SISD

Free SISD practice questions

10 questions from our SISD practice tests, each with the correct answer and a written explanation. Read them with the answers shown, or switch to quiz mode and answer first.

Back to the SISD practice exams

Question 1 of 10, Discovery Pattern Design. Choose 2.

A team is building a Discovery pattern for a new in-house application on Windows and Linux. Which two practices follow ServiceNow guidance for custom patterns? (Choose two.)

  1. ACreate separate identification sections for different OS types or variants of the application Correct answer
  2. BWrite results directly to CMDB tables with script steps instead of using CI attribute variables
  3. CEdit an out-of-box pattern for a similar product and rename it to the new application name
  4. DSkip the identification rule for the new CI class because patterns already identify the CI
  5. ETest each section in debug mode against representative hosts before activating the pattern Correct answer

Answer and explanation: Patterns can contain multiple identification sections, which keeps OS-specific logic separate, and debug mode against real hosts validates each step before production. Renaming a modified out-of-box pattern creates an upgrade-unsafe customization; bypassing CI attribute variables bypasses the IRE; the new class still needs an identification rule for the IRE.

Question 2 of 10, Discovery Pattern Design. Choose 2.

A pattern runs successfully in debug mode but produces no CI when triggered by a scheduled Discovery. Which two causes should be checked first? (Choose two.)

  1. AThe pattern is not set as active or is not referenced by the classifier or trigger Correct answer
  2. BDebug mode results are always written to a temporary CI class
  3. CThe credential used in debug mode differs from what the scheduled MID Server can use Correct answer
  4. DThe pattern has more than one identification section for the same CI class
  5. EThe CI class has an identification rule, so the IRE ignores patterns

Answer and explanation: An inactive pattern, or one not linked from the classifier or process trigger, never runs on schedule, and differing credentials or MID Server reach between debug and scheduled runs commonly cause silent failures. Multiple identification sections are normal; identification rules are required, not blocking; debug runs do not write to a temporary class.

Question 3 of 10, Discovery Pattern Design

While building a new pattern, a developer wants to run each step against a real host and see the variable values returned at every step. What should they use?

  1. AThe Identification Simulation tool in CI Class Manager for the pattern's main class
  2. BDebug mode in Pattern Designer, connecting through a MID Server to the target host Correct answer
  3. CThe Discovery Status record of the last scheduled run, opened in the ECC Queue view
  4. DThe pattern's Preview tab, which renders the expected CI form using sample data

Answer and explanation: Debug mode executes the pattern step by step through a MID Server against a live host and shows the resulting variables and errors. A previous Discovery Status only shows logs of a past run; Identification Simulation tests IRE rules with sample payloads, not pattern steps; there is no preview tab that executes steps against sample data.

Question 4 of 10, Discovery Pattern Design

In Pattern Designer, what is the difference between a scalar variable and a tabular variable?

  1. AA scalar is visible to every pattern, while a tabular variable is local to one step
  2. BA scalar is written to the CMDB, while a tabular variable is never saved
  3. CA scalar holds one value, while a tabular variable holds rows and columns of values Correct answer
  4. DA scalar is always numeric, while a tabular variable is always text-based

Answer and explanation: Scalar variables store a single value, whereas tabular variables store multiple rows with named columns, such as a list of processes or ports. Scope, data type, and CMDB persistence are not what distinguishes them: both are local to the pattern run, both can hold text, and tabular CI variables are commonly written to related tables.

Question 5 of 10, Discovery Configuration

A Discovery schedule targets 10.20.0.0/16 and runs through a cluster of three MID Servers. Some Windows hosts must be explored only with WinRM, while UNIX hosts in the same range use SSH, and only two of the three MIDs have network access to the UNIX subnet. Which feature addresses this with the LEAST custom work?

  1. ASeparate Discovery schedules per OS that each point to a quick range and a hand-picked MID Server
  2. BMID Server capabilities set to Windows on one MID and UNIX on the others with auto-select enabled
  3. CA Discovery definition update that removes the SSH port probe from the Shazzam probe for the whole instance
  4. DDiscovery behaviors that assign the Windows and UNIX functionality and their protocols to specific MID Servers Correct answer

Answer and explanation: Behaviors let one schedule assign functionality definitions and protocols (for example WinRM for Windows, SSH for UNIX) to specific MID Servers, which is designed for multi-MID scans with different network reach. Separate schedules per OS duplicate configuration and cannot split one range by OS before classification; removing the SSH probe breaks UNIX scanning; capabilities help MID selection but cannot express protocol-per-functionality rules inside a single scan.

Question 6 of 10, Discovery Configuration

A company has one MID Server in its London data center and another in Singapore. Each should only scan the subnets local to it, and schedules should pick the right MID automatically. What should the administrator configure?

  1. APut both MID Servers in a single load-balancing cluster so that the instance spreads probes evenly across both sites
  2. BCreate Discovery behaviors that list only the Windows functionality for London and the UNIX functionality for Singapore
  3. CCreate a separate Discovery schedule for each site and hard-code that site's MID Server on the schedule record itself
  4. DSet the schedule to auto-select the MID Server and give each MID its local IP ranges and the Discovery application Correct answer

Answer and explanation: With auto-select, the instance chooses a MID Server whose assigned IP ranges cover the target and whose applications include Discovery, so each site's MID scans its own subnets. Hard-coding MIDs per schedule works but is not automatic and adds maintenance; a load-balancing cluster would send London probes through Singapore; behaviors define protocols per MID, not site ownership of subnets.

Question 7 of 10, Discovery Configuration

Discovery finds 40 Linux servers, but their serial numbers are empty and some attributes are missing, while hostname and OS are populated. The SSH credential is a non-root account. What is the MOST likely cause?

  1. AThe Linux servers are classified with the SNMP classifier instead of the UNIX classifier for the class
  2. BThe MID Server is missing the Linux CPU and memory Shazzam port probes for the UNIX device family
  3. CThe account cannot run privileged commands such as dmidecode, so sudo rights must be granted for them Correct answer
  4. DThe identification rule for Linux servers is set to match on name only and discards the serial number

Answer and explanation: Commands that read hardware data such as serial numbers (for example dmidecode) need elevated rights, so a non-root account without sudo for those commands returns partial data. SNMP classification would not populate OS details over SSH; Shazzam only probes ports and has no CPU/memory probes; identification rules decide matching, they do not stop attributes from being written.

Question 8 of 10, Discovery Configuration

A network device answers SNMP but Discovery classifies it as a generic 'Network Gear' class instead of a firewall. The vendor is supported, but this model's sysObjectID is new. Which change gives the MOST precise fix?

  1. AAdd an SNMP OID classification entry mapping that sysObjectID to the firewall class Correct answer
  2. BChange the SNMP classifier to classify every device of this vendor as a firewall
  3. CLower the SNMP credential's version to v1 so the device returns the vendor's legacy object IDs
  4. DAdd an identification rule on the firewall class that uses the device's name as the match key

Answer and explanation: SNMP classification uses the OID classification table to map a sysObjectID to a class, so a new entry fixes this model without affecting others. Downgrading to SNMP v1 does not change the sysObjectID and weakens security; identification rules match records but do not decide class; classifying the entire vendor as firewalls would misclassify switches and routers.

Question 9 of 10, Configuration Management Database

Discovery and a third-party tool both create network adapters for servers. Some servers now show duplicate adapters with the same MAC address. Where is the fix MOST likely needed?

  1. AIn the MID Server's IP ranges, so Discovery stops scanning the adapters' subnets
  2. BIn the identification rule for the adapter class, so both sources match the same adapter Correct answer
  3. CIn the CMDB Health dashboard thresholds, so that adapter duplicates are no longer counted against the class
  4. DIn the server class's data refresh rules, so older adapter values are removed

Answer and explanation: Dependent CIs such as network adapters are matched by their identification rule (for example MAC address plus the parent server); if the third-party payload does not satisfy it, duplicates appear. Health thresholds only hide the issue, changing ranges stops discovery, and refresh rules control attribute authority, not matching.

Question 10 of 10, Discovery Engagement Readiness

Production is cloned over the test instance. What should the team do right after the clone to avoid Discovery problems?

  1. AMake sure test MID Servers, schedules, and credentials cannot scan production Correct answer
  2. BDisable the IRE in test because cloned data already passed identification
  3. CPoint production MID Servers to the test instance to double the scan capacity
  4. DDelete the CMDB in test so that Discovery can rebuild it again from scratch

Answer and explanation: After a clone, schedules and credentials from production exist in test, so MID Server records and schedules must be checked so the test instance does not scan production unexpectedly. Deleting the CMDB is unnecessary, pointing production MIDs at test breaks production, and the IRE cannot be disabled.