Skip to main content

Alerts Quick Reference & Cheatsheet | Nife Deploy

Quick answers to common alert questions.


I Want To...​

Create an Alert Rule​

Alerts → Alert Rules → New Rule
↓
Choose what to monitor (CPU, errors, etc.)
↓
Set threshold (> 80%, < 10%, etc.)
↓
Choose severity (Critical/Warning/Info)
↓
Name it clearly
↓
Save and Enable

Time: 2-3 minutes


Get Alert Notifications​

Alerts → Alert Config → Add Channel
↓
Choose type (Email/Slack/PagerDuty/Webhook)
↓
Enter your details
↓
Test the channel
↓
Link to your alert rule

Time: 5-10 minutes


Respond to a Firing Alert​

Get notification
↓
Click link or go to SRE → Alerts
↓
Find the alert with Firing status
↓
Click Acknowledge
↓
Investigate and fix the issue
↓
Click Resolve

Time: Varies on issue complexity


Find a Specific Alert​

Use the filters:

Status: Firing / Acknowledged / Resolved
Severity: Critical / High / Medium

Or scroll through the list at SRE → Alerts


Fix Too Many Alerts​

Step 1: Find the noisy rule

Alerts → Alert Rules → Look for frequently triggered rule

Step 2: Increase threshold

Example: Change from CPU &gt; 70% to CPU &gt; 85%

Step 3: Test and adjust as needed


Test if Notifications Work​

Alerts → Alert Config
↓
Find your channel
↓
Click "Test"
↓
Check if you received it

Disable Alerts Temporarily​

Alerts → Alert Rules
↓
Find the rule
↓
Toggle Enabled OFF
↓
(Remember to turn back ON later!)

Quick Reference Tables​

Alert Statuses​

StatusIconColorMeaningAction
Firing🔔RedActive alert, needs attentionAcknowledge & Fix
Acknowledged⏱️YellowSomeone investigatingWait or help
Resolved✓GreenIssue is fixedNone

Severity Levels​

SeverityIconWhenResponse Time
Critical🔴Immediate action neededMinutes
Warning🟠Important, needs attention1 hour
Info🟡FYI informationAs time allows

Notification Channels​

ChannelSpeedBest ForSetup Time
EmailSlowDocumentation1 min
SlackFastTeam visibility2 min
PagerDutyImmediateCritical alerts5 min
WebhookDependsCustom integration10+ min

Common Alert Examples​

Website/API​

AlertThresholdSeverity
Service DownHTTP 503Critical
Slow Response> 5 secondsWarning
High Error Rate> 5%Warning
CPU High> 80%Warning

Database​

AlertThresholdSeverity
Service DownNot respondingCritical
CPU High> 80%Warning
Memory High> 90%Warning
Low Disk Space< 10%Critical

Infrastructure​

AlertThresholdSeverity
Server DownUnreachableCritical
Disk Full< 5%Critical
High Latency> 100msWarning
Network Error> 1% lossWarning

Troubleshooting​

Alert Won't Fire​

Check:

  1. Is rule Enabled? (toggle should be ON)
  2. Is threshold correct? (Test with extreme value)
  3. Is metric actually hitting threshold?

Fix:

  • Enable the rule
  • Lower threshold temporarily to test
  • Verify correct metric is selected

Not Getting Notifications​

Check:

  1. Click "Test" on your channel
  2. Did you receive test?
  3. Check email spam folder
  4. Check Slack channel settings
  5. Verify email/phone is correct

Fix:

  • Resend test
  • Check spam folder
  • Verify email/channel settings
  • Check connection/credentials

Too Many Alerts​

Solutions:

  1. Increase threshold (less sensitive)
  2. Disable low-priority rules
  3. Use digest instead of single emails
  4. Combine related alerts

Example:

Before: CPU &gt; 70% = 50 alerts/day
After: CPU &gt; 85% = 5 alerts/day

False Alarms​

Why: Threshold too sensitive

Fix:

  1. Review when it fired
  2. Increase threshold
  3. Make condition more specific
  4. Or delete if not important

Duplicate Alerts​

Problem: Same issue creates multiple alerts

Fix:

  1. Delete duplicate rules
  2. Keep only one rule per metric
  3. Or adjust thresholds so only one fires

Time-Saving Tips​

Fastest to Respond​

Get Slack notification (2 sec)
↓
Click link (1 sec)
↓
Acknowledge (5 sec)
↓
Investigate (5-30 min)
↓
Resolve (5 sec)
Total: < 1 minute to acknowledge

Fastest to Create Rule​

New Rule → Pick template → Set threshold
→ Name it → Save → Enable
Total: ~1 minute

Fastest to Test Channel​

Alert Config → Find channel → Click Test
→ Receive notification
Total: ~30 seconds

Key Locations​

NeedPath
Create ruleAlerts → Alert Rules
Configure notificationsAlerts → Alert Config
View alertsSRE → Alerts
Help? button on page

Common Mistakes​

MistakeImpactFix
Too many alertsAlert fatigueIncrease thresholds
Vague namesConfusionUse specific names
No testingAlerts don't workAlways test
Never adjustingGets worse over timeReview monthly
Not documentingTeam confusionDocument each rule

Best Threshold Examples​

CPU Usage​

Normal: 20-50%
Peak: 60-70%
Alert: 80% ← Not too sensitive
Critical: 95% ← Only if really bad

Memory Usage​

Normal: 30-60%
Peak: 70-80%
Alert: 90% ← Getting concerning
Critical: 95% ← Critically high

Response Time​

Normal: 100-300ms
Slow: 500-1000ms
Alert: 2000ms ← Warning level
Critical: 5000ms ← Very slow

Error Rate​

Normal: 0-0.5%
Alert: 1% ← One in hundred
Critical: 5% ← One in twenty

Checklist: First Alert Setup​

  • Created alert rule
  • Set realistic threshold
  • Chose appropriate severity
  • Added clear name
  • Set up notification channel
  • Tested notification
  • Linked rule to channel
  • Enabled the rule
  • Documented the rule
  • Tested end-to-end

Do's and Don'ts​

Do ✅​

  • ✅ Create alerts for important things
  • ✅ Use clear, specific names
  • ✅ Test before relying
  • ✅ Respond quickly to alerts
  • ✅ Review and adjust regularly

Don't ❌​

  • ❌ Create too many rules
  • ❌ Use vague names
  • ❌ Skip testing
  • ❌ Ignore alerts
  • ❌ Let system grow stale

Quick Definitions​

Alert Rule Defines WHEN an alert should fire (the condition)

Notification Channel Defines HOW you get notified (email, Slack, etc.)

Severity How serious the alert is (Critical/Warning/Info)

Threshold The value that triggers the alert (> 80%, < 10%, etc.)

Status Current state of alert (Firing/Acknowledged/Resolved)


Response Times​

ActionTime
Create rule2-3 min
Setup channel5-10 min
Acknowledge alert30 sec
Resolve alert30 sec
Test channel30 sec

NeedContact
Questionssupport@nife.io
Help on page? button
DocumentationAlerts Overview
Best PracticesBest Practices


Last Updated: January 2026

Need Help? support@nife.io