Advanced
|
Sync Network Model
|
Click Sync to refresh the synchronization of the network
model. You can use this option if the network model audit has unresolved
discrepancies or if the information displayed for the model is out
of sync.
When you sync the network model, this is what happens
behind the scenes:
Information associated with the network model (nodes,
links, LSPs, interfaces, SRLGs, and user-defined parameters) remains
intact. Nothing is purged from the database.
Note: Device profiles are not affected.
The network model is repopulated with live data learned
from topology acquisition.
Table 2 describes the effects
on various elements in the network when you reset or synchronize the
model.
|
Reset Network Model
|
Warning: The Reset Network Model operation is typically appropriate for a lab rather than a production
environment. We strongly recommend that you perform this
action only if the Juniper Networks Technical Assistance
Center (JTAC) suggests that you do so.
There are two circumstances under which you must reset
the network model in order to keep the model in sync with the actual
network:
The node ISO network entity title (NET) address changes—This
can happen when configuration changes are made to support IS-IS.
The routing device’s IP address (router ID) changes—The
router ID is used by BGP and OSPF to identify the routing device from
which a packet originated. The router ID is usually the IP address
of the local routing device. If a router ID has not been configured
for the device, the IP address of the first interface (of the device)
that comes online is used, which is usually the loopback interface.
Otherwise, the first hardware interface with an IP address is used.
If either of these addresses changes, and you do not perform
the Reset Network Model operation, the network model in the Pathfinder
database becomes out of sync with the live network.
When you reset the network model, this is what happens
behind the scenes:
Information associated with the network model (nodes,
links, LSPs, interfaces, SRLGs, and user-defined parameters) is purged
from the database (so, we recommend that you do not reset the network
model unless it is absolutely required, and that you perform this
action only under JTAC supervision).
Note: Device profiles are not affected.
The network model is repopulated with live data learned
from topology acquisition.
Table 2 describes the effects
on various elements in the network when you reset or synchronize the
model.
|
Config Server
|
Settings
|
publish-aslink—If you enable this toggle button, the ConfigServer publishes the AS link (created
by the getipconf command) to
Pathfinder. By default, this toggle button is
disabled.
- update-topology—If you enable this toggle button, the config
server adds new nodes and links (not discovered through
BGP-LS) in the topology. If you disable (default) this
toggle button, new nodes and links aren’t added in the
topology. This parameter hasn't been tested in all
environments and hence, configuring this parameter may not
provide the desired results.
include-interface-type—From the list, select one
or more interface types that you want the ConfigServer to add to Pathfinder:
physical (physical interfaces)
loopback-mgmt (loopback and management interfaces)
vrf-if (interfaces associated with a VRF)
links-if (interfaces on links)
all (all interfaces)
Note: ConfigServer publishes to all components only the interface
types that you specify. The GUI and data collection receive information
about the interfaces representing those interface types only. Later,
if you modify this setting by clearing the previously selected interface
types (that are already represented by interfaces in the Pathfinder
model), the information collected for those interfaces remains in
the model.
|
Log Destination
|
name—Displays the name that is used to identify the log destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error
condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Device Profile
Adapter
|
Settings
|
redis-poll-interval—Specify the frequency (in seconds)
with which the device profile adapter polls the Redis service for
the Redis status.
Enter 0 or a negative value to disable this setting.
Example: 10s or 10seconds
Default: 5s
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Mladapter
|
Settings
|
polling-interval—Specify the polling interval (in
seconds) for interfaces without notification support. To disable polling,
enter 0.
Default: 3600
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Netconf Connection
Controller
|
Device-Connection Pool
|
Device-Connection Pool:
batch-size—Specify the maximum number of devices (per cycle) for which a NETCONF session must be
established with the NETCONF connection
controller.
Default: 10
reconnect-delay—Specify the delay (in seconds) after which the NETCONF connection controller
attempts to reconnect with the device, when the
NETCONF session fails.
Default: 30s
disconnect-delay—Specify the delay (in seconds) after which an attempt is made to disconnect the
NETCONF session between the NETCONF connection
controller and the device, when the session is
migrated.
Example: 250s or 250seconds
Default: 300s
Netconf:
keepalive-interval—Specify the interval (in seconds) at which the NETCONF connection controller
periodically sends messages to the devices to
ensure that the NETCONF session isn’t disconnected
due to inactivity.
Default: 0s
port—Specify a TCP port number to change the default
port for NETCONF from 830. In some installations, port 22 is preferred.
Default: 830
|
In-memory-Datastore
|
connection-pool-size—Specify the maximum number of connections to be maintained, between the
microservice and in-memory datastore, in the connection
pool.
Default: 5
reconnect-delay— Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the in-memory datastore, when the
connection between the two fails.
Default: 1000
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the in-memory datastore, when the connection
between the two fails.
Default: 1000
|
Infra
|
Cache Memory:
Msg Broker:
max-queue-length—Specify the maximum allowed length
of the queue used by the message broker.
Default: 0
If you use the default value, the queue is set without any maximum
length.
prefetch-count—Specify the number of messages to
be sent at the same time by the message broker.
Default: 200
|
Network Data Backend
|
backend-connection—From the list, select either folder orin-memory datastore as the backend
datastore from which you want to fetch the topology
information:
in-memory-datastore—If you select redis, the values
that you configured in netconf-connection-controller > in-memory-datastore are applied, and those values are displayed here.
folder-path—If you select file-system, specify the
path to the network definition folder.
Default: /opt/northstar/data/network_data/
|
Registry
|
check-delay—Specify the delay interval (in seconds)
between checking registered NETCONF connection controller instances
in Redis.
Default: 45s
publish-delay—Specify the delay interval (in seconds)
between attempting registration of NETCONF connection controller instances
in Redis.
Default: 30s
|
Rpc Service
|
workers—Specify the number of threads to be used by the remote procedure call (RPC) service
provider.
Default: 10
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Netconfd
|
Device Connection Pool
|
publish-connected-device-status—Specify the interval (in seconds) for publishing the connection
status of all the devices managed by netconfd.
Default: 0s
If you use the default value, the device connection status is
not published.
|
In-memory Datastore
|
connection-pool-size—Specify the maximum number of connections to be maintained, between the
microservice and in-memory datastore, in the connection
pool.
Default: 5
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the in-memory datastore, when the
connection between the two fails.
Default: 1000
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the in-memory datastore, when the connection
between the two fails.
Default: 1000
|
Infra
|
Cache Memory:
|
Network Data Backend
|
backend-connection—Select the backend datastore (either folder or in-memory datastore) from which
you want to fetch the topology information:
in-memory-datastore—If you select redis, the values that you configured in netconfd >
in-memory-datastore are applied, and those
values are displayed in this field.
folder-path—If you select file-system, specify the
path to the network definition folder.
Default: /opt/northstar/data/network_data/
|
Proxy
|
workers—Specify the number of threads to be used by the proxy service provider.
Default: 10
|
Registry
|
check-delay—Specify the delay (in seconds) between checking registered netconfd instances in
Redis.
Default: 45s
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Path Computation
Server
|
Settings
|
ignore-rro-check—Click the toggle button to enable
or disable the Path Computation Server (PCS) to ignore explicit route
objects (ERO) and record route objects (RRO) during LSP initial node
consolidation. By default, this toggle button is disabled.
northstar-vpn—Click the toggle button to enable
or disable Pathfinder VPN features in the PCS. By default, this toggle
button is disabled.
lsp-latency-interval—Specify the frequency (in seconds)
at which the PCViewer calculates LSP delay and display the data in
the Tunnel tab of the GUI (Network > Topology > Tunnels tab >
View > Delay).
license-check-interval—Specify the interval for
the PCS to check for the license (npatpw) file.
Default: 3600s
route-over-logical-SRLink—Click the toggle button
to enable or disable the PCS to calculate a route over a logical segment
routed (SR) link. By default, this toggle button is disabled.
SRLG-provisioning—Click the toggle button to enable
or disable the provisioning of shared risk link groups (SRLG). By
default, this toggle button is disabled.
provisioning-include-lsp-metric—Click the toggle
button to enable or disable the PCS to include the LSP metric when
provisioning the LSP. By default, this toggle button is disabled.
- round-trip-delay-on-remote-end—Click the toggle button to
enable or disable (default) configuring the link latency at
the remote-end router as the same value as the link latency
at the local-end router. Link latency is equal to half the
value of the measured delay, which is calculated as the RTT
in one direction.
disable-ecmp-tree-calculation—Click the toggle button to enable or disable the calculation of
equal-cost multi-paths (ECMP). If you enable this
toggle button, equal-cost multi-paths (ECMP) are not
calculated. If you disable (default) this toggle button,
ECMPs are calculated; if paths to a remote destination
have the same cost, then traffic is distributed between
them in equal proportion.
disable-node-sid-calculation—Click the toggle button to enable or disable node segment ID (SID)
calculation. If you enable this toggle button, SIDs for
the nodes are not calculated. If you disable (default)
this toggle button, segment IDs (SID) for the nodes are
calculated.
lsp-provision-queue-size—Specify the size for the
LSP provisioning queue.
Default: 50
pcep-speaker-id—Specify the PCEP speaker ID to be
used for P2MP flow mapping.
Default: northstar
zero-bandwidth-signalling—Click the toggle button to enable or disable the zero-bandwidth
signaling feature. By default, this toggle button is
disabled.
ecmp-placement-method—Specify one of the following
options as the ECMP placement method. After considering link metrics,
if there are multiple ECMPs available, Pathfinder selects a path based
on the value you specify here. If you don't specify a value, the first
random path is selected for all the LSPs.
Random—Pathfinder randomly selects one of the ECMP paths.
Least Fill—Pathfinder selects the path with the maximum
available bandwidth.
Most Fill—Pathfinder selects the path with the minimum
available bandwidth.
monitor-mode—Click the toggle button to enable or
disable the monitor mode in the PCS. By default, this toggle button
is disabled.
If you enable monitor mode, the PCS doesn't send LSP provisioning
orders to the PCC. You use this option when you want to monitor the
topology but don’t want the LSPs provisioned. If you disable
monitor mode, the PCS sends LSP provisioning orders to the PCC.
|
| Analytics |
- link-utilization-threshold—Specify the threshold value for
link utilization. When traffic on a link exceeds this value,
the controller triggers rerouting for label-switched paths
(LSPs). If a threshold isn’t specified, the LSP is not
rerouted. If 0 is specified, links are blocked.
Range: 0
through 100.
- packet-loss-threshold—Specify the threshold value for
packet loss on all links. If the packet loss on a
link exceeds this value, the link is considered
unstable, and the PCS triggers a maintenance event on
the link. If you don’t specify a value or specify 0, the
PCS doesn’t act in case of packet loss.
Range: 0
through 100.
- reroute-minimum-interval—Specify the minimum interval (in
minutes or m) after which the controller reacts to any
traffic or delay violations. If you don’t specify an
interval, the LSP is not rerouted in case of a violation.
Range: 1m through 300m.
|
BGP Steering
|
target-tag—Specify the target prefix for the programmable
routing protocol process (PRPD) filtering community.
Default: 42
target-tag-cookie-range-start—Set the starting number for allocating path cookies. EPE static
routes added by Pathfinder are allocated an unused path
cookie starting from the value specified here.
Default: 42
steering-route-preference—Specify the relative preference value for the PRPD steering route. If
you specify a sign (+ or -) for the value, the
preference is considered the relative increase or
decrease of the preference for static routes (relative
to BGP routes). If you don’t specify a sign, the value
is considered an absolute value.
Range: -127 through 127
Default: 1
steering-aggregate-colors—Click the toggle button
to enable the creation of a steering entry (per ingress prefix) that
contains the route target with all the colors. By default, this toggle
button is disabled.
Note: This parameter hasn't been tested in all environments and hence, configuring this parameter may
not provide the desired results.
|
In Memory Datastore
|
disable-pipeline—If you enable this toggle button, the in-memory datastore pipeline feature is
disabled (meaning, multiple entries are written to the
in-memory datastore one by one). If you disable this
toggle button, multiple entries are written to the
in-memory datastore in a single operation.
pipeline-threshold—Specify the buffer threshold value for the in-memory datastore pipeline.
Default: 65536
|
Link Flap
|
flap-interval—Specify the interval (in seconds)
after which the link flap count is reset. If a link remains in the
same status (Up or Down) longer than this interval, the counter is
reset and the link is no longer considered flapped.
Range: 1 through 300s
flap-count—Specify the maximum value for the link
flap count. When a link goes from Up to Down, the Path Computation
Element (PCE) increments the counter on that link. When the counter
reaches the maximum link flap count, the link is considered flapped.
Flapped links carry a large penalty, so are not preferred by the PCS.
|
LSP To Path Computation Instance
|
LSP Request Discriminator SR NodeSID:
Instance-type—From the list, select the instance
of the PCS that manages the segment routing (SR) LSPs tagged with Use node SIDs:
LSP Request Discriminator SR Test:
From the list, select the instance of the PCS that manages
the SR LSPs:
|
Segment Routing Policy
|
segment-routing-policy:
originator-asn—Specify the originator Autonomous System Number (ASN) from which the segment
routing policy originates.
Default: 0
originator-ip—Specify the IP address from where the segment routing policy originates.
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Path Optimization
|
Timer
|
Specify the frequency (in minutes) with which path optimization
is triggered automatically.
Note: The optimization is based on the current network, and
not on the most recent Path Analysis report.
|
Programmable
RPD Client
|
Settings
|
retry-interval—Specify the frequency (in seconds) with which the programmable routing protocol
process (PRPD) client retries to connect with the PRPD
daemon (on the router) after the connection between the
two fails.
Example: 60s or 60seconds
Default: 30s
top-prefix-filter—Specify how the top prefix filter is applied:
- Onwards—Only the routes learned in the future will
be checked against the filter and published if the
routes meet the filter criteria; existing routes are
retained.
- Immediately—Existing routes are cleared and all
routes (existing and those learned in the future)
are checked against the filter and published only if
the routes meet the filter criteria.
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
System
|
In Memory Datastore
|
connection-pool-size—Specify the maximum number of connections to be maintained, between the
microservice and in-memory datastore, in the connection
pool.
Default: 5
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the in-memory datastore, when the
connection between the two fails.
Default: 1000
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the in-memory datastore, when the connection
between the two fails.
Default: 1000
|
Messaging Bus
|
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the messaging bus, when the connection between the
two fails; 0 indicates no attempts and -1 indicates
infinite retries.
Default: -1
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the messaging bus, when the connection
between the two fails.
Default: 1000
max-channels—Specify the maximum number of channels that can be multiplexed over a single
connection between the messaging bus and the
microservice.
Default: 128
|
Persistent Datastore
|
|
Scheduler
|
device-profile-update-interval—Specify the interval
(in milliseconds) after which the internal timer in the scheduler
polls for device profile updates.
Default: 5000
|
Scheduler > Distributed Task Queue
|
thread-pool-size—Specify the number of threads that can be run for the Device Collection
task.
Range: 1 through 255
Default: 10
collector-max-pool-size—Specify the maximum size (in bytes) of queues to be allowed in the
distributed task collector messaging bus.
Default: 524288000
|
Scheduler >
Tasks
|
Collection Cleanup
|
rollup-data-retention-duration—Specify the retention period (in days) for aggregated data. Enter
0d to disable the retention of aggregated data.
Example: 800d or 800days
Default: 180d
raw-data-retention-duration—Specify the retention period (in days) for raw data logs. Enter 0d to
disable the retention of raw data logs.
Example: 10days or 10d
Default: 14d
interval—Specify how often (in days) the collection
cleanup task is executed. Enter 0 to disable the cleanup task.
Example: 7d or 7days
Default: 1d
|
Demand Reports
|
demand-reports:
as-demand-bucket-size—Specify the maximum number of autonomous system (AS) demand report entries
that can be stored in the bucket. This parameter isn't required
currently but may be required in future releases.
Default: 100
|
Device Collection
|
data-path—Specify the path to the directory for storing the data collected by the Device
Collection task.
Default: /opt/northstar/data
- enable-ptalk-logging—If you enable this toggle button, a
ptalk log file is generated for each device that is part of
the device collection task, when the task is run. You can
view the log files from the /opt/pcs/log/ directory
in the ptalkserver container and use the logs for debugging.
If you disable (default) this toggle button, the ptalk log
file isn’t generated.
timeout—Specify the time (in seconds) after which
the device collection task times out.
Example: 3000s or 3000seconds
Default: 1800s
enable-update-device-profile—If you enable this toggle button, the device-type attribute
is updated in device profiles as part of the Device
Collection task. If you disable (default) this toggle
button, the device-type attribute is not
updated.
- ptalk-timeout—Specify the time (in seconds) until which the
ptalk server waits for a response (to the ptalk request)
from each device. If the server doesn’t receive a response
from a device within the specified time, the ptalk request
for that device times out. You must then increase the
timeout period and re-run the device collection task for
those devices. Default: 600s.
purge-cutoff—Specify the purge cut-off period (in days). The device collection task deletes all
files that are older than the cut-off period specified
here.
Example: 10d or 10days
Default: 7d
|
File Transfer
|
interval—Specify the interval (in seconds) at which
the system transfers files from the active node to other cluster nodes.
Example: 3000s or 3000seconds
Default: 3600s
paths—Specify the absolute directory path to one or more directories in other cluster nodes for
transferring files from the active node.
|
Rollup
|
interval—Specify how often the ESRollup system task
is run (in hours). The ESRollup system task executes the esrollup.py
script to aggregate the previous hour’s data. The ESRollup task
is called from the Pathfinder server. You can view (but not modify)
the rollup task on the Task Scheduler page (Administration >
Task Scheduler).
Example: 3h or 3hours
Default: 1h
Note: We recommend that you do not change this default value
except to disable aggregation. If you want to disable data aggregation,
set the value to 0h.
max-worker-process—Specify the maximum number of
worker processes that can be used for the ESRollup system task.
Default: 4 or equivalent to the number of CPUs (whichever value
is smaller)
bulk-insert-record-count—Specify the maximum number of records in bulk, which can be inserted to
the PostgreSQL database in a single operation.
Default: 5000
|
Scheduled LSP Sizing
|
|
SNMP Collection
|
timeout—Specify the time (in seconds) after which
the SNMP polling task times out.
Example: 10s or 10seconds
Default: 3s
retries—Specify the maximum number of retries to
be allowed for an SNMP poll.
Default: 3
|
Log Destination
|
name—Specify a unique name for the log destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Topology Filter
|
Data Persistence
|
persistent-storage-init-wait—Specify the wait time (in seconds) for a connection to be
established between the topology filter and persistent
storage before exiting with an initialization failure.
Default: 45s
file-store-path—Specify the default file path for storing output data of the topology filter if
persistent storage isn’t available.
Default: /opt/northstar/data
|
Messaging Bus
|
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the messaging bus, when the connection between the
two fails. 0 indicates no attempts and -1 (default)
indicates infinite retries.
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the messaging bus, when the connection
between the two fails.
Default: 1000
max-channels—Specify the maximum number of channels that can be multiplexed over a single
connection between the messaging bus and the
microservice.
Default: 128
|
Persistent Datastore
|
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
Topology Server
|
Application
|
use-live-rsvp-bw-over-configured—Click the toggle button to enable the usage of live RSVP
bandwidth (obtained from the live network) instead of
the RSVP bandwidth that you configured.
By default, this toggle button is disabled.
use-nokia-path-workaround—Click the toggle button to append a double colon (::) between the
device name and routing path name strings for Nokia
devices.
By default, this toggle button is disabled.
use-unnumbered-interface-workaround—Click the toggle button to enable the usage of interface
index (IfIndex) to correlate link events for unnumbered
interfaces.
If you enable this toggle button, IfIndex is not used for correlating link events.
By default, this toggle button is disabled.
use-prefix-link-matching—Click the toggle button to enable the usage of prefixes to correlate
bidirectional links events.
By default, this toggle button is disabled.
do-not-publish-rest-updates—If you enable this toggle button, the topology server doesn’t publish
object updates to the REST server. If you disable
(default) this toggle button, the topology server will
publish object updates to the REST server.
do-not-suppress-beacon-message—If you enable this toggle button, the beacon messages will be
written to log files. If you disable (default) this
toggle button, the beacon messages will be
suppressed.
no-netconf-pathname-allocation—If you enable this toggle button, the topology server doesn't
check for path names allocated to LSPs configured
through NETCONF and also doesn't set the path name for
the planned LSP. If you disable (default) this toggle
button, the topology server will check the path names
and set the path name for the planned LSP.
|
Application Tuning
|
pce-beacon-holddown—Specify the time (in seconds)
within which the topology server must receive a beacon from the PCE
server. If the PCE beacon is not received within the specified time,
the connection between the topology server and the PCEP server is
marked as down.
Example: 80s or 80 seconds
Default: 90s
resync-unresolved-node-threshold—Specify a threshold
value for unresolved objects.
During the discovery and updation of the topology, the topology
server maintains a list of network objects that cannot be resolved
to nodes. If the number of unresolved objects exceeds the threshold
value specified here, a resynchronization is automatically triggered
with the BGP-LS topology source.
Default: 10
lsp-topo-sync-timeout—Specify the frequency (in seconds) with which the topology server
synchronizes the PCEP LSP database globally.
Default: 120s
pce-restart-holddown—Specify the maximum time (in
seconds) until which the topology server waits, after the PCEServer
starts, to request the PCEP LSP database.
Default: 30s
message-queue-low-watermark—Specify the low watermark
for the messaging queue. If the number of elements in the messaging
queue reaches the value specified here, the topology server resumes
LSP provisioning.
Default: 5000
message-queue-high-watermark—Specify the high watermark
for the messaging queue. If the number of elements in the messaging
queue reaches the value specified here, the topology server pauses
LSP provisioning until the number of elements in the messaging queue
reaches the low watermark.
Default: 500000
message-queue-overflow-watermark—Specify the overflow
watermark for the messaging queue. If the number of elements in the
messaging queue reaches the value specified here, the topology server
purges the messaging queue.
Default: 0 (meaning no overflow watermark is set and thus, purging
is disabled.)
pce-watermark-interval—Specify the frequency (in
seconds) with which the topology server checks the high and low watermark
rates for PCEP messages.
Default: 10s
pce-high-watermark-timeout—Specify the maximum time
(in milliseconds) that the topology server waits for the PCEP message
rate to reach the low watermark, after which, it resumes the processing
of PCEP messages.
Default: 60000
pce-high-watermark-rate—Specify the PCEP message
rate to trigger the high watermark. If the PCEP message rate reaches
the value specified here, the topology server pauses the processing
of PCEP messages.
Default: -1.0 (to disable the triggering of low watermark)
pce-high-watermark-count-threshold—Specify the number
of times the PCEP message rate must reach the high watermark, after
which, it exceeds the high watermark threshold. If the count exceeds
the threshold specified here, the counters for the received messages
and for watermark are reset, and the pce-high-watermark-timeout is
updated.
Default: 3
pce-low-watermark-rate—Specify the PCEP message
rate to trigger the low watermark. If low watermark is triggered,
the topology server resumes the processing of PCEP messages.
Default: -1.0 (to disable the triggering of low watermark)
pce-low-watermark-count-threshold—Specify the number
of times the PCEP message rate must reach the low watermark, after
which, it exceeds the low watermark threshold. If the count exceeds
the threshold specified here, the counters for the received messages
and for watermark are reset.
Default: 3
|
Data Persistence
|
persistent-storage-init-wait—Specify the maximum time (in seconds) until which the topology
server can wait for a connection to be established with
the persistent datastore. If the wait time exceeds the
specified value, the topology server exits the topology
server application with an initialization failure.
Example: 10s or 10seconds
Default: 45s
network-snapshot-store-path—Specify the output directory in which you want to save a snapshot of
the network topology.
Default: /opt/northstar/data/network_archive/NorthStar
file-store-path—Specify the default directory for storing event data files.
Default: /opt/northstar/data
debug-file-store-path—Specify the default directory for storing the debug output files.
Default: /opt/northstar/data
network-snapshot-store-interval—Specify the maximum time (in seconds) until which the network
topology snapshot can be stored in the persistent
datastore.
Example: 3000s or 3000seconds
Default: 3600s
persist-lsp-topology-object—Click the toggle button to enable or disable the persistence of the
LSP topology object in the persistent datastore.
By default, this toggle button is disabled.
do-not-persist-pcep-lsp-events—If you enable this toggle button, PCEP LSP events are not stored
in the persistent datastore. If you disable (default)
this toggle button, PCEP LSP events are stored in the
persistent datastore.
persist-pcep-pcc-node-events—Click the toggle button to enable or disable the persistence of PCEP
node events in the persistent datastore.
By default, this toggle button is disabled.
do-not-persist-topology-advertisements—If you enable this toggle button, BGP-LS advertisements
are not stored in the persistent datastore. If you
disable (default) this toggle button, BGP-LS
advertisements are stored in the persistent datastore.
do-not-persist-provision-requests—If you enable this toggle button, provisioning requests are not
stored in the persistent datastore. If you disable
(default) this toggle button, provisioning requests are
stored (in JSON format) in the persistent datastore.
do-not-persist-lsp-events—If you enable this toggle button, LSP events (event description and
object data) events are not stored in the persistent
datastore. If you disable (default) this toggle button,
LSP events are stored in the persistent datastore.
do-not-persist-node-events—If you enable this toggle button, node events (event description and
object data) are not stored in the persistent datastore.
If you disable (default) this toggle button, node events
are stored in the persistent datastore.
do-not-persist-link-events—If you enable this toggle button, link events (event description and
object data) are not stored in the persistent datastore.
If you disable (default) this toggle button, link events
are stored in the persistent datastore.
do-not-persist-lsps—If you enable this toggle button, LSP data is not stored in the persistent
datastore.
If you disable (recommended) this toggle button, LSP data
is stored in the persistent datastore.
Note: We recommend that you do not disable the persistence of
LSP data (that is, we recommend that you do not enable the toggle
button).
persist-topology-snapshot—Click the toggle button to enable or disable the persistence of the
topology snapshot in the persistent datastore.
By default, this toggle button is disabled.
persist-demand-events—Toggle the button to enable or disable (default) the persistence of demand
events in the persistent datastore.
Note:
For demand events to be persistent, LSP events should
be persistent.
persist-topology-objects-to-file—Click the toggle
button to enable or disable saving the BGP Monitoring Protocol (BMP)
events to a file in binary format.
By default, this toggle button is disabled.
|
In-memory Datastore
|
in-memory-datastore:
connection-pool-size—Specify the maximum number of connections to be maintained, between the
microservice and in-memory datastore, in the connection
pool.
Default: 5
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the in-memory datastore, when the
connection between the two fails.
Default: 1000
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the in-memory datastore, when the connection
between the two fails.
Default: 1000
|
Messaging Bus
|
reconnect-retries—Specify the maximum number of times a microservice can attempt to reconnect
with the messaging bus, when the connection between the
two fails. 0 indicates no attempts and -1 (default)
indicates infinite retries.
reconnect-delay—Specify the delay (in milliseconds) after which the microservice attempts to
reconnect with the messaging bus, when the connection
between the two fails.
Default: 1000
max-channels—Specify the maximum number of channels that can be multiplexed over a single
connection between the messaging bus and the
microservice.
Default: 128
|
Persistent Datastore
|
|
Topology Acquisition
|
retry-before-exit-count—Specify the number of times the topology server can request a topology
refresh (after it connects to a BMP topology source) to
obtain the topology that contains the links. If the
topology server doesn’t receive the topology even after
the specified refresh requests, it exits the topology
server application.
Default: 0 (meaning infinite retries)
retry-delay—Specify the maximum time (in seconds) that the topology server can wait before
requesting a topology refresh again (after it connects
to a BMP topology source).
Example: 30s or 30seconds
Default: 5s
reconnection-count—Specify the number of times the topology server can try to reconnect to the
BGP-LS topology source before exiting the topology
server application.
Default: 0 (meaning infinite retries)
reconnection-delay—Specify the maximum time (in seconds) that the topology server can wait before
attempting to reconnect to the BGP-LS topology source.
Example: 30s or 30seconds
Default: 5s
refresh-holddown—Specify the maximum time (in seconds) that the topology server can wait before
requesting for a topology refresh after it connects to a
BMP topology source but doesn’t receive the
topology.
Example: 100s or 100seconds
Default: 300s
eor-timeout—: Specify the maximum time (in seconds) that the topology server can wait to receive
the topology from the BGP-LS topology source. After the
complete topology is sent, the topology source sends an
end-of-Routing Information Base (end-of-RIB or EOR)
message that indicates the completion of topology
update. If the topology server doesn’t receive this EOR
message within the time that you specify here, an EOR
timeout is triggered and the topology server sends
another topology refresh request.
Example: 15s or 15seconds
Default: 20s
|
Log Destination
|
-
name—Displays the name that is used to identify the log
destination configuration.
level—From the list, select the severity level of
the log messages. The available options are:
emergency—The system is unusable.
alert—Immediate action is needed.
critical—Critical condition exists.
error—Error condition.
warning—Warning condition (this is the default value).
notice—Normal but significant condition.
info—Information message.
debug—Debug message.
trace—Trace message.
– (none)—No severity level.
|
