Hi, changing my Kue to use 'ioredis'. However I have issues with backwards data since it wrapps the prefix with {}. At redis.js:72-76 it does that, mentioning that "is needed in using ioredis cluster to keep they keys in same node", however it does so with an OR check. I'm using Sentinel & ioredis, and there is no reason to add the brackets. Am I correct?
@conorcodes eventually I just solved it extracting the db value and setting it as Kue was expecting.
import URL from 'url';
const parsed = 'redis://yourredisurl?db=10'
parsed.pathname = parsed.query.db;
const jobQueue = kue.createQueue({
redis: URL.format(parsed),
});I submitted the bug on github and provided a pull request, but nothing happened :P, this solves it temporarily.
Does anyone know how to release the open redis connection after Job completed? For example, if there 10 jobs are on queue, 20 connections are opening for redis and this can be seen using
I need those connections to be closed to make sure the file open limit can be shared across other processes.
sudo lsof -i tcp:6379 | wc -lI need those connections to be closed to make sure the file open limit can be shared across other processes.
does kue support remote worker setup
@abasaro_twitter Do you mean like this:
https://olg.io/blog/2016/distributed-job-processing-with-nodejs/
https://olg.io/blog/2016/distributed-job-processing-with-nodejs/
Hello poeple!
Is there a way to handle "dissapearing" workers with Kue?
I'm trying to apply it in a distributed setting where the workers resides on remote nodes that occasionally get killed for arbitrary reasons.
Now these jobs just remain in the queue as active forever.
Is there a way to do some kind of timeout or similar?
Is there a way to handle "dissapearing" workers with Kue?
I'm trying to apply it in a distributed setting where the workers resides on remote nodes that occasionally get killed for arbitrary reasons.
Now these jobs just remain in the queue as active forever.
Is there a way to do some kind of timeout or similar?
Hi there - I'm new to kue and I'm having an issue with getting started. Do I need any other dependencies before installing kue via npm? Right now I'm getting a "Redis connection to 127.0.0.1:6379 failed - connect ECONNREFUSED 127.0.0.1:6379" error when trying to use kue.createQueue(); - Do I also need to install redis via npm?
hi. got some trouble.
"test": "export NODE_ENV=test && mocha server/test i am exporting 'test' to NODE_ENV. my app picks it up and loads the respective configuration but the new worker processes do no pick up the NODE_ENV and are using a different loaded config (default = development)
is there some common practice in how to handle this situation?
Hello, on the
queue.process() can I get the first parameter as a variable? I'm having like 6 different jobs, and would love to be able to write just one queue process just to keep my code down, as right now there are 6 exact equal functions just calling other functions with strings instead of variables.
I was thinking something like
queue.process(type, 1, (job, done) => { jobs[type](job.data).then(result) => {done();} });
I have a usecase in which I have to abort scheduled jobs, the best solution would be to find them by some root attributes of the job-data
Ideally - kue can feature that.
Also acceptable - query it on the REDIS
Also acceptable - query it on the REDIS
ideas?
There is a main server.js file listening on port 80, and in it a task is created for kue, in a parallel process kue picks up these tasks and executes, which open new port. Is that statement right ?
anyway kue works pretty well and we are using it.
(as a support for delayed messages in RabbitMQ)
But you can use redis directly as well. Keys are HSET, so run redis-cli, then
KEYS * will give you what's inside, HGETALL key will give you all information about job. HGET key data will return your message.
