in_node_exporter: more pluggable structure

[nokute78]

This patch is make more pluggable structure.
It is to remove similar codes for maintainability.

• Define common collector struct flb_ne_collector
• Add/rename common/platform specific source and header
  • Add ne_filefd.[ch], ne_stat.[ch] and ne_vmstat.[ch]
  • Rename ne_systemd.[ch] -> ne_systemd_linux.[ch].

struct flb_ne_collector

This struct represents common collector like plugin instance.

struct flb_ne_collector {
    const char *name;
    int coll_fd;
    int interval;
    int activated;

    /* callbacks */
    int (*cb_init) (struct flb_ne *ctx);
    int (*cb_update) (struct flb_input_instance *ins, struct flb_config *conf, void *in_context);
    int (*cb_exit) (struct flb_ne *ctx);

    struct mk_list _head;
};

Name	Description
name	Collector name. e.g. cpufreq
interval	Polling interval to collect metrics in second. If collector.%name%.scrape_interval is defined on config map, this value is set by the value of config map. Default : scrape_interval
coll_fd	File descriptor that is returned from flb_input_set_collector_time
activated	If enabled, this collector is enabled and collect metrics.
cb_init	Callback to initialize this collector. If it is NULL, this collector is not supported. It is useful for not supported platform.
cb_update	Callback to poll metrics.
cb_exit	Callback to cleanup this collector. If it is NULL, this collector is not needed to clean up.

---

Enter [N/A] in the box, if an item is not applicable to your change.

Testing
Before we can approve your change; please submit the following in a comment:

• Example configuration file for the change
• Debug log output from testing the change

• Attached Valgrind output that shows no leaks or memory corruption was found

If this is a change to packaging of containers or native binaries then please confirm it works for all targets.

• Run local packaging test showing all targets (including any new ones) build.
• Set ok-package-test label to test for all targets (requires maintainer to do).

Documentation

• Documentation required for this feature

Backporting

• [N/A] Backport to latest stable release.

Configuration

[INPUT]
    Name node_exporter_metrics
    scrape_interval 1

[OUTPUT]
    Name null

Debug/Valgrind output

$ valgrind --leak-check=full bin/fluent-bit -c a.conf 
==147192== Memcheck, a memory error detector
==147192== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==147192== Using Valgrind-3.18.1 and LibVEX; rerun with -h for copyright info
==147192== Command: bin/fluent-bit -c a.conf
==147192== 
Fluent Bit v2.2.0
* Copyright (C) 2015-2023 The Fluent Bit Authors
* Fluent Bit is a CNCF sub-project under the umbrella of Fluentd
* https://fluentbit.io

[2023/10/22 09:49:14] [ info] [fluent bit] version=2.2.0, commit=a9a2eb1c59, pid=147192
[2023/10/22 09:49:14] [ info] [storage] ver=1.5.1, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2023/10/22 09:49:14] [ info] [cmetrics] version=0.6.4
[2023/10/22 09:49:14] [ info] [ctraces ] version=0.3.1
[2023/10/22 09:49:14] [ info] [input:node_exporter_metrics:node_exporter_metrics.0] initializing
[2023/10/22 09:49:14] [ info] [input:node_exporter_metrics:node_exporter_metrics.0] storage_strategy='memory' (memory only)
[2023/10/22 09:49:14] [ info] [input:node_exporter_metrics:node_exporter_metrics.0] path.procfs = /proc
[2023/10/22 09:49:14] [ info] [input:node_exporter_metrics:node_exporter_metrics.0] path.sysfs  = /sys
[2023/10/22 09:49:14] [ info] [input:node_exporter_metrics:node_exporter_metrics.0] thread instance initialized
[2023/10/22 09:49:14] [ info] [output:null:null.0] worker #0 started
[2023/10/22 09:49:14] [ info] [sp] stream processor started
^C[2023/10/22 09:49:17] [engine] caught signal (SIGINT)
[2023/10/22 09:49:17] [ warn] [engine] service will shutdown in max 5 seconds
[2023/10/22 09:49:18] [ info] [engine] service has stopped (0 pending tasks)
[2023/10/22 09:49:18] [ info] [output:null:null.0] thread worker #0 stopping...
[2023/10/22 09:49:18] [ info] [output:null:null.0] thread worker #0 stopped
==147192== 
==147192== HEAP SUMMARY:
==147192==     in use at exit: 0 bytes in 0 blocks
==147192==   total heap usage: 47,205 allocs, 47,205 frees, 7,832,452 bytes allocated
==147192== 
==147192== All heap blocks were freed -- no leaks are possible
==147192== 
==147192== For lists of detected and suppressed errors, rerun with: -s
==147192== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)

---

Fluent Bit is licensed under Apache 2.0, by submitting this pull request I understand that this code will be released under the terms of that license.

[nokute78]

How to add new collector

1. Create common source
2. Create common header
3. Create source/header for specific platform
4. Add mk_list_add
5. Define config map
6. Add source to CMakeLists.txt

1. Create common source

e.g.
ne_new.c:

#ifdef __linux__
#include "ne_new_linux.c"
#elif __APPLE__
#include "ne_new_darwin.c"
#else
/* Not supported */
struct flb_ne_collector new_collector = {
    .name = "new",
    .cb_init = NULL,
    .cb_update = NULL,
    .cb_exit = NULL
};

#endif

2. Create common header

#ifndef FLB_IN_NE_NEW_H
#define FLB_IN_NE_NEW_H

#include "ne.h"

extern struct flb_ne_collector new_collector;

#endif

4. Add mk_list_add

ne.c:

    mk_list_add(&systemd_collector._head, &ctx->collectors);
    mk_list_add(&processes_collector._head, &ctx->collectors);

    mk_list_add(&new_collector._head, &ctx->collectors); // add here !

    mk_list_foreach(head, &ctx->collectors) {
        coll = mk_list_entry(head, struct flb_ne_collector, _head);
        collectors_common_init(coll);
    }

5. Define config map

Property name should be collector.%name%.scrape_interval . e.g. collector.new.scrape_interval.

ne.c:

    {
     FLB_CONFIG_MAP_TIME, "collector.new.scrape_interval", "0",
     0, FLB_FALSE, 0,
     "scrape interval to collect new metrics from the node."
    },

6. Add source to CMakeLists.txt

Modify plugins/in_node_exporter_metrics/CMakeLists.txt to add ne_new.c.

[cosmo0920]

It would be really good to have! I love this patch.
I confirmed that this patch does not have any side effect on macOS. Everything is working as expected.