I know I’m very late to the party probably everybody has already moved on to Telemetry. But during the Annual Shutdown of Dec ‘24, I had some spare time, so I wanted to set this up in my personal lab and play around. At work, I have been testing Model Driven Telemetry(MDT) on Cisco’s IOS-XR for a while now, so I wanted to share some of my learnings through this post. I will be primarily focusing on Cisco IOS-XR’s implementation of MDT, more precisely GRPC DIAL-OUT in this post.
I’m using Cisco IOS-XRv 9000 running 7.9.1 for this exercise.
There are already tons of content on setting up the TIG(Telegraf, InfluxDB, Grafana) stack, I just want to mention here the Telegraf configuration for a quick reference.
# telegraf.conf
# Global Agent Configuration
[agent]
hostname = "telemetry-container"
flush_interval = "15s"
interval = "15s"
# gRPC Dial-Out Telemetry Listener
[[inputs.cisco_telemetry_mdt]]
transport = "grpc"
service_address = ":57000"
# Output Plugin InfluxDB
[[outputs.influxdb_v2]]
bucket = "mdtlab"
urls = ["http://127.0.0.1:8086"]
token = "enter the token here"
organization = "lab"
Configuration on IOS-XR #
As mentioned earlier, I wil be using grpc dial-out option, and the required configuration on Cisco IOS-XR for this:
!
grpc
no-tls
!
telemetry model-driven
destination-group DGroup2
address-family ipv4 192.168.100.1 port 57000
encoding self-describing-gpb
protocol grpc no-tls
!
!
sensor-group ISIS
sensor-path Cisco-IOS-XR-clns-isis-oper:isis/instances/instance/summary
sensor-path Cisco-IOS-XR-ip-rib-ipv4-oper:rib/vrfs/vrf/afs/af/safs/saf/ip-rib-route-table-names/ip-rib-route-table-name/protocol/isis/as/information
!
sensor-group MPLSTE
sensor-path Cisco-IOS-XR-mpls-te-oper:mpls-te/tunnels/summary
!
sensor-group SGroup2
sensor-path Cisco-IOS-XR-wdsysmon-fd-oper:system-monitoring/cpu-utilization
sensor-path Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary
!
sensor-group ipv4_bgp
sensor-path Cisco-IOS-XR-ipv4-bgp-oper:bgp/instances/instance/instance-active/vrfs/vrf/process-info/global
sensor-path Cisco-IOS-XR-ipv4-bgp-oper:bgp/instances/instance/instance-active/default-vrf/process-info/global
sensor-path Cisco-IOS-XR-ip-rib-ipv4-oper:rib/vrfs/vrf/afs/af/safs/saf/ip-rib-route-table-names/ip-rib-route-table-name/protocol/bgp/as
!
sensor-group rib-ipv4
sensor-path Cisco-IOS-XR-ip-rib-ipv4-oper:rib/rib-table-ids/rib-table-id/information
sensor-path Cisco-IOS-XR-ip-rib-ipv4-oper:rib/vrfs/vrf/afs/af/safs/saf/ip-rib-route-table-names/ip-rib-route-table-name/routes/route
!
sensor-group INT_STATS
sensor-path Cisco-IOS-XR-infra-statsd-oper:infra-statistics/interfaces/interface/cache/data-rate/input-data-rate
sensor-path Cisco-IOS-XR-infra-statsd-oper:infra-statistics/interfaces/interface/cache/data-rate/output-data-rate
!
subscription ISIS
sensor-group-id ISIS sample-interval 30000
destination-id DGroup2
!
subscription ISIS
sensor-group-id ISIS sample-interval 30000
destination-id DGroup2
!
subscription Sub2
sensor-group-id SGroup2 sample-interval 30000
destination-id DGroup2
!
subscription MPLSTE
sensor-group-id MPLSTE sample-interval 60000
destination-id DGroup2
!
subscription ipv4_bgp
sensor-group-id ipv4_bgp sample-interval 30000
destination-id DGroup2
!
subscription rib-ipv4
sensor-group-id rib-ipv4 sample-interval 30000
destination-id DGroup2
!
subscription INT_STATS
sensor-group-id INT_STATS sample-interval 30000
destination-id DGroup2
!
!
I have setup a bunch of things around system parameters, control plane and data plane. Please refer to Cisco Feature Navigator to get the YANG models as per your requirement.
Checks on IOS-XR #
RP/0/RP0/CPU0:PE1#show telemetry model-driven summary
Tue Dec 31 07:23:58.009 UTC
Subscriptions Total: 6 Active: 6 Paused: 0
Destination Groups Total: 1
Destinations grpc-tls: 0 grpc-nontls: 1 tcp: 0 udp: 0
dialin: 0 Active: 1 Sessions: 6 Connecting: 0
Sensor Groups Total: 6
Num of Unique Sensor Paths : 12
Sensor Paths Total: 12 Active: 12 Not Resolved: 0
Max Sensor Paths : 1000
Max Containers per path : 16
Minimum target defined cadence : 30000
Target Defined cadence factor : 2
RP/0/RP0/CPU0:PE1#
Looking at one particular subscription
RP/0/RP0/CPU0:PE1#show telemetry model-driven subscription Sub2 internal
Tue Dec 31 07:26:47.149 UTC
Subscription: Sub2
-------------
State: ACTIVE
Sensor groups:
Id: SGroup2
Sample Interval: 30000 ms
Heartbeat Interval: NA
Sensor Path: Cisco-IOS-XR-wdsysmon-fd-oper:system-monitoring/cpu-utilization
Sensor Path State: Resolved
Sensor Path: Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary
Sensor Path State: Resolved
Destination Groups:
Group Id: DGroup2
Destination IP: 192.168.100.1
Destination Port: 57000
Encoding: self-describing-gpb
Transport: grpc
State: Active
TLS : False
Total bytes sent: 4140731
Total packets sent: 73
Last Sent time: 2024-12-31 07:26:47.3658113843 +0000
Collection Groups:
------------------
Id: 4
Sample Interval: 30000 ms
Heartbeat Interval: NA
Heartbeat always: False
Encoding: self-describing-gpb
Num of collection: 5
Incremental updates: 0
Collection time: Min: 201 ms Max: 328 ms
Total time: Min: 205 ms Avg: 267 ms Max: 372 ms
Total Deferred: 0
Total Send Errors: 0
Total Send Drops: 0
Total Other Errors: 0
No data Instances: 0
Last Collection Start:2024-12-31 07:26:47.3657951022 +0000
Last Collection End: 2024-12-31 07:26:17.3628194656 +0000
Sensor Path: Cisco-IOS-XR-wdsysmon-fd-oper:system-monitoring/cpu-utilization
Sysdb Path: /oper/wdsysmon_fd/gl/*
Count: 5 Method: DATALIST Min: 201 ms Avg: 237 ms Max: 328 ms
Item Count: 11 Status: Active
Missed Collections:0 send bytes: 3376266 packets: 15 dropped bytes: 0
Missed Heartbeats: 0 Filtered Item Count: 0
success errors deferred/drops
Gets 0 0
List 0 0
Datalist 6 0
Finddata 0 0
GetBulk 0 0
Encode 0 0
Send 0 0
Id: 5
Sample Interval: 30000 ms
Heartbeat Interval: NA
Heartbeat always: False
Encoding: self-describing-gpb
Num of collection: 5
Incremental updates: 0
Collection time: Min: 10 ms Max: 45 ms
Total time: Min: 13 ms Avg: 25 ms Max: 48 ms
Total Deferred: 0
Total Send Errors: 0
Total Send Drops: 0
Total Other Errors: 0
No data Instances: 0
Last Collection Start:2024-12-31 07:26:17.3628195821 +0000
Last Collection End: 2024-12-31 07:26:17.3628216972 +0000
Sensor Path: Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary
Sysdb Path: /oper/showd/node/*/show_mem
Count: 5 Method: GET Min: 10 ms Avg: 23 ms Max: 45 ms
Item Count: 11 Status: Active
Missed Collections:0 send bytes: 3310 packets: 5 dropped bytes: 0
Missed Heartbeats: 0 Filtered Item Count: 0
success errors deferred/drops
Gets 11 0
List 6 0
Datalist 0 0
Finddata 6 0
GetBulk 0 0
Encode 0 0
Send 0 0
RP/0/RP0/CPU0:PE1#
Querying the InfluxDB
select "free_physical_memory","node_name" from "Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary"
Interactive Table View (press q to exit mode, shift+up/down to navigate tables):
Name: Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ index ┃ time ┃ free_physical_memory ┃ node_name ┃
┣━━━━━━━╋━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━━┫
┃ 1┃2024-12-30T04:28:52.419Z ┃ 652320768.0000000000┃0/RP0/CPU0 ┃
┃ 2┃2024-12-30T04:28:52.423Z ┃ 2699067392.0000000000┃0/0/CPU0 ┃
┃ 3┃2024-12-30T04:29:22.373Z ┃ 651173888.0000000000┃0/RP0/CPU0 ┃
┃ 4┃2024-12-30T04:29:22.377Z ┃ 2699079680.0000000000┃0/0/CPU0 ┃
┃ 5┃2024-12-30T04:29:52.357Z ┃ 650625024.0000000000┃0/RP0/CPU0 ┃
┃ 6┃2024-12-30T04:29:52.361Z ┃ 2699071488.0000000000┃0/0/CPU0 ┃
┃ 7┃2024-12-30T04:30:22.341Z ┃ 648966144.0000000000┃0/RP0/CPU0 ┃
┃ 8┃2024-12-30T04:30:22.346Z ┃ 2696978432.0000000000┃0/0/CPU0 ┃
┃ 9┃2024-12-30T04:30:52.431Z ┃ 647417856.0000000000┃0/RP0/CPU0 ┃
┃ 10┃2024-12-30T04:30:52.441Z ┃ 2694713344.0000000000┃0/0/CPU0 ┃
┃ 11┃2024-12-30T04:31:22.355Z ┃ 647303168.0000000000┃0/RP0/CPU0 ┃
┃ 12┃2024-12-30T04:31:22.371Z ┃ 2694729728.0000000000┃0/0/CPU0 ┃
┃ 13┃2024-12-30T04:31:52.344Z ┃ 647397376.0000000000┃0/RP0/CPU0 ┃
┃ 14┃2024-12-30T04:31:52.347Z ┃ 2694721536.0000000000┃0/0/CPU0 ┃
┃ 15┃2024-12-30T04:32:22.334Z ┃ 647389184.0000000000┃0/RP0/CPU0 ┃
┃ 16┃2024-12-30T04:32:22.339Z ┃ 2694733824.0000000000┃0/0/CPU0 ┃
┃ 17┃2024-12-30T04:32:52.356Z ┃ 645771264.0000000000┃0/RP0/CPU0 ┃
┃ 18┃2024-12-30T04:32:52.361Z ┃ 2694598656.0000000000┃0/0/CPU0 ┃
┃ 19┃2024-12-30T04:33:22.357Z ┃ 645820416.0000000000┃0/RP0/CPU0 ┃
┃ 20┃2024-12-30T04:33:22.361Z ┃ 2694590464.0000000000┃0/0/CPU0 ┃
┃ 21┃2024-12-30T04:33:52.39Z ┃ 645697536.0000000000┃0/RP0/CPU0 ┃
┃ 22┃2024-12-30T04:33:52.394Z ┃ 2694574080.0000000000┃0/0/CPU0 ┃
┃ 23┃2024-12-30T04:34:22.392Z ┃ 645849088.0000000000┃0/RP0/CPU0 ┃
┃ 24┃2024-12-30T04:34:22.396Z ┃ 2702364672.0000000000┃0/0/CPU0 ┃
┃ 25┃2024-12-30T04:34:52.364Z ┃ 645808128.0000000000┃0/RP0/CPU0 ┃
┃ 26┃2024-12-30T04:34:52.369Z ┃ 2693156864.0000000000┃0/0/CPU0 ┃
┣━━━━━━━┻━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━━┫
┃ 4 Columns, 5942 Rows, Page 1/229┃
┃ Table 1/1, Statement 1/1┃
┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┛
Equivalent query for Grafana in Flex Language Syntax
from(bucket: "mdtlab")
|> range(start: v.timeRangeStart, stop: v.timeRangeStop)
|> filter(fn: (r) => r["_measurement"] == "Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary")
|> filter(fn: (r) => r["_field"] == "free_physical_memory")
|> aggregateWindow(every: v.windowPeriod, fn: mean, createEmpty: false)
|> yield(name: "mean")
Personally, I found the Flex Language syntax a bit tricky to write from scratch, thankfully, we can use Query Builder option in InfluxDB GUI to get the appropriate Flex Language query for your needs.
Finally, it all comes together in Grafana dashboard.
Troubleshooting #
It’s all well and good when things work fine but when they don’t, we need to know what logs/traces to check to narrow down the issue.
I have simulated a failure scenario, due to which none of the sessions are Active, all are stuck in Connecting state.
RP/0/RP0/CPU0:PE1#show telemetry model-driven subscription Sub2 internal
Tue Dec 31 07:56:34.848 UTC
Subscription: Sub2RP/0/RP0/CPU0:PE1#show telemetry model-driven summary
Tue Dec 31 07:56:00.701 UTC
Subscriptions Total: 6 Active: 0 Paused: 0
Destination Groups Total: 1
Destinations grpc-tls: 0 grpc-nontls: 1 tcp: 0 udp: 0
dialin: 0 Active: 0 Sessions: 0 Connecting: 6
Sensor Groups Total: 6
Num of Unique Sensor Paths : 12
Sensor Paths Total: 12 Active: 0 Not Resolved: 0
Max Sensor Paths : 1000
Max Containers per path : 16
Minimum target defined cadence : 30000
Target Defined cadence factor : 2
Checking one particular Subscription, there are no active collection groups.
RP/0/RP0/CPU0:PE1#
-------------
State: NA
Sensor groups:
Id: SGroup2
Sample Interval: 30000 ms
Heartbeat Interval: NA
Sensor Path: Cisco-IOS-XR-wdsysmon-fd-oper:system-monitoring/cpu-utilization
Sensor Path State: Resolved
Sensor Path: Cisco-IOS-XR-nto-misc-oper:memory-summary/nodes/node/summary
Sensor Path State: Resolved
Destination Groups:
Group Id: DGroup2
Destination IP: 192.168.100.1
Destination Port: 57000
Encoding: self-describing-gpb
Transport: grpc
State: NA
TLS : False
Collection Groups:
------------------
No active collection groups
RP/0/RP0/CPU0:PE1#
Checking the telemetry model-driven traces, we can see that TCP Connection is failing to Establish.
RP/0/RP0/CPU0:PE1#show telemetry model-driven trace all reverse
Tue Dec 31 07:59:49.160 UTC
24118 wrapping entries (26496 possible, 24704 allocated, 0 filtered, 351346 total)
Dec 31 07:59:44.446 m2m/mdt/go-info 0/RP0/CPU0 t31604 5469 [mdt_go_trace_info]: mdtDialer:279 1: Reuse dialerChan 0xc000010218 for default
Dec 31 07:59:44.446 m2m/mdt/go-info 0/RP0/CPU0 t31604 5468 [mdt_go_trace_info]: getAddDialerChanWithLock:153 1:Found existing dialer for namespace default
Dec 31 07:59:44.446 m2m/mdt/go-info 0/RP0/CPU0 t31604 5467 [mdt_go_trace_info]: mdtDialer:269 1: namespace default, args 192.168.100.1:57000
Dec 31 07:59:44.445 m2m/mdt/go-info 0/RP0/CPU0 t31604 5466 [mdt_go_trace_info]: mdtConnEstablish:268 dial: target 192.168.100.1:57000
Dec 31 07:59:44.445 m2m/mdt/go-info 0/RP0/CPU0 t31604 5465 [mdt_go_trace_info]: mdtConnEstablish:229 1: req 503, TLS is false, cert , host , compression .
Dec 31 07:59:44.445 m2m/mdt/go-info 0/RP0/CPU0 t31604 5464 [mdt_go_trace_info]: mdtConnEstablish:229 1: Dialing out to 192.168.100.1:57000, req 503
Dec 31 07:59:44.445 m2m/mdt/go-info 0/RP0/CPU0 t31604 5463 [mdt_go_trace_info]: dialReqAgent:489 1: req 501, entries when dequeue is 3
Dec 31 07:59:44.445 m2m/mdt/subdb 0/RP0/CPU0 t31604 5462 [mdt_conn_grpc_establish_done]: Failed to establish conn 0
Dec 31 07:59:44.445 m2m/mdt/subdb 0/RP0/CPU0 t31604 5461 [mdt_conn_grpc_establish_done]: Got grpc establish callback 501, chan 18446744073709551615, status 1
Dec 31 07:59:44.445 m2m/mdt/go-info 0/RP0/CPU0 t31604 5460 [mdt_go_trace_error]: mdtConnEstablish:295 1: grpc service call failed, ReqId 501, 192.168.100.1:57000, rpc error: code =
Unavailable desc = connection error: desc = "transport: error while dialing: dial tcp 192.168.100.1:57000: i/o timeout"
grpc traces also reveal the same.
RP/0/RP0/CPU0:PE1#show grpc trace all reverse | i 07:59:44
Tue Dec 31 08:37:28.700 UTC
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] Channel Connectivity change to CONNECTING
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] pickfirstBalancer: UpdateSubConnState: 0xc00001e8e0, {CONNECTING <nil>}
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Subchannel picks a new address "192.168.100.1:57000" to connect
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Subchannel Connectivity change to CONNECTING
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] blockingPicker: the picked transport is not ready, loop back to repick
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Channel switches to new LB policy "pick_first"
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] ClientConn switching balancer to "pick_first"
Dec 31 07:59:44.446 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] ccResolverWrapper: sending update to cc: {[{192.168.100.1:57000 <nil> 0 <nil>}] <nil> <nil>}
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] scheme "" not registered, fallback to default scheme
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] parsed scheme: ""
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Subchannel Connectivity change to SHUTDOWN
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Channel Connectivity change to SHUTDOWN
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Channel Connectivity change to TRANSIENT_FAILURE
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] pickfirstBalancer: UpdateSubConnState: 0xc000019fb0, {TRANSIENT_FAILURE connection error: desc = "transport: error wh
ile dialing: dial tcp 192.168.100.1:57000: i/o timeout"}
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] Subchannel Connectivity change to TRANSIENT_FAILURE
Dec 31 07:59:44.445 ems/grpc 0/RP0/CPU0 t31604 EMS-GRPC: [core] grpc: addrConn.createTransport failed to connect to {192.168.100.1:57000 192.168.100.1:57000 <nil> 0 <nil>}. Err: con
nection error: desc = "transport: error while dialing: dial tcp 192.168.100.1:57000: i/o timeout"
RP/0/RP0/CPU0:PE1#
We can also check the TCP connection state on the node. Here the TCP connection is not moving past SYN_SENT, as it has not received SYN+ACK from the Server, eventually it times out.
RP/0/RP0/CPU0:PE1#bash netstat -an
Tue Dec 31 07:59:44.895 UTC
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 1 211.0.0.9:61390 192.168.100.1:57000 SYN_SENT
tcp6 0 0 :::57400 :::* LISTEN
Active UNIX domain sockets (servers and established)
Proto RefCnt Flags Type State I-Node Path
unix 2 [ ACC ] STREAM LISTENING 14646215 /var/lib/malloc_stat/9580
unix 2 [ ] DGRAM 14646222
RP/0/RP0/CPU0:PE1#
In this particular case, Server doesn’t have a route to Router’s Loopback 211.0.0.9, due to which it is unable to respond to Router’s TCP SYN, as soon I added the route, Connections started coming up.
RP/0/RP0/CPU0:PE1#show grpc trace all reverse
Tue Dec 31 08:50:05.358 UTC
83163 wrapping entries (549888 possible, 328448 allocated, 0 filtered, 149016 total)
Dec 31 08:48:04.729 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] Channel Connectivity change to READY
Dec 31 08:48:04.729 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] pickfirstBalancer: UpdateSubConnState: 0xc0001646c0, {READY <nil>}
Dec 31 08:48:04.729 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] Subchannel Connectivity change to READY
Dec 31 08:48:04.709 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] Channel Connectivity change to CONNECTING
Dec 31 08:48:04.709 ems/grpc 0/RP0/CPU0 t31542 EMS-GRPC: [core] pickfirstBalancer: UpdateSubConnState: 0xc0001646c0, {CONNECTING <nil>}
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] Subchannel picks a new address "192.168.100.1:57000" to connect
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] Subchannel Connectivity change to CONNECTING
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] blockingPicker: the picked transport is not ready, loop back to repick
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] Channel switches to new LB policy "pick_first"
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] ClientConn switching balancer to "pick_first"
Dec 31 08:48:04.708 ems/grpc 0/RP0/CPU0 t31659 EMS-GRPC: [core] ccResolverWrapper: sending update to cc: {[{192.168.100.1:57000 <nil> 0 <nil>}] <nil> <nil>}
RP/0/RP0/CPU0:PE1#bash netstat -an
Tue Dec 31 08:51:09.756 UTC
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 211.0.0.9:60665 192.168.100.1:57000 ESTABLISHED
tcp 0 0 211.0.0.9:60658 192.168.100.1:57000 ESTABLISHED
tcp 0 0 211.0.0.9:60659 192.168.100.1:57000 ESTABLISHED
tcp 0 0 211.0.0.9:60668 192.168.100.1:57000 ESTABLISHED
tcp 0 0 211.0.0.9:60661 192.168.100.1:57000 ESTABLISHED
tcp 0 0 211.0.0.9:60660 192.168.100.1:57000 ESTABLISHED
tcp6 0 0 :::57400 :::* LISTEN
Active UNIX domain sockets (servers and established)
Proto RefCnt Flags Type State I-Node Path
unix 2 [ ACC ] STREAM LISTENING 13474827 /var/lib/malloc_stat/2458
unix 2 [ ] DGRAM 13474834
RP/0/RP0/CPU0:PE1#
Validating YANG Data model using built-in Python script #
IOS-XR comes with a handy built-in python script, mdt_exec.py which can be invoked to quickly check the YANG model for any particular path.
RP/0/RP0/CPU0:PE1#run mdt_exec.py -s Cisco-IOS-XR-clns-isis-oper:isis/instance$
Tue Dec 31 08:56:16.248 UTC
{
"encoding_path": "Cisco-IOS-XR-clns-isis-oper:isis/instances/instance/summary",
"subscription_id_str": "app_TEST_200000001",
"collection_start_time": "1735635376550",
"msg_timestamp": "1735635376562",
"collection_end_time": "1735635376562",
"node_id_str": "PE1",
"data_json": [
{
"keys": [
{
"instance-name": "1"
}
],
"timestamp": "1735635376561",
"content": {
"l2-adjacencies": 3,
"l1-routers": 0,
"l1ls-ps": 0,
"l2ls-ps": 5,
"l2-routers": 5,
"running-levels": "isis-levels-2",
"l1-adjacencies": 0,
"l1l2-adjacencies": 0
}
}
],
"collection_id": "185"
},
^CDone
This can also help in verifying if the sensor-path you’re trying is valid or not. Using an invalid path will show up as Not Resolved while checking the sensor-path state.
RP/0/RP0/CPU0:PE1#show telemetry model-driven subscription INVALID_SUB
Fri Jan 3 04:09:42.372 UTC
Subscription: INVALID_SUB
-------------
State: NA
Sensor groups:
Id: INVALID_PATH
Sample Interval: 90000 ms
Heartbeat Interval: NA
Sensor Path: Cisco-IOS-XR-dnx-driver-fabric-plane-oper:fabric/plane-table/plane
Sensor Path State: Not Resolved
Destination Groups:
Group Id: DGroup2
Destination IP: 192.168.100.1
Destination Port: 57000
Encoding: self-describing-gpb
Transport: grpc
State: NA
TLS : False
Collection Groups:
------------------
No active collection groups
RP/0/RP0/CPU0:PE1#
RP/0/RP0/CPU0:PE1#run mdt_exec.py -s Cisco-IOS-XR-dnx-driver-fabric-plane-oper$
Fri Jan 3 04:10:06.983 UTC
{"Sub_id" : "200000001", "error" : "Module 'Cisco-IOS-XR-dnx-driver-fabric-plane-oper' not recognized or supported for the specified operation"}
{"Sub_id" : "200000001", "error" : "Module 'Cisco-IOS-XR-dnx-driver-fabric-plane-oper' not recognized or supported for the specified operation"}
^CDone
RP/0/RP0/CPU0:PE1#
This was just scratching the surface when it comes to possibilities with Telemetry but it was great fun setting everything up and to finally see those shinny graphs. Cisco has some very good documentation on this subject matter, please check them out if you’re thinking of setting this up at home or at work.
Reference #