search

Juniper / open-nti

Open Network Telemetry Collector build with open source tools
Apache License 2.0
233 stars 93 forks source link

Fresh install - Grafana graphs empty (MX routers, Junos 17.4, Ubuntu server) #249

Open nzjc opened 5 years ago

nzjc commented 5 years ago

Hi there

I have followed the instructions (https://open-nti.readthedocs.io/en/latest/install.html) which are a tad out of date, but basically on my Ubuntu 14.04 server I ran the usual apt updates, then:

$ git clone https://github.com/Juniper/open-nti.git $ cd open-nti $ make start

Everything starts up as it should.

Docker PS

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES fd84fc77e8ca opennti_input-snmp "/source/start-inp..." 4 minutes ago Up 4 minutes 0.0.0.0:162->162/udp opennti_input_snmp 214ca3966a9d juniper/open-nti-input-jti:latest "/bin/sh -c /fluen..." 4 minutes ago Up 4 minutes 0.0.0.0:50000->50000/udp, 24284/tcp, 40020/udp, 0.0.0.0:50020->50020/udp opennti_input_jti e5cd0f02cc31 quay.io/influxdb/chronograf:1.5.0.1 "/usr/bin/chronogr..." 4 minutes ago Up 4 minutes 0.0.0.0:8888->8888/tcp chronograf_con d164ebbfc4da juniper/open-nti-input-syslog:latest "/bin/sh -c /home/..." 4 minutes ago Up 4 minutes 5140/tcp, 24220/tcp, 24224/tcp, 0.0.0.0:6000->6000/udp opennti_input_syslog d5111e0671b4 kapacitor:1.5.0 "/entrypoint.sh ka..." 4 minutes ago Up 4 minutes 0.0.0.0:9092->9092/tcp kapacitor aaf81a1a0be0 opennti_input-oc "/entrypoint.sh /s..." 4 minutes ago Up 4 minutes 8092/udp, 8125/udp, 8094/tcp, 0.0.0.0:50051->50051/udp opennti_input_oc 580c44001075 juniper/open-nti:latest "/sbin/my_init" 4 minutes ago Up 4 minutes 0.0.0.0:80->80/tcp, 0.0.0.0:3000->3000/tcp, 0.0.0.0:8083->8083/tcp, 0.0.0.0:8086->8086/tcp, 0.0.0.0:8125->8125/udp opennti_con

When I tcpdump, I can see udp traffic coming in from my MX routers (MX960s, MX480s, MX204s, on a variety of Junos versions, 4/5 of them on 17.4, one on 15.1F6):

me@lab-docker:~/open-nti$ sudo tcpdump -i eth1.999 udp tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth1.999, link-type EN10MB (Ethernet), capture size 262144 bytes 12:28:23.077599 IP 192.168.254.1.50000 > 192.168.254.58.50000: UDP, length 1304 12:28:23.082022 IP 192.168.254.1.50000 > 192.168.254.58.50000: UDP, length 1073 12:28:23.894927 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 549 12:28:23.895002 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 157 12:28:23.896354 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 180 12:28:23.899546 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 1145

My Juniper config is like so:

me@mx5> show configuration services analytics | no-more streaming-server open-nti-lab-docker { remote-address 192.168.254.58; remote-port 50000; } export-profile open-nti-lab { local-address 192.168.254.9; local-port 50000; reporting-rate 10; format gpb; transport udp; } sensor phy-interfaces { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/interface/; } sensor log-interfaces { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/interface/logical/usage/; } sensor cpu-mem { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/cpu/memory/; } sensor pfe-lsp-usage { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/services/label-switched-path/usage/; } sensor firewall-stats { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/firewall/; } sensor pfe-fabric { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/fabric/; } sensor npu-memory { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/npu/memory/; }

In influx, when I switch to the 'Juniper' database and run 'show measurements' I get the following:

name

juniperNetworks.cpu_memory_util_ext juniperNetworks.fabricMessageExt juniperNetworks.jnprLogicalInterfaceExt juniperNetworks.jnpr_firewall_ext juniperNetworks.jnpr_interface_ext juniperNetworks.jnpr_lsp_statistics_ext

However, I see nothing in any of the graphs in Grafana. Has something changed in one of the tools and not updated in another?

I can give any other details required, happy to share screens etc if useful.

Cheers

agould123 commented 5 years ago

Try chronograf and see if you can see telem data in there….

http://192.168.254.58:8888

click “Data Explorer” in the left pane/column

click “juniper.four_weeks”

click “juniperNetworks.jnpr_interface_ext”

… > device – (some number)

Do you see your devices listed there?

From: James [mailto:notifications@github.com] Sent: Monday, December 17, 2018 6:35 AM To: Juniper/open-nti Cc: Subscribed Subject: [Juniper/open-nti] Fresh install - Grafana graphs empty (MX routers, Junos 17.4, Ubuntu server) (#249)

Hi there

I have followed the instructions (https://open-nti.readthedocs.io/en/latest/install.html) which are a tad out of date, but basically on my Ubuntu 14.04 server I ran the usual apt updates, then:

$ git clone https://github.com/Juniper/open-nti.git $ cd open-nti $ make start

Everything starts up as it should.

Docker PS

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES fd84fc77e8ca opennti_input-snmp "/source/start-inp..." 4 minutes ago Up 4 minutes 0.0.0.0:162->162/udp opennti_input_snmp 214ca3966a9d juniper/open-nti-input-jti:latest "/bin/sh -c /fluen..." 4 minutes ago Up 4 minutes 0.0.0.0:50000->50000/udp, 24284/tcp, 40020/udp, 0.0.0.0:50020->50020/udp opennti_input_jti e5cd0f02cc31 quay.io/influxdb/chronograf:1.5.0.1 "/usr/bin/chronogr..." 4 minutes ago Up 4 minutes 0.0.0.0:8888->8888/tcp chronograf_con d164ebbfc4da juniper/open-nti-input-syslog:latest "/bin/sh -c /home/..." 4 minutes ago Up 4 minutes 5140/tcp, 24220/tcp, 24224/tcp, 0.0.0.0:6000->6000/udp opennti_input_syslog d5111e0671b4 kapacitor:1.5.0 "/entrypoint.sh ka..." 4 minutes ago Up 4 minutes 0.0.0.0:9092->9092/tcp kapacitor aaf81a1a0be0 opennti_input-oc "/entrypoint.sh /s..." 4 minutes ago Up 4 minutes 8092/udp, 8125/udp, 8094/tcp, 0.0.0.0:50051->50051/udp opennti_input_oc 580c44001075 juniper/open-nti:latest "/sbin/my_init" 4 minutes ago Up 4 minutes 0.0.0.0:80->80/tcp, 0.0.0.0:3000->3000/tcp, 0.0.0.0:8083->8083/tcp, 0.0.0.0:8086->8086/tcp, 0.0.0.0:8125->8125/udp opennti_con

When I tcpdump, I can see udp traffic coming in from my MX routers (MX960s, MX480s, MX204s, on a variety of Junos versions, 4/5 of them on 17.4, one on 15.1F6):

me@lab-docker:~/open-nti$ sudo tcpdump -i eth1.999 udp tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth1.999, link-type EN10MB (Ethernet), capture size 262144 bytes 12:28:23.077599 IP 192.168.254.1.50000 > 192.168.254.58.50000: UDP, length 1304 12:28:23.082022 IP 192.168.254.1.50000 > 192.168.254.58.50000: UDP, length 1073 12:28:23.894927 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 549 12:28:23.895002 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 157 12:28:23.896354 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 180 12:28:23.899546 IP 192.168.254.4.50000 > 192.168.254.58.50000: UDP, length 1145

My Juniper config is like so:

me@mx5> show configuration services analytics | no-more streaming-server open-nti-lab-docker { remote-address 192.168.254.58; remote-port 50000; } export-profile open-nti-lab { local-address 192.168.254.9; local-port 50000; reporting-rate 10; format gpb; transport udp; } sensor phy-interfaces { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/interface/; } sensor log-interfaces { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/interface/logical/usage/; } sensor cpu-mem { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/cpu/memory/; } sensor pfe-lsp-usage { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/services/label-switched-path/usage/; } sensor firewall-stats { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/firewall/; } sensor pfe-fabric { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/fabric/; } sensor npu-memory { server-name open-nti-lab-docker; export-name open-nti-lab; resource /junos/system/linecard/npu/memory/; }

In influx, when I switch to the 'Juniper' database and run 'show measurements' I get the following:

name

juniperNetworks.cpu_memory_util_ext juniperNetworks.fabricMessageExt juniperNetworks.jnprLogicalInterfaceExt juniperNetworks.jnpr_firewall_ext juniperNetworks.jnpr_interface_ext juniperNetworks.jnpr_lsp_statistics_ext

However, I see nothing in any of the graphs in Grafana. Has something changed in one of the tools and not updated in another?

I can give any other details required, happy to share screens etc if useful.

Cheers

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Juniper/open-nti/issues/249 , or mute the thread https://github.com/notifications/unsubscribe-auth/Am_xaJt-n0e_WRPtHfe55Ujyqt96cW6nks5u5492gaJpZM4ZWRLs . https://github.com/notifications/beacon/Am_xaJ44vT_NtWTav9t0eiq1SY9ekZ0eks5u5492gaJpZM4ZWRLs.gif

nzjc commented 5 years ago

Thanks @agould123 - I have just been reading older closed issues (I am quite bad at using github it seems!) and saw that Cronograf is the replacement for InfluxDB's front end, even though the latter still seems present.

I do see some telemetry graphs by clicking around in there, yes. I see 4 of my 5 hosts present. From this, I take it that the Grafana dashboards that Open-NTI comes with are no longer valid? The queries don't seem to work any longer - did I miss some documentation update somewhere?

Thanks for your reply

panks21 commented 5 years ago

There is certain need of the document update. It seems that the old DB root for InfluxDB has been changed to juniper.four_weeks. The queries needs to be updated like this SELECT mean("interface_stats.egress_stats.if_pkts") AS "mean_interface_stats.egress_stats.if_pkts", mean("interface_stats.ingress_stats.if_pkts") AS "mean_interface_stats.ingress_stats.if_pkts" FROM "juniper"."four_weeks"."juniperNetworks.jnpr_interface_ext" WHERE time > now() - 1h AND "interface_stats.if_name"='ge-0/0/0' GROUP BY time(:interval:) FILL(null)

cylmad commented 5 years ago

Hi there, I'm facing the same issue with grafana not showing any data points in the dashboards (but it works fine using Chronograf). Any idea to solve this grafana pb ? I tried many queries adjustments (including the one provided above) but nothing helps. Thx for your help.

[EDIT] I finally succeed in graphing something in grafana. The provided dashboard has to be update to take in account the four_weeks retention policy.

nzjc commented 5 years ago

Thanks for the suggestions. I am not really that well versed in the nuts and bolts here.. Is it likely this will be updated and I can just pull the containers down in a working order? :)

lwbnet commented 4 years ago

Hi there, I'm facing the same issue with grafana not showing any data points in the dashboards (but it works fine using Chronograf). Any idea to solve this grafana pb ? I tried many queries adjustments (including the one provided above) but nothing helps. Thx for your help.

[EDIT] I finally succeed in graphing something in grafana. The provided dashboard has to be update to take in account the four_weeks retention policy.

Hi @cylmad - I'm hitting this issue also, I can see data in Chronograf but nothing in Grafana - can you tell me how to update the Grafana dashboards to take into account the four_weeks policy?

agould123 commented 4 years ago

I also had more success with displaying telemetry data using Chronograf.

While at Juniper NXTWORK 2019 in Vegas, I was in a session on telemetry and a very knowledge presenter on the material showed my later how to make Grafana work.

It was simple as you just have to select the proper data source (influxdb, or perhaps whatever data source you are using) and then select field and time and grouped by and format.

-Aaron

From: lwbnet [mailto:notifications@github.com] Sent: Saturday, April 11, 2020 4:01 PM To: Juniper/open-nti Cc: Aaron Gould; Mention Subject: Re: [Juniper/open-nti] Fresh install - Grafana graphs empty (MX routers, Junos 17.4, Ubuntu server) (#249)

Hi there, I'm facing the same issue with grafana not showing any data points in the dashboards (but it works fine using Chronograf). Any idea to solve this grafana pb ? I tried many queries adjustments (including the one provided above) but nothing helps. Thx for your help.

[EDIT] I finally succeed in graphing something in grafana. The provided dashboard has to be update to take in account the four_weeks retention policy.

Hi @cylmad https://github.com/cylmad - I'm hitting this issue also, I can see data in Chronograf but nothing in Grafana - can you tell me how to update the Grafana dashboards to take into account the four_weeks policy?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Juniper/open-nti/issues/249#issuecomment-612512656 , or unsubscribe https://github.com/notifications/unsubscribe-auth/AJX7C2COTUKK2JVMMRZXZ5LRMDLABANCNFSM4GKZCLWA . https://github.com/notifications/beacon/AJX7C2DWV3GMTXKSE7WJAATRMDLABA5CNFSM4GKZCLWKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOESBDHEA.gif