I'm still quite new to pm2 and currently trying to get a node app running. This workes fine except the fact that the app gets restarted without any reason.
The app runs fine without any crash on the development machine. The log files do not show any crash. (Already testet that the log files would log a crash.)
Did I configure pm2 wrong?
ecosystem.json
{
"name" : "app1",
"cwd" : "/home/pm2/apps/app1/prod",
"script" : "dist/main.js",
"log_date_format" : "YYYY-MM-DD HH:mm Z",
"error_file" : "/home/pm2/apps/app1/logs/stderr.log",
"out_file" : "/home/pm2/apps/app1/logs/stdout.log",
"pid_file" : "/home/pm2/apps/app1/pids/app1.pid",
"instances" : 1, //or 0 => 'max'
"min_uptime" : "200s", // 200 seconds, defaults to 1000
"max_restarts" : 10, // defaults to 15
"max_memory_restart": "1M", // 1 megabytes, e.g.: "2G", "10M", "100K", 1024 the default unit is byte.
"cron_restart" : "1 0 * * *",
"watch" : false,
"ignore_watch" : "./logs",
"exec_interpreter" : "node",
"env": {
"NODE_ENV": "production"
}
}
logs/stdout.log
2016-10-24 14:28 +02:00: server listening on port 3100
2016-10-24 14:28 +02:00: server listening on port 3100
2016-10-24 14:29 +02:00: server listening on port 3100
2016-10-24 14:29 +02:00: server listening on port 3100
2016-10-24 14:30 +02:00: server listening on port 3100
2016-10-24 14:30 +02:00: server listening on port 3100
logs/stderr.log
//empty...
This directive could be problem: 'max_memory_restart'. One megabyte for app is pretty low value. Try to increase to ~200MB - depending on your app memory requirements.
EDIT: this directive is expressing maximum amount of memory which can be consumpted. If process will allocate more, it will be restarted.
Related
I am trying to deploy a meteor Application, But I am receiving this error message on the Verifying Deployment section with the following error message -
------------------------------------STDERR------------------------------------
: (7) Failed to connect to 172.17.0.2 port 3000: Connection refused
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
curl: (7) Failed to connect to 172.17.0.2 port 3000: Connection refused
=> Logs:
=> Setting node version
NODE_VERSION=14.17.4
v14.17.4 is already installed.
Now using node v14.17.4 (npm v6.14.14)
default -> 14.17.4 (-> v14.17.4 *)
=> Starting meteor app on port 3000
=> Redeploying previous version of the app
When I do the sudo netstat -tulpn | grep LISTEN in the server it shows this
tcp 0 0 10.0.3.1:53 0.0.0.0:* LISTEN 609/dnsmasq
tcp 0 0 127.0.0.53:53 0.0.0.0:* LISTEN 406/systemd-resolve
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 745/sshd: /usr/sbin
tcp6 0 0 :::22 :::* LISTEN 745/sshd: /usr/sbin
When I run sudo docker ps i receive the following message -
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
e51b1b4bf3a3 mup-appName:latest "/bin/sh -c 'exec $M…" About an hour ago Restarting (1) 49 seconds ago appName
68b723183f3d mongo:3.4.1 "/entrypoint.sh mong…" 9 days ago Restarting (100) 9 seconds ago mongodb
In my firewall i have also opened the Port 3000
If I check the Docker is running it seems like there is no docker running!!
Also in my mup.js file I am using http and not https
module.exports = {
servers: {
one: {
host: 'xx.xx.xxx.xxx',
username: 'ubuntu',
pem: '/home/runner/.ssh/id_rsa'
}
},
meteor: {
name: 'appName',
path: '../../',
docker: {
image: 'zodern/meteor:latest',
},
servers: {
one: {}
},
buildOptions: {
serverOnly: true
},
env: {
PORT: 3000,
ROOT_URL: 'http://dev-api.appName.com/',
NODE_ENV: 'production',
MAIL_URL: 'smtp://xxxx:xxx/eLPCB3nw3jubkq:#email-smtp.eu-north-1.amazonaws.com:587',
MONGO_URL: 'mongodb+srv://xxx:xx#xxx.iiitd.mongodb.net/Development?retryWrites=true&w=majority'
},
deployCheckWaitTime: 15
}
proxy: {
domains: 'dev.xxx.com',
ssl: {
letsEncryptEmail: 'info#xxx.com'
}
}
}
Any idea what might cause this issue?
I don't know why, but in the MUP docs the correct image name is zodern/meteor:root
If your app is slow to start, increase the deployCheckWaitTime . In my complex apps I put 600, just to ensure the app is up.
I'm trying to run Logstash in an EC2 Ubuntu instance,
but when I run:
logstash-5.2.0/bin/logstash -f logstash.conf --debug
I get:
Starting puma
Trying to start WebServer {:port=>9600}
start
Trying to start WebServer {:port=>9601}
[api-service] start
Successfully started Logstash API endpoint {:port=>9601}
PeriodicPoller: Stopping
stopping pipeline {:id=>"main"}
Closing inputs
stopping {:plugin=>"LogStash::Inputs::Http"}
Closed inputs
This is logstash.conf
input
{
http
{
host => "127.0.0.1"
port => 31311
}
}
output
{
elasticsearch
{
hosts => ["localhost:9200"]
}
stdout
{
codec => rubydebug
}
}
When I run
curl 'http://localhost:9200/?pretty'
I get:
{
"name" : "QrRfI_U",
"cluster_name" : "elasticsearch",
"cluster_uuid" : "LLdvAaAsQSCULTfl_b4xIA",
"version" : {
"number" : "5.2.0",
"build_hash" : "24e05b9",
"build_date" : "2017-01-24T19:52:35.800Z",
"build_snapshot" : false,
"lucene_version" : "6.4.0"
},
"tagline" : "You Know, for Search"
}
So elasticsearch is running fine.
What if you have your hosts as:
hosts => "localhost"
And make sure that the http port which you've mentioned above is not bound to any other process.
If that's not the case just to make sure, run plugin list and check whether http-input plugin does exist.
We have created a real time chat application, with a socket.io, nodejs & mongodb. It is working fine on to a local & modulus server, but not working as expected on AWS.
Socket gets disconnected randomly with a ping timeout after around 60 sec. I also set "Heart Timeout" & "Heart Interval" but still it gets disconnected.
Here is the attached config file for node,
var config = {
local : {
mode : "LOCAL",
port : 8080,
db_path : "mongodb://localhost/local_db",
site_loc : "http://dummy.local/",
api_loc : "http://dummy.dummy.com/"
},
dev : {
mode : "DEV",
port : 8080,
db_path : "mongodb://dbath:27017/dev_db",
site_loc : 'http://dummy.dummy.com/',
api_loc : 'http://dummy.dummy.com/'
},
stage : {
mode : "STAGE",
port : 3000,
db_path : "mongodb://localhost:27017/stage_db",
site_loc : 'http://dummy.dummy.com/',
api_loc : 'http://dummy.dummy.com/'
},
production : {
mode : "PROD",
port : 443,
db_path : "mongodb://localhost:27017/live_db",
site_loc : 'https://dummy.com/',
api_loc : 'https://dummy.dummy.com/'
}
}
module.exports = function(mode) {
return config[mode || process.argv[2]] || config.local;
}
I resolved this issue, by just adding a port number besides link like,
http://www.chat.com:3000/
by d way thanks everyone...
Installed an ELK server via: https://www.digitalocean.com/community/tutorials/how-to-install-elasticsearch-logstash-and-kibana-elk-stack-on-centos-7
It seems to work except for the filebeat connection; filebeat does not appear to be forwarding anything or at least I can't find anything in the logs to indicate anything is happening.
My filebeat configuration is as follows:
filebeat:
prospectors:
-
paths:
- /var/log/*.log
- /var/log/messages
- /var/log/secure
encoding: utf-8
input_type: log
timeout: 30s
idle_timeout: 30s
registry_file: /var/lib/filebeat/registry
output:
logstash:
hosts: ["my_elk_fqdn:5044"]
bulk_max_size: 1024
compression_level: 3
worker: 1
tls:
certificate_authorities: ["/etc/pki/tls/certs/logstash-forwarder.crt"]
shipper:
logging:
to_files: true
files:
path: /var/log/filebeat
name: filebeat.log
rotateeverybytes: 10485760 # = 10MB
keepfiles: 7
level: debug
The log file output I keep getting from filebeat is just not very helpful:
2016-07-14T17:32:21-04:00 DBG Start next scan
2016-07-14T17:32:31-04:00 DBG Start next scan
2016-07-14T17:32:41-04:00 DBG Start next scan
2016-07-14T17:32:46-04:00 DBG Flushing spooler because of timeout. Events flushed: 0
2016-07-14T17:32:51-04:00 DBG Start next scan
Is there anything wrong with my configuration file?
When I test on the ELK server to see if I am getting anything:
[root#my_elk_server ~]# curl -XGET 'http://localhost:9200/filebeat-*/_search?pretty'
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 0,
"successful" : 0,
"failed" : 0
},
"hits" : {
"total" : 0,
"max_score" : 0.0,
"hits" : [ ]
}
}
Oh and my logstash configuration for filebeats:
input {
beats {
port => 5044
ssl => true
ssl_certificate => "/etc/pki/tls/certs/logstash-forwarder.crt"
ssl_key => "/etc/pki/tls/private/logstash-forwarder.key"
}
}
UPDATE: It is not filebeat. Somewhat relieved that messages are indeed being passed but still have an issue I can't track:
Discovered it wasn't filebeat that was causing the issue. It appears that the configuration file in logstash to send to elasticsearch is not properly labeling the index (and the type) to make it searchable as shown in the question. Instead of putting filebeat in the index name it gives a result like this:
"_index" : "%{[#metadata][beat]}-2016.07.14",
The elasticsearch output put in the file turned out to be incorrect in the
output {
elasticsearch {
hosts => "my_elk_fqdn:9200"
sniffing => true
manage_template => false
index => "%{[#metadata][beat]}-%{+YYYY.MM.dd}"
document_type => "%{[#metadata][type]}"
}
}
Apparently this #metadata is not being passed in correctly. Has anyone been able to get the _index and _type fields to populate correctly???
This might be a bug with filebeat??
https://github.com/logstash-plugins/logstash-input-beats/issues/6
I'm developing an ExpressJS app.
I use pm2 to load it:
myapp$ pm2 start bin/www
This works fine, except that adding the --watch flag doesn't seem to work; every time I change the JS source I need to explicitly restart it for my changes to take effect:
myapp$ pm2 restart www
What am I doing wrong? I've tried the --watch flag with a non-ExpressJS app and it worked as expected.
See this solution in Stack Overflow
The problem is relative to the path where pm2 is watching, and if it is relative to the execution file or the actual root path of the project.
2021 Feb.
Things have changed a bit now. Gave a full example below from my project. Below works:
1 . Create config file. File: ecosystem.config.js
module.exports = {
apps: [
{
name: 'api',
script: './bin/www', // --------------- our node start script here like index.js
// ------------------------------------ watch options - begin
watch: ['../'],
watch_delay: 1000,
ignore_watch: ['node_modules'],
watch_options: {
followSymlinks: false,
},
// ------------------------------------ watch options - end
env: {
NODE_ENV: 'development',
PORT: 3001,
DEBUG: 'api:*',
MONGODB_URI:
'mongodb://localhost:27017/collection1?readPreference=primary&ssl=false',
},
env_production: {
NODE_ENV: 'production',
},
},
],
deploy: {
production: {
// user: "SSH_USERNAME",
// host: "SSH_HOSTMACHINE",
},
},
};
2 . Run server (dev/ prod)
pm2 start ecosystem.config.js
pm2 start ecosystem.config.js --env production
3 . More information :
https://pm2.keymetrics.io/docs/usage/watch-and-restart/
You need to specify the app location to the --watch option
myapp$ pm2 start bin/www --watch /your/location/to/app
I never managed to make default watch settings work in Ubuntu, however using polling via advanced watch options worked:
"watch": true,
"ignore_watch" : ["node_modules"],
"watch_options": {
"usePolling": true,
"interval": 1000
}
More info:
https://github.com/buunguyen/PM2/blob/master/ADVANCED_README.md#watch--restart
https://github.com/paulmillr/chokidar#api