graphile_worker 0.13.3

High performance Rust/PostgreSQL job queue (also suitable for getting jobs generated by PostgreSQL triggers/functions out into a different work queue)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148

use std::sync::Arc;

use futures::FutureExt;
use tracing::{debug, error, warn};

use crate::background_tasks::BackgroundTasks;
use crate::Worker;
use graphile_worker_queries::worker_heartbeat::registration::{
    worker_deregister, worker_heartbeat,
};
use graphile_worker_recovery::{sweep_stale_workers, SweepStaleWorkersOptions};
use graphile_worker_runtime as runtime;

pub(crate) async fn register_worker(
    worker: &Worker,
    metadata: Option<serde_json::Value>,
) -> Result<(), crate::errors::GraphileWorkerError> {
    if !worker.recovery_config.enabled {
        return Ok(());
    }

    worker_heartbeat(
        &worker.database,
        &worker.schema,
        &worker.worker_id,
        metadata,
    )
    .await
}

pub(crate) async fn deregister_worker(
    worker: &Worker,
) -> Result<(), crate::errors::GraphileWorkerError> {
    if !worker.recovery_config.enabled {
        return Ok(());
    }

    worker_deregister(&worker.database, &worker.schema, &worker.worker_id).await
}

pub(crate) struct RecoveryTasks {
    tasks: BackgroundTasks,
}

impl RecoveryTasks {
    pub(crate) fn empty() -> Self {
        Self {
            tasks: BackgroundTasks::new("recovery"),
        }
    }

    pub(crate) async fn stop(self) {
        self.tasks.stop().await;
    }
}

pub(crate) fn spawn_recovery_tasks(worker: &Worker) -> RecoveryTasks {
    if !worker.recovery_config.enabled {
        return RecoveryTasks::empty();
    }

    let worker = Arc::new(worker.clone_for_recovery());
    let heartbeat_worker = worker.clone();
    let heartbeat_handle = runtime::spawn(async move {
        if let Err(error) = run_heartbeat_loop(heartbeat_worker).await {
            error!(error = %error, "Worker heartbeat loop failed");
        }
    });

    let sweep_worker = worker;
    let sweep_handle = runtime::spawn(async move {
        if let Err(error) = run_sweeper_loop(sweep_worker).await {
            error!(error = %error, "Worker recovery sweeper failed");
        }
    });

    let mut tasks = BackgroundTasks::new("recovery");
    tasks.push(heartbeat_handle);
    tasks.push(sweep_handle);

    RecoveryTasks { tasks }
}

async fn run_heartbeat_loop(worker: Arc<Worker>) -> Result<(), crate::errors::GraphileWorkerError> {
    let mut interval = runtime::interval(worker.recovery_config.heartbeat_interval);
    let mut shutdown_signal = worker.shutdown_signal.clone();

    loop {
        futures::select_biased! {
            _ = (&mut shutdown_signal).fuse() => break,
            _ = interval.tick().fuse() => {
                worker_heartbeat(
                    &worker.database,
                    &worker.schema,
                    &worker.worker_id,
                    worker_recovery_metadata(),
                )
                .await?;
            }
        }
    }

    Ok(())
}

async fn run_sweeper_loop(worker: Arc<Worker>) -> Result<(), crate::errors::GraphileWorkerError> {
    let mut interval = runtime::interval(worker.recovery_config.sweep_interval);
    let mut shutdown_signal = worker.shutdown_signal.clone();

    loop {
        futures::select_biased! {
            _ = (&mut shutdown_signal).fuse() => break,
            _ = interval.tick().fuse() => {
                if let Err(error) = sweep_once(worker.clone()).await {
                    warn!(error = %error, "Worker recovery sweep failed");
                }
            }
        }
    }

    Ok(())
}

async fn sweep_once(worker: Arc<Worker>) -> Result<(), crate::errors::GraphileWorkerError> {
    let result = sweep_stale_workers(
        &worker.database,
        &worker.schema,
        Some(&worker.hooks),
        &worker.worker_id,
        &worker.recovery_config,
        SweepStaleWorkersOptions {
            ..Default::default()
        },
    )
    .await?;

    if result.worker_ids.is_empty() {
        debug!("No stale workers found during recovery sweep");
    }

    Ok(())
}

fn worker_recovery_metadata() -> Option<serde_json::Value> {
    Some(serde_json::json!({
        "pid": std::process::id(),
    }))
}