Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Navigation  Back up to About Overview 

Known Issues

Table 2 lists known issues in NorthStar Controller Release 3.0.0. The identifier associated with each entry is the tracking number in the Juniper Networks Problem Report (PR) tracking system.

Table 2: Known Issues in NorthStar Controller 3.0.0

Identifier

Description

1267861

The show command output of a SPRING LSP on the PCC displays an incorrect node IP address; the previous hop address is displayed instead. This has no impact on the operation of the LSP.

1269162

Spring TE LSPs using node SIDs created by NorthStar are not turned down when the following command is launched on the LSP tail end: deactivate protocols isis source-packet-routing node-segment.

1278728

NorthStar Controller Release 3.0.0 does not report the correct RRO in the web UI and via the REST API when routers are configured with Junos OS Release 17.2R1. Instead of showing a list of link adjacency SIDs, the web UI and REST API report a list of “zero” labels. This issue has been fixed in Junos OS Releases 17.2.R1-S1 and 17.2R2.

1287508

During very rare cases, pce_server can linger during HA switchover and not stop completely which can be resolved manually by issuing killall -9 pce_server on the standby server. In NorthStar Controller 3.1.0, it will be part of HA agent failover sequences to ensure the pce_server completely stops during HA switchover.

1290455

We recommend the following workflow when creating or modifying a collection task for all routers in the network:

  1. In the Modify Task - Netconf window, click Selected devices rather than All devices.
  2. Select the check boxes for all devices individually in the Collect column.

This workflow will be improved in NS 3.1.0. If All devices is selected, the collection task for all routers does still work, but the status of the collection is not properly displayed.

1290466

During scale tests with tens of thousands of LSPs, it has been (inconsistently) observed that the NorthStar Controller does not always re-provision existing PCE-initiated LSPs when the PCEP session is not stable. PCC-controlled LSPs and PCE-delegated LSPs are not impacted by this issue.

A workaround is to increase the queue_high_watermark parameter in the northstar.cfg settings file. We recommend a queue_high_watermark value of two times the number of LSPs in the network.

1290503

In some circumstances, PCE-initiated LSPs do not properly switch to the secondary path as expected when the primary path is no longer valid, such as when an interface that belongs to the LSP return path is disabled on a transit router. This issue is not always reproducible. LSPs with a standby path are not impacted. We believe the issue only occurs for PCE-initiated LSPs.

Modified: 2017-07-24