Start by locating the failure boundary, not by rebooting. From a network that is independent of the server, test DNS, TCP ports, and HTTP separately. If only HTTP fails, investigate the web service or its dependencies. If SSH and other ports also fail, move to the provider’s serial, KVM, or physical console. Preserve console output, journal entries, memory statistics, and provider-event details before restarting anything; a blind reboot can erase the evidence needed to explain the crash.
Need a server to try this on? Antihost offers Linux VPS plans with NVMe storage, full root access and a 14-day free trial.
This runbook covers sudden website outages caused by DNS or networking, failed services, exhausted memory or swap, startup dependency errors, kernel panics, and hard crashes. It also shows how to make the next incident diagnosable with persistent logs, monitoring, snapshots, console access, and tested crash capture.
As an Amazon Associate I earn from qualifying purchases.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →1. Establish exactly what is down
Use an external machine, mobile connection, or another network segment so that a local routing or DNS problem does not look like a server failure. Record the first time you observed the outage and the time of each test.
Test name resolution
dig +short example.com
dig example.com A
dig example.com AAAACompare the result with the address you expect. An empty answer, an old address, or an unexpected IPv6 address can make a healthy host appear down. Test the authoritative DNS service or a second resolver if the result is inconsistent. DNS changes can also be delayed by TTL caching, so note the resolver and timestamp.
#1 Best Overall
- Read Before You Buy — No Video Output: These adapters support charging and USB 2.0 data transfer, but cannot transmit video signals. Except for standard USB webcams (which use USB data only), they are not compatible with HDMI/DisplayPort cables, video-capable USB-C hubs, or docking stations with video output.
- Convert USB-A Ports to USB-C: Designed to connect USB-C earphones, cables, flash drives, card readers, and other USB-C accessories to standard USB-A ports. Plug-and-play with no drivers or software required.
- Aluminum Alloy Housing: Built with a sturdy aluminum alloy shell that aids in heat dissipation and protects against daily wear and scratches. Designed to maintain a stable and secure connection.
- Compact & Travel-Friendly: The ultra-compact design allows the adapter to stay plugged into your device without blocking adjacent ports or adding bulk, reducing wear and tear on your original USB ports.
- 12-Month Warranty: Backed by a 12-month manufacturer warranty for peace of mind. Designed to meet strict quality control standards for reliable everyday performance.
Test TCP reachability
nc -vz example.com 80
nc -vz example.com 443
nc -vz example.com 22A refused connection means that something answered but no process accepted the connection (or a firewall actively rejected it). A timeout can indicate filtering, a routing failure, a stopped host, or a saturated machine. Test the server’s IP address as well as its hostname to separate DNS from transport problems.
Test HTTP independently of the browser
curl -I --connect-timeout 10 https://example.com
curl -v --connect-timeout 10 https://example.com/healthCheck the status code, TLS negotiation, redirect chain, and response time. A 502 or 503 generally means a proxy or web server is reachable but its upstream application is not. A connection timeout with failed SSH points to a broader host, network, or provider problem.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallMap the symptoms to a likely layer
| Observed result | Most useful next check | What it suggests |
|---|---|---|
| DNS is wrong or absent; IP tests work | DNS records, authoritative nameservers, TTLs | DNS or delegation problem, not necessarily a crashed server |
| Port 22 works; HTTP fails | systemctl status and service journals for the web server, proxy, and application |
Service, configuration, certificate, or dependency failure |
| All tested ports time out | Provider status, firewall, route, serial/KVM console | Network path, host hang, power event, or infrastructure failure |
| Console responds but normal services do not | Current-boot journal, failed units, memory and disk checks | Boot or service-startup failure |
| Console shows a panic or spontaneous reboot | Previous-boot journal and kdump/vmcore artifacts | Kernel, hardware, or low-level driver failure |
Do not repeatedly restart a host while its state is unknown. Each reboot can rotate volatile logs, terminate processes that still contain useful state, and overwrite clues about whether the machine hung, rebooted cleanly, or was forcibly reset.
2. When SSH fails, use an out-of-band console
If port 22 is unreachable, use the cloud provider’s serial console, KVM, web console, or a physical management interface. These paths do not depend on the server’s normal network stack or the SSH daemon. Confirm that you have console credentials and know which account has permission to open the session before an incident.
What to capture before a reboot
- The console’s visible panic, watchdog, filesystem, or OOM messages.
- The provider’s event timeline, including power actions, host migration, maintenance, or network incidents.
- The exact outage time and the last successful external test.
- Current process, memory, swap, disk, and mount information if the shell responds.
- Recent deployments, package upgrades, kernel changes, configuration edits, and scheduled jobs.
date -Is
uptime
uname -a
free -h
swapon --show
df -hT
ps aux --sort=-%mem | head -n 20
ps aux --sort=-%cpu | head -n 20
systemctl --failedCopy output to a separate system. If the console is interactive but the machine is badly overloaded, collect the smallest useful set first; expensive diagnostic commands can worsen an I/O or memory crisis.
When a reboot is unavoidable
Reboot only after you have captured the available evidence, or when the provider confirms an unresponsive host requires a reset. Record whether the action was a graceful reboot, forced power cycle, or hypervisor reset. That distinction matters when you later compare shutdown and boot records.
Rank #2
- 5-in-1 USB-C Hub: Experience comprehensive connectivity featuring a Power Delivery input, two USB-A 2.0 ports, a USB-A 3.0 port, and an HDMI port. (Note: The USB-C power delivery input port is only for connecting an external wall charger to power your laptop and cannot power peripheral devices.)
- 90W Pass-Through Charging: Achieve optimal charging with 90W pass-through power to your laptop, supported by a total input of 100W, with the hub reserving 10W for operational efficiency. (Note: Wall charger not included.)
- Quick Data Transfers: Accelerate your productivity with rapid data transfers using a high-speed 5Gbps USB 3.0 port and two 480Mbps USB 2.0 ports.
- 4K HDMI Display: Enhance your visual experience with a hub capable of delivering 4K resolution at 30Hz in both mirror and extend modes. Please note that this hub is compatible with MacBook (macOS 12 and newer), Windows 10 and 11, ChromeOS, and laptops equipped with DP Alt Mode and Power Delivery. Note: This device is not compatible with Linux.
- What You Get: Anker USB-C Hub (5-in-1, 4K HDMI), welcome guide, 18-month warranty, and our friendly customer service.
3. Inspect the current boot and the failing service
Once you have shell or console access, identify the service layer before changing configuration. Replace the example unit with the actual service name.
systemctl status example.service
systemctl is-enabled example.service
systemctl is-active example.service
systemctl --failedRead the current boot’s journal rather than relying only on a service’s own log file. systemd’s boot-debugging guidance recommends starting with:
journalctl -b
journalctl -p err -b
journalctl -u example.service -b
journalctl --since "2026-10-15 02:00" --until "2026-10-15 03:00"-b selects the current boot. -p err shows error and higher priorities, which is useful for an initial scan but can omit warnings that explain the lead-up. The unit filter isolates one service, while the time window prevents an incident review from being buried in unrelated history.
Validate before restarting
Use the service’s built-in configuration check when one exists. For example, validate a web server configuration with that server’s documented test command, then inspect its journal for the exact file and line that failed. A restart without validation can turn a recoverable process failure into a persistent configuration outage.
Check dependencies and listening sockets:
systemctl list-dependencies example.service
ss -ltnp
systemctl show example.service -p ActiveState,SubState,ExecMainStatus,ExecMainCodeIf a reverse proxy is healthy but the application socket is missing, follow the dependency chain instead of repeatedly restarting the proxy. If the service exits with a nonzero status, preserve the surrounding journal lines before attempting a fix.
4. Determine whether the machine rebooted or merely lost the network
A failed SSH connection does not prove that Linux crashed. The host may still be running with a broken route, a firewall change, a wedged SSH daemon, or a saturated CPU and disk. Compare boot identifiers and shutdown records.
Rank #3
- Sleek 7-in-1 USB-C Hub: Features an HDMI port, two USB-A 3.0 ports, and a USB-C data port, each providing 5Gbps transfer speeds. It also includes a USB-C PD input port for charging up to 100W and dual SD and TF card slots, all in a compact design.
- Flawless 4K@60Hz Video with HDMI: Delivers exceptional clarity and smoothness with its 4K@60Hz HDMI port, making it ideal for high-definition presentations and entertainment. (Note: Only the HDMI port supports video projection; the USB-C port is for data transfer only.)
- Double Up on Efficiency: The two USB-A 3.0 ports and a USB-C port support a fast 5Gbps data rate, significantly boosting your transfer speeds and improving productivity.
- Fast and Reliable 85W Charging: Offers high-capacity, speedy charging for laptops up to 85W, so you spend less time tethered to an outlet and more time being productive.
- What You Get: Anker USB-C Hub (7-in-1), welcome guide, 18-month warranty, and our friendly customer service.
List available boots
journalctl --list-bootsIn the standard numbering, boot 0 is the current boot, -1 is the immediately previous boot, and lower numbers refer to older boots. Inspect the previous boot directly:
journalctl -b -1
journalctl -p err -b -1
journalctl -u example.service -b -1Check recorded shutdown and reboot events
last -Fxn2 shutdown rebootA clean shutdown sequence followed by a new boot points toward an intentional or orderly restart. An abrupt end to the previous journal, with no shutdown messages, is consistent with a hard reset, power loss, host failure, or kernel panic, but it is not proof of one particular cause. Correlate it with provider events, console output, and crash-dump artifacts.
Why persistent journaling must be enabled in advance
Many installations keep the journal only in volatile storage. Create the persistent journal directory and flush the running journal before an incident:
sudo mkdir -p /var/log/journal
sudo systemd-tmpfiles --create --prefix /var/log/journal
sudo journalctl --flushSet retention deliberately in /etc/systemd/journald.conf (for example, limits for total and per-file size), then restart or reload journald according to your distribution’s procedure. Verify that records survive a planned reboot with journalctl --list-boots. Red Hat notes that previous-boot analysis is unavailable when persistence was not configured before the failure.
5. Check memory, swap, and disk exhaustion
A host that uses all physical memory and swap can become extremely slow because of heavy I/O, stop accepting connections, or invoke the kernel’s out-of-memory (OOM) killer. OOM messages identify a victim process, not necessarily the component that caused the pressure; a leak or workload spike may have started elsewhere.
Capture the state
free -h
vmstat 1 5
swapon --show
ps aux --sort=-%mem | head -n 25
journalctl -k -b | grep -i -E 'oom|out of memory|killed process'
df -hT
df -ihfree -h shows available RAM and swap. vmstat reveals sustained swap activity and I/O wait. Check both block-space and inode-space exhaustion; either can prevent services from writing files, sockets, or logs.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Rank #4
- Dual Converters, Infinite Potential：Includes 2× USB C male to USB A female adapters and 2× USB A male to USB C female adapters. Perfect for a wide range of uses—tablets with Bluetooth keyboards, expand USB ports on macbook, and more. Two different converters for all your daily needs
- Next-Level 10Gbps & 3A Charging: No more slow 480Mbps, this usb to usb c adapter has a transfer speed of up to 10Gbps, allowing you to do more transferring in less time. This usb adapter fits both USB A and USB C charger, supporting up to 3A fast charging
- Upgraded Exquisite Craftsmanship: With an aluminum alloy housing and metal connector, the usbc to usb adapter is extremely durable and sturdy. Rigorously tested to withstand more than 10,000 times of plugging and unplugging, ensuring long-lasting performance
- Broad Compatible: The usb c to usb adapter widely supports all USB C/ USB A devices like laptops, tablets, cellphones, car chargers, and phone chargers. Such as compatible with MacBook Pro/Air 2023/2022, Thunderbolt 4/3 Devices,Apple MagSafe Watch 9/8/7/SE/Ultra, iPad Pro 2022/2021, Samsung Galaxy S23/S20/S10, and iPhone 17/16/15 Pro. Plug and play
- Please Note: To reach 10Gbps speed, keep the cable under 3.3 ft. For USB A Male to USB C adapters, try flipping the USB C connector. USB C Male to USB A adapters support bidirectional 10Gbps transfer within 3.3 ft
Choose a fix based on the cause
- Workload spike: reduce concurrency, pause a batch job, rate-limit a queue, or temporarily disable a nonessential worker.
- Memory leak: identify the process and deploy a corrected version; a restart is only a temporary containment measure.
- Insufficient capacity: add RAM or move the workload after measuring a representative peak.
- No swap or undersized swap: add appropriately sized swap where operationally acceptable, while recognizing that swap is slower than RAM and can extend an outage under sustained pressure.
- Disk or inode exhaustion: remove or rotate known-safe files, expand storage, and fix the retention policy. Do not delete unknown database or queue files merely to make space.
After containment, preserve the OOM lines and process information. Record whether the kernel killed the web server, database, SSH daemon, or another process, then verify service health and memory behavior over time.
6. Recover a host that fails during normal startup
If the kernel boots but the machine hangs or fails while starting ordinary services, use the console to select a rescue target. systemd documents adding systemd.unit=rescue.target to the kernel command line; traditional systems may refer to runlevel 1.
- Open the serial, KVM, or physical console and interrupt the bootloader during the maintenance window.
- Edit the Linux kernel entry and append
systemd.unit=rescue.target. - Boot the edited entry and authenticate at the rescue shell.
- Read
journalctl -b,systemctl --failed, and the journal for the last unit that started. - Correct the specific configuration, mount, package, or dependency problem. Validate the affected service before enabling it again.
- Return to the normal target only after the failing component and its prerequisites are understood.
Rescue mode is for diagnosis and repair, not a substitute for finding the cause. A broken filesystem mount, expired certificate, invalid environment file, missing secret, or dependency ordering error can all stop a normal boot without indicating a kernel crash.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.7. Investigate a kernel panic or hard crash
Ubuntu defines a kernel crash dump as a portion of volatile RAM copied to disk when kernel execution is disrupted. Documented triggers include a kernel panic, non-maskable interrupt, machine-check exception, hardware failure, and manual intervention. A panic message on the console or an abrupt reboot with no clean shutdown record warrants crash-dump investigation.
Free tools Windows power users keep installed
One-click scans. No signup required.
Check kdump readiness
kdump-config show
systemctl status kdump-toolsAlso verify that the kernel command line reserves crash memory (the crashkernel setting), that the dump destination has enough space, and that the configured storage remains available when the primary system is damaged. Ubuntu documents crash artifacts under /var/crash. An “not ready” state can result from insufficient reserved memory, so treat readiness as a capacity check rather than assuming the package is configured correctly.
Interpret the evidence
- vmcore present: preserve it and its accompanying metadata before rotation or cleanup.
- Kernel log names a driver or hardware exception: correlate with recent kernel updates, firmware changes, hardware alerts, and provider events.
- No dump and no persistent previous boot: state that the cause cannot be reconstructed confidently from the remaining evidence; do not manufacture certainty from the reboot alone.
Analyze a dump on a separate machine with the matching kernel debug symbols and the distribution’s documented crash-analysis tools. Keep the original artifact unchanged and work from a copy.
Best Value
- 5-in-1 Connectivity: Equipped with a 4K HDMI port, a 5 Gbps USB-C data port, two 5 Gbps USB-A ports, and a USB C 100W PD-IN port. Note: The USB C 100W PD-IN port supports only charging and does not support data transfer devices such as headphones or speakers.
- Powerful Pass-Through Charging: Supports up to 85W pass-through charging so you can power up your laptop while you use the hub. Note: Pass-through charging requires a charger (not included). Note: To achieve full power for iPad, we recommend using a 45W wall charger.
- Transfer Files in Seconds: Move files to and from your laptop at speeds of up to 5 Gbps via the USB-C and USB-A data ports. Note: The USB C 5Gbps Data port does not support video output.
- HD Display: Connect to the HDMI port to stream or mirror content to an external monitor in resolutions of up to 4K@30Hz. Note: The USB-C ports do not support video output.
- What You Get: Anker 332 USB-C Hub (5-in-1), welcome guide, our worry-free 18-month warranty, and friendly customer service.
8. Do not trigger a crash test over SSH
Ubuntu warns that SysRq crash testing reboots the machine and can cause data loss under load. The SSH connection will be lost, so testing requires a local or serial console, an idle or lightly loaded host, an approved maintenance window, current backups, and a recovery plan.
echo c > /proc/sysrq-triggerNever run that command on a production server as an improvised diagnostic. If you need to verify kdump, schedule a controlled maintenance event, confirm that the console is reachable, check that crashkernel memory and dump storage are ready, and define how the resulting dump will be collected and deleted after analysis.
Recommended Free Tools
9. Select the least-destructive recovery path
| Recovery approach | Evidence preserved | Access path | Best fit | Main risk |
|---|---|---|---|---|
| Fix or restart one failed service | High when journals are copied first | SSH or console | HTTP-only outage with a responsive host | Restart can erase transient state or hide the original failure |
Boot to rescue.target |
High, including startup ordering | Serial/KVM/physical console | Service, mount, or dependency failure during boot | Editing the wrong unit or mount can make recovery harder |
| Graceful reboot | Moderate if logs and metrics are saved first | SSH or console | Known software deadlock with a documented recovery plan | Volatile evidence and process state are lost |
| Forced reset or provider power cycle | Low unless console and kdump capture state | Provider console | Host is completely unresponsive | Filesystem corruption, data loss, and missing root-cause evidence |
| Restore snapshot or backup | Depends on snapshot and incident records | Provider or backup console | Corruption or unrecoverable configuration | Recent writes may be lost; the underlying defect may recur |
Choose the fastest option that still preserves enough information to prevent a repeat. Restoration and diagnosis are separate objectives: bring service back, then investigate with the captured records.
10. Build a complete incident record
Create one time-ordered record rather than scattered notes. Include:
- First observed outage time, timezone, and the independent networks used for testing.
- DNS answers, TCP results, HTTP status and latency, and the exact hostname or IP tested.
- Provider event IDs, console screenshots or transcripts, and whether a reset was graceful or forced.
- Current and previous boot IDs from
journalctl --list-boots. - Relevant
journalctlexcerpts, service status, kernel messages, and the earliest OOM or panic lines with surrounding context. free -h,vmstat, process, disk-space, inode, and mount readings.- Recent deployments, package and kernel changes, scheduled jobs, certificate renewals, and configuration edits.
- Any kdump status,
vmcorepath, dump size, and analysis location.
Preserve the earliest relevant lines, not just the final “killed process” or “panic” message. A single isolated line rarely establishes causality.
Quick Recap
11. Prevent the next unexplained outage
Make logs survive a reboot
- Enable persistent journald storage under
/var/log/journal. - Set retention and size limits that fit the disk.
- Verify that
journalctl --list-bootsshows the previous boot after a planned restart. - Forward critical logs to separate storage if losing the server would also lose its journal.
Monitor the failure signals
- Memory availability, swap-in and swap-out activity, and OOM-killer events.
- Disk space, inode usage, I/O wait, and filesystem errors.
- HTTP status, latency, TLS expiry, DNS resolution, and TCP reachability from outside the host.
- Service health, process restarts, failed systemd units, and unexpected reboot events.
Protect recovery options
- Keep tested backups and provider snapshots; document retention and restoration points.
- Document serial, KVM, and provider-console access, including who owns the credentials.
- Configure kdump where the distribution and hardware support it; reserve adequate crash memory and dump storage.
- Practice a restoration and log-collection runbook during a maintenance window.
- After every kernel or major package change, confirm that the normal boot, previous-boot journal, monitoring, and console access still work.
12. Quick command reference
| Question | Command |
|---|---|
| What failed in the current boot? | journalctl -p err -b |
| What did one service report? | journalctl -u example.service -b |
| What happened before the reboot? | journalctl -b -1 |
| Which boots are retained? | journalctl --list-boots |
| Was there a recorded shutdown? | last -Fxn2 shutdown reboot |
| Is memory or swap exhausted? | free -h, vmstat 1 5, swapon --show |
| Are filesystems full? | df -hT and df -ih |
| Is kdump ready? | kdump-config show and systemctl status kdump-tools |
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problems
