Skip to content

Jira Cloud

Since Syncer 3.8.2 the Jira Cloud plugin imports Assets/Insight objects into the Syncer; since 4.1 it can also export host fields back into Jira Cloud Assets.

Account Settings

Configure on the Account record:

  • Address: your https://<workspace>.atlassian.net URL
  • Username: Jira account username (email)
  • Password: API token (not your password)

In Additional Settings:

  • workspace_id — Atlassian Assets workspace id. Discover it via GET <address>/rest/servicedeskapi/assets/workspace.
  • ql_query — AQL filter used by import only, e.g. objectType = "Hardware Server".
  • verify_certTrue / False.

Import

cmdbsyncer jira import_cloud <account>

Walks the configured ql_query and stores each Asset object as a Host record with one label per Asset attribute.

Export

The export writes selected host attributes back into Jira Cloud Assets objects. One Export Rule = one target object type. Multiple rules can run in a single export pass, e.g. push Linux hosts to Red Hat Linux and Windows hosts to Windows Server without re-iterating the host list.

One-off: cache the Jira schema

The GUI needs to know which object types and attributes exist to offer autocomplete; the export plugin uses the same cache to resolve attribute names to ids without doing a live Jira call per host.

cmdbsyncer jira sync_schema <account>

Re-run after schema edits in Jira. The cache is visible under Modules → Jira Cloud → Schema Cache.

Define an Export Rule

Modules → Jira Cloud → Export Rules → Create:

  • Conditions: standard rule conditions — pick which hosts the rule applies to.
  • Field Mapping: one row per Jira attribute to write. Pick the Target (a single dropdown over every cached object-type / attribute pair, e.g. CMDB / Hardware Server / Zweck), then enter the Value as a Jinja template rendered against the host's labels and inventory; {{HOSTNAME}} is available. Multiple rows targeting different object types in the same rule write to one Jira object per type per host.

Missing Jira objects are created on the fly — the hostname is filled into the Name attribute. If you want certain hosts to be left alone entirely, add a Filter rule (see below).

Filter (exclude hosts)

Modules → Jira Cloud → Filter uses the standard Filter rule format: match conditions on the host, then an Ignore Matching Hosts action drops the host from the export entirely (no update, no create). This is the canonical place to scope the export — narrower than Account → Plugin Settings → object_filter (which only filters by object type).

Lookup is always by hostname → Jira's Name attribute. If you need to match by something else, use a label / inventory rewrite to put the desired value into the hostname (or open an issue if this stops covering your case).

Run

cmdbsyncer jira export_cloud <account>

For each enabled rule the plugin:

  1. Loads every existing object of the target type in one paginated AQL call.
  2. Iterates matching hosts and renders the field mapping.
  3. Compares with the current values and PUTs only when something changed (unchanged objects are not re-sent).
  4. POSTs a new object when the lookup yields nothing and Create if missing is on.

The run logs updated=, created=, unchanged= and skipped= per rule so an idempotent re-run produces visible "no changes" output instead of silent traffic.

Cron

Both jira sync_schema and jira export_cloud are registered as cronjobs and can be scheduled under Cron → Cronjobs.