you. Thanks for your support!