lnopenmetrics / go-lnmetrics.reporter

:bar_chart: Reference implementation written in Go to collect and report of the lightning node metrics :bar_chart:

📈 go-lnmetrics-reporter 📊

Reference implementation written in Go lang to collect and report of the lightning node metrics

Table of Content

• Introduction
• Install Procedure
• How to Use
• How to Contribute
• Build With
• License

Introduction

go-lnmetrics.reporter is the reference implementation of a lnmetrics client (see definition of lnmetrics client here). It is written in Go lang to use the power of the statically compiled binary that a user can easily run on the target machine. This help to speedup the data collection process very easily.

In addition, this plugin support the following metrics, and you can also see the in the spec what data of your ln node are shared with the server.

• Metric One: See the definition on the specification lnmetrics.spec

Install Procedure

It is suggested to work with the tagged version that you can find in the release page.

The installation process required just to download the right binary from your host machine, and configure c-lightning (that it is the only implementation supported right now) to run the plugin and pass the server link to lightnind.

The configuration suggested is to use the config file, with the following content

... other stuff...
 
plugin=/path/binary/go-lnmetrics-{archirecture}
lnmetrics-urls=https://api.lnmetrics.info/query

Where the lnmetrics-urls give the possibility to specify the server where the plugin need to report the data if any. You can specify more that one server with just append another url divided by a comma.

In addition, there are the following optional parameters:

• lnmetrics-noproxy: Force the disabling of the proxy

How to Use

After running the plugin you will have the possibility to run the following rpc command from lightning-cli, and they are described below:

• metric_one start end: RPC command that give you the possibility to query the internal db and make access to the metric data collected by the plugin. The start and end need to be a string that is the timestamp or you can use the following query to make some particular query:
  • metric_one start="now": Give you the possibility to query the metric data that the plugin have in memory;
  • metric_one start="last": Give you the possibility to query the metric data that the plugin committed to the server last time.
• lnmetrics-info: RPC command that give you access to the plugin information, like version, go version and architecture this will be useful when there is some bug report or just consult the version of the plugin that the user is running.

How to Contribute

You can contribute in three different way, and them are described below:

• Bug and feature request through Github discussion/issue or discord channel;
• Bug fixing through PR;
• New metric support, this required to follow the lnmetrics.spec guide line

In addition, if you want build the project or you can start to play with it, you need the golang compiler (suggested the last one) and the golangci (see Build With section) to compile the code with the make command.

Build With

• golang-standards/project-layout
• glightning
• golangci

License

TODO

go-lnmetrics.reporter's Issues

Adding Sys library

See this library to improve the quality of system info https://github.com/zcalusic/sysinfo

metric_one: support the multi fund channel

Adding the graphql client architecture

graphql client, check the error in the payload and return back the error in go style

Typo to json, change up_times to up_time

Change to up_times

go-lnmetrics.reporter/internal/plugin/metrics_one.go

Line 87 in 40f824b

UpTimes []*channelStatus `json:"up_times"`

Improve the performance when users wants to use the plugin without the server

We can discover if the user uses the server or not from the list of URLs provided by the user.

If the user doesn't want to use the server, we can give the possibility to persist the data and keep the memory empty.

disable the plugin without unregister it

Sometime it is useful disable the plugin and stop to collect the data for some reason, so we should provide a plugin option to disable the collection like --disable-lnmetrics

Adding the feature support in the channels info

With the listeners we can have information about all the node, and this mean that we could use better the listpers and void stupid pinging around the code.

A json example can be

  {
         "id": "0242a4ae0c5bef18048fbecf995094b74bfb0f7391418d71ed394784373f41e4f3",
         "connected": true,
         "netaddr": [
            "3.124.63.44:9735"
         ],
         "features": "0252a1",
         "channels": [
            {
               "state": "CHANNELD_NORMAL",
               "scratch_txid": "eb5d7ba111d58a7296e83a873e743ec9e9d105b02d9f370490fa0a0862f0411a",
               "last_tx_fee_msat": "184000msat",
               "feerate": {
                  "perkw": 253,
                  "perkb": 1012
               },
               "owner": "channeld",
               "short_channel_id": "697850x2450x0",
               "direction": 1,
               "channel_id": "903a0a085bb5e0d94f64517d21cdcc5564cbde882e38d62c5af88dfb8eedeb6b",
               "funding_txid": "6bebed8efb8df85a2cd6382e88decb6455cccd217d51644fd9e0b55b080a3a90",
               "close_to_addr": "bc1qgmt7mrjpeqrn4f67chfyxtsvgtsxrpjeh7f326",
               "close_to": "001446d7ed8e41c8073aa75ec5d2432e0c42e0618659",
               "private": false,
               "opener": "local",
               "features": [
                  "option_static_remotekey"
               ],
               "funding": {
                  "local_msat": "234857000msat",
                  "remote_msat": "0msat"
               },
               "msatoshi_to_us": 184849898,
               "to_us_msat": "184849898msat",
               "msatoshi_to_us_min": 184849898,
               "min_to_us_msat": "184849898msat",
               "msatoshi_to_us_max": 234857000,
               "max_to_us_msat": "234857000msat",
               "msatoshi_total": 234857000,
               "total_msat": "234857000msat",
               "fee_base_msat": "0msat",
               "fee_proportional_millionths": 2000,
               "dust_limit_satoshis": 546,
               "dust_limit_msat": "546000msat",
               "max_htlc_value_in_flight_msat": 18446744073709551615,
               "max_total_htlc_in_msat": "18446744073709551615msat",
               "their_channel_reserve_satoshis": 2348,
               "their_reserve_msat": "2348000msat",
               "our_channel_reserve_satoshis": 2348,
               "our_reserve_msat": "2348000msat",
               "spendable_msatoshi": 181961898,
               "spendable_msat": "181961898msat",
               "receivable_msatoshi": 47659102,
               "receivable_msat": "47659102msat",
               "htlc_minimum_msat": 0,
               "minimum_htlc_in_msat": "0msat",
               "their_to_self_delay": 144,
               "our_to_self_delay": 144,
               "max_accepted_htlcs": 30,
               "state_changes": [
                  {
                     "timestamp": "2021-08-27T16:15:58.497Z",
                     "old_state": "CHANNELD_AWAITING_LOCKIN",
                     "new_state": "CHANNELD_NORMAL",
                     "cause": "user",
                     "message": "Lockin complete"
                  }
               ],
               "status": [
                  "CHANNELD_NORMAL:Reconnected, and reestablished.",
                  "CHANNELD_NORMAL:Funding transaction locked. Channel announced."
               ],
               "in_payments_offered": 0,
               "in_msatoshi_offered": 0,
               "in_offered_msat": "0msat",
               "in_payments_fulfilled": 0,
               "in_msatoshi_fulfilled": 0,
               "in_fulfilled_msat": "0msat",
               "out_payments_offered": 1,
               "out_msatoshi_offered": 50007102,
               "out_offered_msat": "50007102msat",
               "out_payments_fulfilled": 1,
               "out_msatoshi_fulfilled": 50007102,
               "out_fulfilled_msat": "50007102msat",
               "htlcs": []
            }
         ]
      },

Add a method in the metrics like OnInt to inizialize the metrics when the node is up

Adding in the metrics on the payment information

At the moment we have only the payment forwards information on the metrics one, it is useful to have also the payments made and received by the node.

Give the possibility to define the log path.

It is cool to have the metrics plugin in the lightning network folder.

Checking confusion on log message

I think this debug messages are pretty confusing

➜  VincentSSD cat .lightning/bitcoin/metrics.log
time="2021-10-26T09:12:26+02:00" level=debug msg="Home dir /media/vincent/VincentSSD/.lightning/bitcoin"
time="2021-10-26T09:12:26+02:00" level=info msg="Created and Connected to the database at /media/vincent/VincentSSD/.lightning/bitcoin/metrics/db"
time="2021-10-26T09:12:26+02:00" level=info msg="Loading metrics with id 1 end name metric_one"
time="2021-10-26T09:12:26+02:00" level=info msg="Metrics available on DB, loading them."
time="2021-10-26T09:12:26+02:00" level=error msg="leveldb: not found"
time="2021-10-26T09:12:26+02:00" level=info msg="No metric with empity key"
time="2021-10-26T09:12:36+02:00" level=debug msg="Calling on time function function"

inside on stop message we need to avoit rpc call

Move the architecture to OS and adding the real architecture.

Adding in the metric one the ln implementation information

I would like to have somethings like that

ln_impl: {
    name: "c-lightning",
    version: "0.10.1",
    on_experimental: ["offers"]
}

Binary blobs in history

While the repository is so young, I would recommend rewriting the history and getting rid of the ~30MB blobs in there.

To find the blobs (and their filenames), save following script as git-find-large.sh:

#!/bin/sh

git rev-list --objects --all |
  git cat-file --batch-check='%(objecttype) %(objectname) %(objectsize) %(rest)' |
  sed -n 's/^blob //p' |
  sort --numeric-sort --key=2 |
  cut -c 1-12,41- |
  $(command -v gnumfmt || echo numfmt) --field=2 --to=iec-i --suffix=B --padding=7 --round=nearest

Now run it and grep for big (MiB) files:

$ sh git-find-large.sh | grep MiB
e18cca20fe54  2.9MiB go-lnmetrics
93d8fe172466  2.9MiB go-lnmetrics
96f8db4416d4  3.1MiB go-lnmetrics
085389b7d797  4.0MiB go-lnmetrics
a08d2baee46a  4.0MiB go-lnmetrics
67fdfa256be6  4.0MiB go-lnmetrics
809698f74644  4.6MiB go-lnmetrics
f42c5dcb17bb  4.6MiB go-lnmetrics
c9260359fd64  4.6MiB go-lnmetrics
ea783e785c3b  4.6MiB go-lnmetrics
8ab6dbfdca16  4.7MiB go-lnmetrics
bb87c77ee906  4.7MiB go-lnmetrics
de3b9e8d1a16  4.7MiB go-lnmetrics
152c6a8bb728  4.8MiB go-lnmetrics
39be3e928398  4.8MiB go-lnmetrics
c6d0b5bd5755  4.8MiB go-lnmetrics
eae5dd9b709b  4.8MiB go-lnmetrics
9a0f78a060b6  4.8MiB go-lnmetrics
d4b142804efe  4.8MiB go-lnmetrics
7e82588f9d75  4.8MiB go-lnmetrics

Now to remove those from history (changing all the following commits), install [git-filter-repo] first and then do this:

$  git-filter-repo --invert-paths --path go-lnmetrics --force

You shall be then left out with rewritten history where your current latest commit c04afd1 becomes 0ad5f11 and the repository ends up to be 333 KiB big (instead of almost 30 MiB).

For a reference, see samhocevar/wincompose#16

crash on raspberry pi 3b at startup

I receive the following crash at the startup of the plugin

➜  ~ lightningd --disable-plugin bcli --daemon
➜  ~ panic: unaligned 64-bit atomic operation

goroutine 29 [running]:
runtime/internal/atomic.panicUnaligned()
	/home/pi/compilers/go/src/runtime/internal/atomic/unaligned.go:8 +0x24
runtime/internal/atomic.Xadd64(0x1cb637c, 0x1)
	/home/pi/compilers/go/src/runtime/internal/atomic/atomic_arm.s:256 +0x14
github.com/niftynei/glightning/jrpc2.(*Client).NextId(...)
	/home/pi/github/go-metrics-reported/vendor/github.com/niftynei/glightning/jrpc2/client.go:229
github.com/niftynei/glightning/jrpc2.(*Client).Request(0x1cb6360, {0x279d6c, 0x3c2db4}, {0x207010, 0x1e0e080})
	/home/pi/github/go-metrics-reported/vendor/github.com/niftynei/glightning/jrpc2/client.go:166 +0x40
github.com/niftynei/glightning/glightning.(*Lightning).GetInfo(...)
	/home/pi/github/go-metrics-reported/vendor/github.com/niftynei/glightning/glightning/lightning.go:1060
github.com/OpenLNMetrics/go-metrics-reported/internal/plugin.(*MetricOne).OnInit(0x1c6a050, 0x1c988d8)
	/home/pi/github/go-metrics-reported/internal/plugin/metrics_one.go:307 +0x50
github.com/OpenLNMetrics/go-metrics-reported/internal/plugin.(*MetricsPlugin).RegisterOneTimeEvt.func1.1(0x3b1350, {0x27d870, 0x1c6a050})
	/home/pi/github/go-metrics-reported/internal/plugin/plugin.go:118 +0x30
created by github.com/OpenLNMetrics/go-metrics-reported/internal/plugin.(*MetricsPlugin).RegisterOneTimeEvt.func1
	/home/pi/github/go-metrics-reported/internal/plugin/plugin.go:117 +0xcc

The method listfund will crash

Check if the cause is the library https://github.com/niftynei/glightning/blob/927c0b8cf00364cc23afd0056d643cb656df1f44/glightning/lightning.go#L2085

the payment forwards in the list channels need to be filtered

We can filter the list forwards and add only the last payments https://lightning.readthedocs.io/lightning-listforwards.7.html and not put inside all the payments pass through the channel

Adding alias of the main node in the payload

The recurrent event doesn't run each 30 minutes

cln v0.12.0 breaking change

If we disable the deprecated api the new msat int filed break all the stuff

time="2022-08-12T16:24:46Z" level=info msg="Register one time event after 10s"
time="2022-08-12T16:24:56Z" level=debug msg="Calling on time function function"
time="2022-08-12T16:24:56Z" level=error msg="Error during the OnInit method; json: cannot unmarshal number into Go struct field NodeInfo.fees_collected_msat of type string"
time="2022-08-12T16:24:56Z" level=error msg="Error during on init call: json: cannot unmarshal number into Go struct field NodeInfo.fees_collected_msat of type string"

Now we need to find a solution for this to support both

OnClose event need to be renamed in onStop to avoid confusion

ping already pending, and channel id not found inside the log as error message

While we are running the online check on cln, I start to note the following error log

time="2022-08-20T17:03:57Z" level=error msg="Error during ping node 02b78caed0f45120acc48efe867aa506e8ea60f0712a23303178471da0ca2213f5: -1:Ping already pending"
time="2022-08-20T17:04:00Z" level=error msg="Error: No channel found for short channel id 580612x1826x0"
time="2022-08-20T17:04:00Z" level=error msg="Error: No channel found for short channel id 580612x1826x0"
time="2022-08-20T17:04:00Z" level=error msg="Error during ping node 03841f794d1380ee3003f223a3c3b83f740fa71793583a1ab4fb2647a21188fd6b: -1:Ping already pending"
time="2022-08-20T17:04:00Z" level=error msg="Error: No channel found for short channel id 588324x277x0"
time="2022-08-20T17:04:01Z" level=error msg="Error: No channel found for short channel id 588324x277x0"
time="2022-08-20T17:04:06Z" level=error msg="Error during ping node 03f2a0d82e057ed46017bc270c8cb0a926af8f46807df23ca7d821783b9cd718f4: -1:Ping already pending"
time="2022-08-20T17:04:06Z" level=error msg="Error: No channel found for short channel id 617793x3130x0"
time="2022-08-20T17:04:06Z" level=error msg="Error: No channel found for short channel id 617793x3130x0"
time="2022-08-20T17:04:06Z" level=error msg="Error during ping node 027d0de66d08f956a8d606c0d1c34e59bda38c05a3b1cc738fdd6378716c644997: -1:Ping already pending"
time="2022-08-20T17:04:06Z" level=error msg="Error: No channel found for short channel id 624949x429x0"
time="2022-08-20T17:04:06Z" level=error msg="Error: No channel found for short channel id 624949x429x0"
time="2022-08-20T17:04:28Z" level=error msg="Error during ping node 035d61eb24f59334d7318061cafd57a481d11d2fc597a873270514f7cdcfb1a12c: -1:Ping already pending"
time="2022-08-20T17:04:38Z" level=error msg="Error during ping node 03baa70886d9200af0ffbd3f9e18d96008331c858456b16e3a9b41e735c6208fef: -1:Ping already pending"
time="2022-08-20T17:04:38Z" level=error msg="Error: No channel found for short channel id 662212x672x1"
time="2022-08-20T17:04:38Z" level=error msg="Error: No channel found for short channel id 662212x672x1"
time="2022-08-20T17:05:24Z" level=error msg="Error during ping node 02b37a7de2b3a6b4b04485a2a6eba03feff0d5d34813a0401d09fbcbfedfbb1a1f: -1:Ping already pending"
time="2022-08-20T17:05:56Z" level=error msg="Error during ping node 039d746c7e5dadaf9e5afb79c9d4a89aba6eb426ddaaa833a811e20fe47428ffd1: -1:Ping already pending"
time="2022-08-20T17:05:56Z" level=error msg="Error: No channel found for short channel id 710182x479x0"
time="2022-08-20T17:05:56Z" level=error msg="Error: No channel found for short channel id 710182x479x0"
time="2022-08-20T17:05:56Z" level=error msg="Error during ping node 03e2408a49f07d2f4083a47344138ef89e7617e63919202c92aa8d49b574a560ae: -1:Ping already pending"
time="2022-08-20T17:06:19Z" level=error msg="Error during ping node 0219c2f8818bd2124dcc41827b726fd486c13cdfb6edf4e1458194663fb07891c7: -1:Ping already pending"
time="2022-08-20T17:06:19Z" level=error msg="Error: No channel found for short channel id 721589x547x1"
time="2022-08-20T17:06:19Z" level=error msg="Error: No channel found for short channel id 721589x547x1"
time="2022-08-20T17:06:31Z" level=error msg="Error during ping node 03f3c108ccd536b8526841f0a5c58212bb9e6584a1eb493080e7c1cc34f82dad71: -1:Ping already pending"

fix the status of the channel where we can take into count the channel

We should avoid example like the following status

     "channels_info": [
        {
          "channel_id": ""
        },
        {
          "channel_id": ""
        },
        {
          "channel_id": ""
        },
        {
          "channel_id": ""
        },
        {
          "channel_id": ""
        },
        {
          "channel_id": ""
        }
      ]

Adding a deprecated procedure

Have a migration procedure that deprecates some functionality it is useful when some change in the plugin and the database need to be migrated.

Error Status offered unexpected

time="2021-12-15T02:45:14Z" level=error msg="Error Status offered unexpected"

OnInit method inside the metrics on return an error

If we have an error during the on init phase, we do not storing the data. There is some error catch that we don't need to catch.

onClose event generate a bad payload?

Maybe we need to call another type the onEvent to collect the data?

plugin: Crash report

INFO    plugin-paytest.py: RPC method 'testinvoice' does not have a docstring.
INFO    plugin-paytest.py: RPC method 'paytest' does not have a docstring.
INFO    connectd: Static Tor service onion address: "beatifulnewonionaddresskindofrandombutrequiresfantasyyes.onion:9735,127.0.0.1:9735" bound from extern port 9735 
INFO    plugin-bcli: bitcoin-cli initialized and connected to bitcoind.
INFO    plugin-autoclean: autocleaning every 3600 seconds
INFO    lightningd: --------------------------------------------------
INFO    lightningd: Server started with public key 755746d3291fae3fa45461c990d4882470a04c0cf730ae8032c19fdf363d787e05, alias HAPPYGIRAFFE-v0.10.2-94-g5a5cf8c (color #a5cf8c) and lightningd v0.10.2-94-g5a5cf8c
panic: leveldb: not found

goroutine 23 [running]:
main.onInit(0xc000114930, 0xc00009b770, 0xc0000c4370)
	/root/go-lnmetrics.reporter/cmd/go-lnmetrics.reporter/main.go:90 +0x813
github.com/vincenzopalazzo/glightning/glightning.InitMethod.Call(0xc0000dc240, 0x35, 0x40, 0xc0000c4370, 0xc000114930, 0x40eb5b, 0xc0000a8318, 0x18, 0x18)
	/root/go/pkg/mod/github.com/vincenzopalazzo/[email protected]/glightning/plugin.go:1127 +0x2a7
github.com/vincenzopalazzo/glightning/jrpc2.Execute(0xc0000a8330, 0x83e680, 0xc00009b5c0, 0x83e680)
	/root/go/pkg/mod/github.com/vincenzopalazzo/[email protected]/jrpc2/server.go:196 +0x35
github.com/vincenzopalazzo/glightning/jrpc2.processMsg(0xc0000e4440, 0xc00012a000, 0x1b7, 0x1b7)
	/root/go/pkg/mod/github.com/vincenzopalazzo/[email protected]/jrpc2/server.go:192 +0x191
created by github.com/vincenzopalazzo/glightning/jrpc2.(*Server).listen
	/root/go/pkg/mod/github.com/vincenzopalazzo/[email protected]/jrpc2/server.go:120 +0x18a
INFO    plugin-go-lnmetrics: Killing plugin: exited before replying to init
INFO    plugin-summary.py: Plugin summary.py initialized

Reported by @jsarenik

Update RPC frormat about channels in up_time

The JSON propriety in the up_time payload about "channels": is unclear. It reports only the number and not the information about the status. I would like to move it on the json object with nodeid and status.

like

{
   "node_id": "0000000",
   "state":  "CHANNELD_AWAITING_LOCKIN",
}

JSON mistakes

Inside the json payload there are the following mistake

• Typos in metric one: forwards instant of forwards

Null properties in the JSON payload

Have null propriety in the JSON payload is somethings very bad, remove it with an empty array

➜  ~ clightning diagnostic 1 | grep "null"         
               "forwards": null,
               "forwards": null,

Load stored data when a node is started

It is useful when a node will start that the plugin is able to get in memory the old information of the metrics.

However, this could be not good at some point, because it could require a lot of memory if the metrics are a lot.

At this point, the question is if the memory db is good.

Filter the list channel by read and pending.

It is more correct report how many ready channels are in the node each time we will make the dump of information

adding a propriety ready in the up time of the channel

For now, when a node where we have a channel with it is down, we identify it with a timestamp = 0, that it is invalid.

So, here we are missing the timestamp information, and we are required to the client to keep track of it.

It is not important, for the first release at the end of the month, but it is good to make the life easy to the user

nil pointer: Error json: Unmarshal(nil *cache.NodeInfoCache)

after #88 the following error happens

me="2022-04-02T08:13:25Z" level=info msg="Register one time event after 10s"
time="2022-04-02T08:13:35Z" level=debug msg="Calling on time function function"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error: channel not exist for direction INCOOMING"
time="2022-04-02T08:13:37Z" level=error msg="Error: Error: channel not exist for direction INCOOMING"
time="2022-04-02T08:13:37Z" level=error msg="Error: Error: channel not exist for direction INCOOMING"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=error msg="Error json: Unmarshal(nil *cache.NodeInfoCache)"
time="2022-04-02T08:13:37Z" level=info msg="Plugin initialized with OnInit event"

adding proxy configuration for the graphql reporter.

As @openoms suggested, we need to have the client able to talk over the tor network, we can configure a client with the JSON payload passed by c-lightning to the plugin

Count the number of shared payment and add in the status struct

Filter payment forwords with success, maybe we need a object

update: up_time record

When the plugin will make a screenshot of the plugin, we are not able to catch if the new payment forwards by the node will fail or not. I would like to move the records "forwards" in an object instance in an accumulator.

From: "forwards": 184,
to:

"forwards": {
   "by_channel": "1x2x3"
   "success": 0,
   "failure": 0,
   "pending": 0,
}

In addition, it will be cool to have also from what channel this payment came, and some information about the payment id

Plugin becomes CPU hog on larger nodes

I'm running a node with ~180 channels. Normally my clightning process uses about 1% CPU. If I start running the metrics collector, things change to ~150% (equally split between the plugin and lightningd). Also (obviously?) interacting with lightningd becomes horribly slow, listpeers etc takes ~7s with the plugin running and 0.03s without.

Sadly that makes it unusable for me.

Give the possibility to enable/disable the proxy with a plugin option

ChannelsInfo as Array is a good choice?

We maintain an array of channels info https://github.com/OpenLNMetrics/go-metrics-reported/blob/0bd0b59c4931c58773019e339516a086870cf967/internal/plugin/metrics_one.go#L81 but this can duplicate some information.

Maybe is better maintains a map of channels (by short channel id) and collect the information inside it?

During the reccurrent event the plugin crash without information

Crash

➜  ~ panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x40 pc=0x6c4d26]

goroutine 5473 [running]:
github.com/OpenLNMetrics/go-lnmetrics.reporter/pkg/graphql.(*Client).MakeRequest.func1(0x0)
	/home/vincent/github/go-metrics-reported/pkg/graphql/client.go:49 +0x26
github.com/OpenLNMetrics/go-lnmetrics.reporter/pkg/graphql.(*Client).MakeRequest(0xc0000f8450, 0xc000110030, 0x81f540, 0xc002d12300)
	/home/vincent/github/go-metrics-reported/pkg/graphql/client.go:69 +0x4b4
github.com/OpenLNMetrics/go-lnmetrics.reporter/pkg/graphql.(*Client).UploadMetrics(0xc0000f8450, 0xc001c52000, 0x42, 0xc000189f00, 0x0, 0xc00011c8d0)
	/home/vincent/github/go-metrics-reported/pkg/graphql/client.go:96 +0x23b
github.com/OpenLNMetrics/go-lnmetrics.reporter/internal/plugin.(*MetricOne).Upload(0xc0000c4000, 0xc0000f8450, 0xc0000c4000, 0x1)
	/home/vincent/github/go-metrics-reported/internal/plugin/metrics_one.go:379 +0xa5
github.com/OpenLNMetrics/go-lnmetrics.reporter/internal/plugin.(*MetricsPlugin).updateAndUploadMetric(0x9f3920, 0x827850, 0xc0000c4000)
	/home/vincent/github/go-metrics-reported/internal/plugin/plugin.go:98 +0xc5
created by github.com/OpenLNMetrics/go-lnmetrics.reporter/internal/plugin.(*MetricsPlugin).RegisterRecurrentEvt.func1
	/home/vincent/github/go-metrics-reported/internal/plugin/plugin.go:110 +0x10e
panic: runtime error: invalid memory address or nil pointer dereference
                                                                       [signal SIGSEGV: segmentation violation code=0x1 addr=0x40 pc=0x6cfe46]

remove the channels with invalid status from the calculation eg. OnChain

In the uptime, we maintain the count of the channels present in the response, in this case, need to remove un the uptime of this invalid channels and maintains the channels info map/set

Periodic task library

maybe it could be a good choice as library https://stackoverflow.com/a/34150696/10854225

plugin: Shows wrong version info, and missing the log and db path

We can add different information to the plugin rpc method to show some running information. We can add information like the
logfile path, and the db path.

In addition, the payload contains a very old version of the plugin.

➜  ~ lightning-cli lnmetrics-reporter
{
   "Name": "go-lnmetrics-reporter",
   "Version": "0.1",
   "LangVersion": "Go lang 1.15.8",
   "Architecture": "amd64"
}

Adding RPC method to add a server to report the metrics.