Workflow Settings

Workflow Settings let you customize how automated workflows behave for your organization. Configure business hours and holiday calendars to control when workflows run, set up SLA tracking with escalation deadlines, define retry policies for failed actions, and monitor workflow backlogs to catch bottlenecks early.

Accessing Workflow Settings

Navigate to Settings → Workflow Settings to configure these options. Admin or Owner role required.

---

Holiday Calendar Settings

Control when workflows consider “business days” for scheduling and deadlines.

Calendar Types

Calendar	Description	Best For
US Federal	US federal holidays (10 days/year)	US-based organizations
US Bank	Banking holidays (12 days/year)	Financial services
Custom	Only your specified dates	International or unique schedules

Configuration Options

Setting	Description	Default
Enabled	Turn holiday handling on/off	On
Calendar Type	Base calendar to use	US Federal
Custom Holidays	Additional dates to treat as holidays	Empty
Skip Holidays	Skip workflow deadlines on holidays	On
Skip Weekends	Skip Saturday/Sunday	On

Adding Custom Holidays

1. Navigate to Settings → Workflow Settings
2. In the Holiday Settings section, click Add Holiday
3. Enter:
   • Date: The holiday date
   • Name: Description (e.g., “Company Retreat Day”)
4. Click Save

Custom holidays supplement the base calendar—you don’t need to re-add federal holidays.

How Holidays Affect Workflows

When a workflow action is scheduled to occur on a holiday or weekend:

Scheduled: Friday 5 PM → Review due Monday
If Monday is a holiday → Review due Tuesday

Scheduled: Invoice processing every day
Skip weekends ON → Processes Mon-Fri only
Skip holidays ON → Skips federal holidays

Common Scenarios

Financial services: Use “US Bank” calendar to align with bank processing days.

International team: Use “Custom” calendar with only your office closures.

24/7 operations: Disable “Skip Holidays” and “Skip Weekends” for continuous processing.

---

Business Hours Settings

Define your organization’s working hours so workflow deadlines, SLA timers, and escalations respect your actual business schedule.

Configuration Options

Setting	Description	Default
Start Time	Business day start	9:00 AM
End Time	Business day end	5:00 PM
Working Days	Days of the week considered business days	Monday-Friday
Timezone	Organization timezone	US/Eastern

Business hours work with holiday calendars — SLA timers and escalation deadlines only count time during business hours on non-holiday working days.

---

SLA Tracking Settings

Set per-state SLA deadlines to track how long workflow items spend in each stage and escalate when deadlines are missed.

Configuration Options

Setting	Description	Default
SLA Deadline	Maximum time allowed in a workflow state	None
Warning Threshold	Percentage of SLA at which to warn	80%
Escalation Action	What happens when SLA is breached	Notify owner
Business Hours Only	Count only business hours toward SLA	On

How SLA Tracking Works

Invoice enters "Pending Approval" state
↓
SLA: 2 business days
↓
After 1.6 business days (80%) → Warning sent to approver
↓
After 2 business days → Escalation triggers (reassign, notify manager, etc.)

SLA deadlines respect business hours and holiday settings — an item entering “Pending Approval” at 4 PM Friday with a 4-hour SLA doesn’t breach until 1 PM Monday.

---

Retry Policy Settings

Configure how workflows handle temporary failures and retry failed actions.

Backoff Strategies

Strategy	Behavior	Best For
Exponential	Delays increase exponentially (30s → 60s → 120s)	API rate limits, external services
Linear	Fixed increase between retries (30s → 60s → 90s)	Predictable recovery times
None	Same delay every retry	Simple failures

Configuration Options

Setting	Description	Default
Max Retries	Maximum retry attempts	3
Retry Delay	Initial delay before first retry	60 seconds
Backoff Strategy	How delay increases	Exponential
Backoff Multiplier	Factor for exponential/linear increase	2.0x
Max Delay	Maximum wait between retries	1 hour
Retry on Errors	Which error types trigger retry	timeout, rate_limit, temporary

Error Types

Error Type	Description	Auto-Retry
timeout	Request timed out	Yes
rate_limit	Too many requests	Yes (with backoff)
temporary	Temporary service issue	Yes
network	Connection problems	Yes
processing	Processing failed	Optional
validation	Invalid data	No
permission	Access denied	No

Example Configurations

High reliability (financial):

Max Retries: 5
Retry Delay: 30 seconds
Backoff Strategy: Exponential
Backoff Multiplier: 1.5x
Max Delay: 30 minutes
Retry on: timeout, rate_limit, temporary, network

Quick fail (real-time):

Max Retries: 2
Retry Delay: 10 seconds
Backoff Strategy: None
Max Delay: 10 seconds
Retry on: timeout only

Conservative (sensitive operations):

Max Retries: 2
Retry Delay: 5 minutes
Backoff Strategy: None
Max Delay: 5 minutes
Retry on: temporary only

How Retries Work

Action fails with "timeout" error
↓
Wait 60 seconds (initial delay)
↓
Retry #1 → fails again
↓
Wait 120 seconds (60 × 2.0 multiplier)
↓
Retry #2 → fails again
↓
Wait 240 seconds (120 × 2.0 multiplier)
↓
Retry #3 → succeeds! ✓

If all retries fail, the workflow enters “Failed” state and can be manually retried or the error investigated.

---

Backlog Monitoring Settings

Monitor workflow queue health and get alerts when processing falls behind.

Why Monitor Backlog?

Large backlogs indicate:

• Processing bottlenecks
• Integration issues
• Capacity problems
• Configuration errors

Early alerts prevent:

• Missed deadlines
• Customer delays
• Compliance issues
• Data staleness

Configuration Options

Setting	Description	Default
Enabled	Turn backlog monitoring on/off	On
Threshold	Items pending before alerting	100
Alert Channels	Where to send alerts	Email
Custom Emails	Additional recipients	Empty
Webhook URL	External service webhook	None
Check Interval	How often to check	15 minutes
Alert Cooldown	Minimum time between alerts	60 minutes

Alert Channels

Channel	Description	Setup
Email	Email to admins	Automatic
Custom Emails	Additional recipients	Add addresses
Webhook	POST to external URL	Provide webhook URL

Webhook Payload

When using webhook alerting, Muin sends:

{
  "alert_type": "workflow_backlog",
  "tenant_id": "your-tenant-id",
  "current_count": 150,
  "threshold": 100,
  "severity": "warning",
  "oldest_pending_at": "2025-01-15T10:30:00Z",
  "timestamp": "2025-01-15T12:00:00Z"
}

Severity Levels

Severity	Condition	Recommended Action
Info	Count at 75-90% of threshold	Monitor
Warning	Count at 90-100% of threshold	Investigate
Critical	Count exceeds threshold	Immediate action

Best Practices

Set appropriate thresholds:

• Low volume organizations: 25-50 items
• Medium volume: 50-100 items
• High volume: 100-500 items

Use multiple channels:

• Email for all alerts
• Webhook for critical systems (PagerDuty, Slack)

Tune check intervals:

• Real-time critical: 5 minutes
• Standard operations: 15 minutes
• Background processing: 30-60 minutes

---

Viewing Backlog Status

The Workflow Settings page shows current backlog status:

Status	Indicator	Meaning
Healthy	Green	Below threshold
Warning	Yellow	Approaching threshold
Critical	Red	Exceeds threshold

Information displayed:

• Current pending count
• Configured threshold
• Oldest pending item age
• Last alert sent time

---

Applying Settings

Immediate Effect

Changes to settings take effect immediately for new workflow actions:

• Holiday settings affect scheduling from this point forward
• Retry settings apply to new failures
• Backlog alerts use new thresholds

Existing Items

In-progress workflows continue with their original settings. To apply new settings:

• Retry policy: Cancel and restart workflow
• Holiday settings: Reschedule pending actions

---

Permissions

Action	Required Role
View settings	Admin, Owner
Modify settings	Admin, Owner
Receive backlog alerts	Admin, Owner (auto), Custom emails

---

Troubleshooting

Workflows Not Skipping Holidays

1. Verify Holiday Settings are Enabled
2. Check the holiday date is correct
3. Confirm the workflow uses business day scheduling
4. Manual “Run Now” actions bypass holiday settings

Too Many Retries

1. Review which error types are enabled
2. Consider reducing max retries
3. Check if errors are truly temporary
4. Some errors (validation, permission) should not retry

Backlog Alerts Not Sending

1. Verify Backlog Monitoring is Enabled
2. Check current count exceeds threshold
3. Verify email addresses/webhook URL
4. Check cooldown hasn’t prevented recent alert

Alert Fatigue

1. Increase threshold to realistic level
2. Extend alert cooldown period
3. Use severity filtering in your monitoring system
4. Fix root cause of backlog growth

---

Related Documentation

• Building Workflows - Create automated workflows
• Workflow Templates - Pre-built workflow patterns
• Exception Reports - View workflow and processing issues