Dynatrace Hub - Cisco Catalyst SD-WAN Manager (Viptela) latest releases

Cisco Catalyst SD-WAN Manager (Viptela) 2.2.3

Wed, 21 Oct 2026 16:19:33 +0000

Update the certifi dependency to address CVE-2026-111943

Cisco Catalyst SD-WAN Manager (Viptela) 2.2.2

Thu, 01 Oct 2026 10:01:36 +0000

New in this version

This extension is now using all rich visualization features of Grail within Infrastructure & Operations App.
Alarm reporting improvements. Cisco SD-WAN raises a new alarm every day for a lasting problem, such as an expired certificate, and for many alarm types it never clears one. The extension now groups alarms by the condition they describe, so one condition raises one problem instead of one per alarm record. The problem stays open while the condition keeps being reported, and closes when the alarm clears or when it stops being reported. Each closing log record says which of the two happened.

Note: Cisco SD-WAN alarms carry no correlation ID, so the extension works out which alarms belong to the same condition on a best-effort basis. If the grouping does not match what you expect, rely on the alarm log records: they hold the alarm payload exactly as Cisco SD-WAN Manager reported it.
Certificates. A new optional certificates feature set reports how many days are left before each device certificate expires. It covers controllers and edge routers.

| Metric | Description | | --- | --- | | cisco.sdwan.device.certificate.days_remaining | Days left until this device certificate expires |

It is split by fabric_infra, system_ip, host_name, site_id, site_name, device_model and device_type.

A negative value means the certificate has already expired. For example, -30 means it expired 30 days ago.

Three alert templates come with it: Certificate Expired, Expiry Imminent (7 days) and Expiring Soon (30 days). They do not overlap, so one certificate raises only one problem.

The new Certificate Collection Frequency setting controls how often certificates are queried: every 30 minutes by default, from 15 to 1440 minutes.
Site names and coordinates on Smartscape nodes. You can now group a Smartscape segment by a readable site name instead of a site ID. Devices and sites also carry map coordinates, so you can place them on a Dynatrace map with no extra work.

⚠ Breaking change. The coordinate fields were renamed from latitude and longitude to geo.location.latitude and geo.location.longitude. Update any query or dashboard that reads the old names. Classic entities and the site screen are not affected.
Device control plane. A new optional device_control_plane feature set turns ten control-plane values into metrics. Each total has a matching up counter — compare the two to find a session that is down. Until now these values were only static device properties. You can now chart them, alert on them and use them in SLOs.

| Metric | Description | | --- | --- | | cisco.sdwan.device.control_plane.control_connections | Total control connections this device has to vSmart, vBond and vManage | | cisco.sdwan.device.control_plane.control_connections_up | Control connections that are currently up | | cisco.sdwan.device.control_plane.vsmart_control_connections | Control connections to vSmart controllers | | cisco.sdwan.device.control_plane.expected_vsmart_connections | Control connections to vSmart controllers this device should have | | cisco.sdwan.device.control_plane.omp_peers | Total OMP peers of this device | | cisco.sdwan.device.control_plane.omp_peers_up | OMP peers that are currently up | | cisco.sdwan.device.control_plane.bfd_sessions | Total BFD sessions of this device | | cisco.sdwan.device.control_plane.bfd_sessions_up | BFD sessions that are currently up | | cisco.sdwan.device.control_plane.total_tloc_num | Total transport locators (TLOCs) of this device | | cisco.sdwan.device.control_plane.up_tloc_num | Transport locators that are currently up |

All ten are split by fabric_infra, system_ip, host_name, site_id, site_name, friendly_site_name, device_model and device_type.

New charts show them on the device and SD-WAN infrastructure screens.

These are totals for the whole device. The Circuit availability metrics count similar sessions per transport, so the two sets of numbers do not add up to each other. That is expected.

⚠ Existing monitoring configurations must enable Device control plane before the metrics start to flow. New configurations have it on by default.
SD-WAN events as log records. Cisco SD-WAN Manager reports events as single points in time. Each one is now a Dynatrace log record with its severity, component, name, device and site, shown in a new SD-WAN Events card on the SD-WAN infrastructure, site and device screens. Events never become Dynatrace problems.

⚠ Collection is off by default. Turn on Report SD-WAN Events as Logs in the monitoring configuration, and set the SD-WAN Event Severity Filter before you do. All severities are collected unless you narrow it, and Minor events such as logins and logouts can be very frequent. Cisco SD-WAN Manager applies the filter itself, so excluded events are never transferred.

The new SD-WAN Events Collection Frequency setting controls how often events are queried: every 15 minutes by default, from 5 to 360 minutes. It also sets how often the event count metric below is reported.
Event volume metric. A new optional events feature set, shown as SD-WAN events, counts the events Cisco SD-WAN Manager reports, so you can chart and alert on event volume.

| Metric | Description | | --- | --- | | cisco.sdwan.device.event_count.gauge | Number of SD-WAN events reported for this device, per severity and per 5-minute interval |

Note: In DQL, use the key cisco.sdwan.device.event_count, without .gauge. The FAQ Why do some metric keys end in .gauge or .count, and which key do I use in DQL? in the extension documentation explains why.

It is split by fabric_infra, system_ip, host_name, site_id, site_name, friendly_site_name, device_model, device.address and event_severity.

A new SD-WAN Events by Severity chart shows it on the SD-WAN infrastructure, site and device screens.

⚠ It is off by default and adds one API call per collection cycle. Existing monitoring configurations must enable SD-WAN events before the metric starts to flow.

⭐ Reporting SD-WAN events as log records is a separate thing and is not affected by this feature set. Log records are controlled only by the Report SD-WAN Events as Logs configuration property. You can have the log records without the metric, or the metric without the log records.
Davis event Smartscape enrichment. Every Cisco SD-WAN alarm Davis event is now enriched with the Smartscape source properties (dt.smartscape_source.id, dt.smartscape_source.type, dt.smartscape.ext_network_device) so that the event links to the correct EXT_NETWORK_DEVICE Smartscape node for RCA and entity-centric views. The node ID is computed deterministically with smartscape_id() from dt-ppx-utils.

Changed in this version

⚠ The alarm count metric now reports one data point per time interval, timestamped with the interval it covers. Charts now line up with the alarm histogram in Cisco SD-WAN Manager, and intervals missed during a short outage are filled in afterwards. Please re-check any dashboard or alert built on this metric — it produces fewer data points than before.

The interval is set by Alarm Count Resolution, 15 minutes by default. This setting was previously called Alarms Collection Frequency, which was misleading: it never changed how often alarms are collected. Renamed only, the behaviour is the same.

Alarm log records and alarm-driven Dynatrace problems are not affected.
The alarm filter field Rule Name is now called Alarm Type, the name Cisco SD-WAN Manager uses in its alarms table. Renamed only: existing filters keep working.
Alarms and SD-WAN events are now grouped into tabs on the SD-WAN infrastructure, site and device screens, instead of five cards stacked down the page.
The alarm distribution charts are now stacked columns over time.

Fixed in this version

⚠ TLOC octets are now reported as a delta metric. cisco.sdwan.tloc.rx_octets and cisco.sdwan.tloc.tx_octets count the bytes received and sent, so they are now reported as a delta instead of a gauge.
- Latest Dynatrace (Grail): no impact.
- Dynatrace classic (Data Explorer, classic dashboards and screens): after the upgrade, newly collected data points must be accessed using the keys cisco.sdwan.tloc.rx_octets.count and cisco.sdwan.tloc.tx_octets.count. Use these keys in Data Explorer and in your own charts; the extension's screens already use them. Data from before the upgrade is not lost: read it in Data Explorer with cisco.sdwan.tloc.rx_octets and cisco.sdwan.tloc.tx_octets.
An invalid Friendly Site Name Pattern no longer breaks collection. Patterns are checked when the extension starts and any problem is reported as a warning.
TLOC Smartscape node name fixed. It now reads host:colour:protocol — for example SDW_BR4:biz-internet:IPSEC. Existing nodes are renamed on the next entity report; no new TLOC nodes are created.
Lower memory use. The extension now needs less memory to collect statistics. This avoids the ActiveGate python memory limit. Nothing changes about which metrics are reported, or their values.

If you still see memory problems, lower the page size in the expert settings. A smaller page needs less memory. But it also sends more requests to Cisco SD-WAN Manager, so you may need a longer metrics collection frequency. The FAQ in the documentation explains how to choose a value.

Cisco Catalyst SD-WAN Manager (Viptela) 2.1.0

Wed, 19 Aug 2026 11:41:56 +0000

New in this version

"Cisco SD-WAN — ISP Circuit Down" alert template. Raises an availability event when both the BFD sessions and the control connections of a WAN transport drop to zero at the same time, indicating that the ISP circuit itself is no longer operational rather than an individual tunnel failing. Controller transports are excluded. The template requires the circuit_availability feature set to be enabled. Its sliding window is sized for the default 5-minute collection frequency — if you change metrics_collection_frequency_minutes, set the template's slidingWindow to 3 × frequency + 1.

Fixed in this version

Fixed an invalid TLOC loss percentage calculation. Tunnels whose remote end is unreachable report exactly 100% loss while carrying only BFD probe packets and no payload traffic. These tunnels were included in the average and dominated it, so the reported loss was far higher than the actual loss on the transport. They are now excluded from the aggregation. On fabrics that have unreachable remote sites, cisco.sdwan.tloc.loss_percentage will drop substantially after this upgrade.
TLOC statistics are now aggregated into 10-minute buckets instead of 5-minute ones, matching the native SD-WAN statistics cycle.
The TLOC statistics aggregation window is now shifted 15 minutes into the past, so that only fully populated statistics buckets are aggregated. Previously the most recent bucket was still being filled while it was read, which produced values calculated from an incomplete dataset.

Cisco Catalyst SD-WAN Manager (Viptela) 2.0.0

Wed, 10 Jun 2026 13:39:51 +0000

⚠️ WARNING ⚠️️: Possible breaking changes and warnings — PLEASE READ before upgrading

This version requires a minimum Dynatrace version of 1.338.0 and a minimum EEC version of 1.333.0.
This version includes a possible breaking change for Dynatrace on Grail.
- With the inclusion of OpenPipeline processing rules, any metrics ingested from the extension will be processed by a dedicated OpenPipeline pipeline instead of the default pipeline.
- This means that any custom dynamic rules you created will no longer apply to metrics from this extension. To keep enforcing them, please create pipeline groups.
- Additionally, some field names are handled differently in OpenPipeline (e.g. capitalization). Please double-check any custom dashboards and alerts you created for this extension.
- You must opt-in to OpenPipeline configuration via Settings API — see the documentation.

New in this version

Smartscape on Grail. This extension will populate new Smartscape nodes and edges via OpenPipeline extraction.
Primary tags and Primary fields (monitoring configuration): two new optional sections on the endpoint configuration make this release ready to support the upcoming global entity tagging functionality.

Cisco Catalyst SD-WAN Manager (Viptela) 1.1.0

Wed, 13 May 2026 12:50:33 +0000

New in this version

Added SD-WAN alarm observability:
- alarm count metric cisco.sdwan.device.alarm_count.gauge
- optional alarm reporting as logs with severity-based filtering
- optional creation of Dynatrace events with configurable rules, including per-condition overrides and suppression
- Infrastructure, Site, and Device screens now include alarm coverage: Alarm Severity, Alarm Component, Alarm Rule Name, and Active/Inactive status
Added documentation note for Cisco SD-WAN alarm API RBAC:
- to enable SD-WAN alarms, update the custom monitoring role to include Alarms: Read
- required custom-role read permissions are documented in the Cisco SD-WAN extension documentation

Improved in this version

Added periodic updates of Site, Device, and Interface properties, reported as log events with configurable frequency:
- Site: Devices Health, Tunnels Health, Apps Health, Latitude, Longitude
- Device: Software Version, Control Connections Up, Control Connections, vSmart Control Connections, Expected vSmart Connections, Up Since, BFD Sessions Up, BFD Sessions, OMP Peers Up, OMP Peers, Board Serial Number, Chassis Number
- Interface: BW Up, BW Down
Updated alarm activation configuration:
- Report Alarm Logs: alarm log output with severity filtering
- Report Alarm Events: Dynatrace event creation per alarm
- Alarm Reporting Rules: per-rule event configuration and suppression
- Entities Reporting Frequency: entity log update frequency (default 60 minutes)
Added scheduling overlap detection: if a monitor run is still active when the next one is due, the new run is skipped with a warning log and DEC code. If this occurs frequently, increase the reporting frequency.
Added maximum limit of 30 seconds for the Single Request Timeout connection setting.
⚠️ Important: If the timeout was previously configured above 30 seconds, update it to a value within the new limit (1–30 seconds) before upgrading.

Cisco Catalyst SD-WAN Manager (Viptela) 1.0.3

Fri, 06 Mar 2026 09:55:21 +0000

Improved in this version

Fixed cisco.sdwan.site.devices_health metric reporting wrong value (was using site health instead of devices health).
Fixed TLOC metrics reporting for devices with multiple protocols.
Added reporting for the dt.ip_addresses property on devices.

Cisco Catalyst SD-WAN Manager (Viptela) 1.0.0

Fri, 06 Feb 2026 09:44:06 +0000

This is the first release of the Cisco SD-WAN extension. Looking forward to your feedback on the Dynatrace Community.