- Home
- Skills
- DevOps & Cloud
- linux server rca
More screenshots
Works with the AI Agents you already use
linux server rca
Diagnose Linux server failures, systemd crashes, and disk I/O bottlenecks with deterministic RCA and recovery steps.
$12
linux server rca
Example session with this skill installed
I have an Ubuntu server where the 'web-api' service won't start, showing status=203/EXEC. Also, 'df -h' shows 40% free but I'm getting 'No space left on device' errors. Analyze these logs.
- Read your context and instructions
- Compiled the linux server rca
I have analyzed the service failure and storage telemetry. The system is suffering from a binary execution failure and inode exhaustion.
Root Cause Analysis Report
| Metric / Component | Observed Value | Threshold / Status | Assessment |
|---|---|---|---|
| Service Status | web-api.service | status=203/EXEC | CRITICAL: Missing/invalid binary path |
| Disk Space | 40% Free | > 10% | NORMAL: Byte capacity is sufficient |
| Inode Usage | 100% | < 90% | CRITICAL: Inode exhaustion detected |
| Process State | ps D state found | N/A | WARNING: High I/O wait on /var/spool |
Executive Incident Summary
The web-api service is failing to start because the systemd ExecStart path points to a file that does not exist or lacks execute permissions. Simultaneously, the filesystem is rejecting new writes despite available byte space due to 100% inode exhaustion, likely caused by a runaway spool queue in /var/spool/postfix.
Step-by-Step Remediation Plan
Fix Inode Exhaustion: Identify and clear accumulated small files in the spool directory without deleting active sessions.
find /var/spool/postfix/maildrop -type f -mtime +7 -delete
- Verify Binary Path: Check the absolute path in the systemd unit file.
systemctl cat web-api.service | grep ExecStart ls -l /usr/local/bin/web-api-binary - Apply Fix: Correct the path or permissions and reload.
chmod +x /usr/local/bin/web-api-binary systemctl daemon-reload systemctl start web-api.service
Permanent Hardening Recommendations
- Implement a
logrotatepolicy for the application logs in/var/log/web-api/. - Set
TasksMax=andMemoryMax=in theweb-api.serviceunit to prevent resource hogging. - Add
df -imonitoring to your Prometheus/Grafana alerts.
Next steps
- Clear the postfix maildrop queue to restore write capabilities.
- Verify the
web-apibinary location matches thesystemdunit configuration. - Audit the cron jobs generating high volumes of small files in
/tmp.
Connects securely to your tools. The creator never sees your data.
About this skill
Server downtime costs money, and trial-and-error troubleshooting in a production environment is a liability. This skill provides a deterministic diagnostic framework for Linux server degradation, replacing manual log grepping with a structured Root Cause Analysis (RCA) pipeline.
What it does
- Systemd Unit Diagnostics identifies failed services by mapping specific exit codes like 203/EXEC or 217/USER to their underlying configuration or permission issues.
- Kernel Event Analysis scans dmesg and journalctl for OOM killer triggers, memory starvation, and process-specific memory leaks.
- Storage Bottleneck Detection correlates block storage capacity with inode table exhaustion to explain write failures on disks that appear half-empty.
- I/O Wait Isolation targets processes stuck in uninterruptible disk sleep (D state) and identifies hardware-level filesystem corruption or read-only remounts.
- Sanitized Reporting generates incident summaries, remediation plans, and Jira-ready escalation briefs while masking sensitive hostnames and IPs.
How it works
- Telemetry Ingestion through the bundled bash collector script, the Python parser, or direct CLI output snippets from tools like
df,free, andiostat. - Diagnostic Branching across five failure modes: service exit codes, memory, inodes/disk, I/O bottlenecks, and filesystem health.
- Evidence Mapping where observed metrics are compared against baseline thresholds and known kernel error patterns.
- Structured Deliverable Generation using a standardized RCA template including safe recovery commands and long-term hardening steps.
Frameworks & tools
This skill works with RHEL, Ubuntu, Debian, and Rocky Linux systems. It utilizes native Linux toolsets including systemd, journalctl, dmesg, procfs, and standard storage utilities.
Why this beats prompting it yourself
Generic prompts often suggest destructive actions like reboot -f or rm -rf when faced with disk pressure. This skill enforces strict read-only telemetry rules and distinguishes between byte-space and inode exhaustion, preventing accidental data loss during triage.
Use cases
- Diagnose why a critical application service is stuck in a crash loop despite having valid configuration.
- Identify which background daemon triggered the OOM killer during a midnight traffic spike.
- Resolve "Disk Full" errors when
df -hshows plenty of available gigabytes. - Triage unresponsive servers reporting high I/O wait times before escalating to hardware vendors.
Known limitations
This skill is designed for standalone Linux hosts or cluster nodes; it requires manual verification of cluster quorum (etcd/Galera) before recommending service restarts in HA environments.
Requires manual verification of cluster quorum for HA systems. Sanitization masks IPs to RFC 5737 prefixes. Focused on systemd-based distributions (RHEL, Ubuntu, Debian, Rocky).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 9 days ago
Requires
Permissions
Allowed Hosts