Change collision records across systems: a key map from ServiceNow to on-call and CMDB

Short answer

Five keys rebuild a clash between two approved changes: the change number, which ties each change to its approval, its conflict result and later its rollback; the CI ID, which ties a change to the components it depends on; the service name, which ties those components to what monitoring watches; the alert ID, which ties monitoring to the incident; and the incident number, which ties the response to the correlated change and the review. ServiceNow holds the change and configuration side, monitoring and on-call tools hold the response side, and the review often sits in a wiki. Break any one key and the case becomes unrelated tickets.

Visual briefFull-size diagram
A ring of five systems, change management, CMDB, monitoring, on-call and review wiki, with keys CI ID, service, alert ID, incident number and change number passing around the ring until one collision case closes the loop.
Illustrative key map for a fictional managed services provider; real field names depend on each company's configuration.

This system map is part of the change collision recovery guide. It answers a single question for an ITSM implementation partner: across the systems a company uses, which identifier joins which record, who owns each system, and what does one fully joined case look like?

The system map: who holds what, and the key between them

SystemRecord it holdsLinking key outTypical owner
Change management (ServiceNow)Change request, approvals, planned window, conflict resultsChange number; affected CIChange manager
CMDB (ServiceNow)CIs and the relationships between themCI ID; relationship recordConfiguration manager
MonitoringAlert, threshold, start and clear timesAlert ID; service namePlatform or SRE lead
On-call (PagerDuty)Incident, responders, change events and correlationsIncident number; change event referenceService delivery director
Review wiki or problem recordPost-incident review, actions, configuration fixLinks to incident and change numbersProblem manager

Each owner sees one slice. The change manager sees approvals but not the alert. The on-call lead sees the incident timing but not the CAB notes. Nobody holds the case; it exists only where the keys line up.

LinkKeyWhat usually breaks it
Change to CIAffected CI field on the changeWork logged against a broad service, not a CI
Change to conflict resultChange numberResults shown on screen but never stored
CI to dependencyCMDB relationship recordRelationships added later, with no history
Change to on-callChange event created by an integrationIntegration never switched on
Alert to incidentAlert ID copied onto the incidentIncidents opened by hand without the alert
Incident to changeCorrelation shown on the incidentNo change events, or no service dependencies configured
Rollback to changeChange number on the back-out taskRollback done under the original record with no note
Review to allIncident and change numbers citedReviews that describe "the Tuesday outage" in prose

Two keys are created by tools rather than people: the change event and the correlation. The other six depend on habits, such as copying an alert ID or citing a ticket number, which is why their quality varies by team and by year.

Chronology: the join that needs no key

Keys say which records belong together. Timestamps say what came first, and this kind of failure is defined by timing: two windows overlapping, an alert inside both, a rollback before recovery.

Three things often damage chronology:

  • CMDB relationships without history. If a dependency was added after the incident, the current configuration data suggests it always existed. Only relationship history shows otherwise.
  • Migrated tickets. Consolidations can reset created dates to the import date, so a change looks approved after it ran.
  • Separate clocks. Monitoring, on-call and ITSM may log in different time zones. Record which.

The ServiceNow archiving and retention guide explains why closed records and their history may already sit in archive tables, which affects whether chronology survives.

Three documented behaviors explain why some links exist at all. ServiceNow's conflict detection needs a configuration item and planned start and end dates, and lists the conflicting change, the CI and the issue type; a check built on CIs cannot flag a dependency the CMDB does not hold. PagerDuty's ServiceNow integration can create change events when a change request opens, and those events never page anyone; they appear as context. On an incident, PagerDuty shows up to three changes correlated from the past 24 hours with a reason, and related-service matches depend on configured service dependencies. If dependencies were not configured, correlation fell back to time, which can be right by coincidence, so note which reason past incidents show.

One fully joined case (Illustrative)

Illustrative. A fictional 900-person managed services provider. On a Tuesday evening a firewall rule update (CHG-1) and a certificate rotation on the API gateway (CHG-2) each passed review, ran in overlapping windows and together cut off a partner API. All numbers are invented.

OrderSystemRecordKey that joins it
1Change managementCHG-1 approved, 21:00 to 23:00Change number; CI: edge firewall
2Change managementCHG-2 standard change from 21:30Change number; CI: API gateway
3Change managementConflict results for both: noneChange numbers
4On-callChange events for CHG-1 and CHG-2Created by the ITSM integration
5MonitoringAlert at 21:52: partner API handshake failuresAlert ID; service: partner API
6On-callIncident at 21:55; CHG-1 and CHG-2 both correlated by timeIncident number; alert ID
7Change managementEmergency rollback of CHG-1 approved at 22:10Change number of CHG-1
8On-callIncident resolved at 22:25Incident number
9Review wikiReview cites the incident and both changesIncident and change numbers
10CMDBPartner API now linked to the firewall and the gateway; relationship dated after the incidentCI IDs; relationship date

Every row holds because a key was present. Remove row 4 and row 6 loses its correlation. Remove the citations in row 9 and the review floats free. Row 6 also shows a limit: with two changes correlated by time, the record alone does not say which was the cause, so the rollback note in row 7 carries the reasoning. A different collision, told step by step with who decided what, is in the worked collision case.

A quick test of whether a client's joins are intact

Ask the service delivery director about one past clash:

  1. Can the team show the stored conflict result that said "no conflict"?
  2. Does the incident show the change that was later rolled back, and with what correlation reason?
  3. Was the rollback approved on its own record, citing the change number?
  4. Does the review cite both the incident and the change numbers?
  5. Can the configuration data show when the missing link was added?

Yes to most means the joins are probably intact. Note the answers as metadata, never as exports. For each system in more depth, see ServiceNow incident data, PagerDuty incident exports and ITSM ticket history. Carry the key facts into the referral brief template.

Qualification and the open questions

The company needs 50+ full-time employees at peak (contractors excluded), a documented operating history, rights to license the records, and an authorized contact. Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 70% of the data is in English.

Until SourceX reviews it, nobody knows whether buyers want this joined history, how customer contracts affect records about hosted services, or whether an offer follows.

Next step

Refer a company. No cap on rewards. Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.

If a client can join a past clash from conflict result to the configuration fix, register as a partner, check the company on who qualifies, and read the program terms before you introduce its service delivery director.

  1. Step 1Share your linkSend your personal link to a company you know.
  2. Step 2Company appliesThe company applies itself at /apply.
  3. Step 3Company signs and is paidThe company signs, delivers its data and is paid by SourceX under its agreement.
  4. Step 4You get your rewardthe greater of $50,000 plus 5% and 15% of the actual supplier payout, with no cap. A reward becomes payable after SourceX has paid the referred company under its signed agreement and the payout conditions in the program terms are met, including finance approval and your tax and payment details.

Common questions

Do we need both the ITSM and the on-call tool to reconstruct a collision?

Usually yes. The ITSM tool holds the approvals, conflict results and CMDB, while the on-call tool holds the alert timing and which change was correlated with the incident. Either one alone shows half the case.

What if change events were never sent to the on-call tool?

Then the correlation link is missing, and the incident-to-change join depends on someone having typed the change number into the incident notes. Note that as a gap; the rest of the chain may still hold.

Sources

  1. docs.servicenow.com/bundle/paris-it-service-management/page/prod…
  2. support.pagerduty.com/docs/change-events
  3. support.pagerduty.com/main/docs/servicenow-change-requests-integra…
  4. refer.sourcex.si/terms

Free resources

Know a company with valuable proprietary data?

Become a referral partner, get your link and introduce the company’s CEO (in an owner-run company, usually the owner). Referral partners can join from most countries. We can't work with partners in countries or regions under U.S. sanctions or on U.S. restricted-party lists. The companies you introduce should be based in the U.S., Canada and the EU, or in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia with at least 70% of their data in English.

Refer a company →

I own a business

Explore licensing your company's data directly to SourceX for AI training and evaluation. Start a short assessment; no uploads needed.

Start an assessment