I have created Python Application which is writing 3 types of log files in /var/log/{envName}/{applicationName} directory. One is stats logs, second is exception logs and third is access logs/application logs
Now I am using Fluentbit DaemonSet for log shipping to my central logging tool. I want to add kubernetes metadata like podName, PodId, DockerId, Hostname etc to my all log files. I am trying to use Kubernetes filter to add all above metadata but they are reading metadata from tags which is present in log file name. i.e. Log file name should be {podname}{namespace}{container_name}_{dockerId}.log and it should be exists in /var/log/container path.
Is there any way to add k8s metadata to my custom application logs files which is present in different directory.
fluentbit_configmap.yaml
[INPUT]
Name tail
Tag services.*
Path /var/log/dev/mario/*/*_service.log
Parser docker
#DB /var/log/flb_kube.db
Mem_Buf_Limit 512MB
Skip_Long_Lines On
Refresh_Interval 10
filter-kubernetes.conf: |
[FILTER]
Name kubernetes
Match *
Kube_URL https://kubernetes.default.svc.cluster.local:443
Merge_Log On
Merge_Log_Key log
K8S-Logging.Exclude On
K8S-Logging.Parser On
[FILTER]
Name nest
Match *
Operation lift
Nested_under kubernetes
Add_prefix kubernetes.
[FILTER]
Name modify
Match *
Copy kubernetes.namespace_name applicationName
Copy kubernetes.container_name subsystemName
Copy kubernetes.host computerName
Remove kubernetes.container_hash
Remove kubernetes.docker_id
Remove kubernetes.pod_id
[FILTER]
Name nest
Match *
Operation nest
Wildcard kubernetes.*
Nest_under kubernetes
Remove_prefix kubernetes.
[FILTER]
Name nest
Match *
Operation nest
Wildcard kubernetes
Wildcard log
Wildcard time
Wildcard stream
Nest_under json
I have mounted the volumes and able to read logs lines. But I am not able to add k8s metadata into this. I am getting error like Invalid Pattern for given tag
Please let me know if you need more information regarding my configurations.
Thanks
you declared 'docker' as parser but haven't defined it.
Add [PARSER] with it's attribute or remove the docker parser line.
Scenario:
You write a program in R or Python, which needs to run on Linux or Windows, you want to log (JSON structured and unstructured) std-out and (mostly unstructured) std-error from this program to a Fluentd instance. Adding a new program or starting another instance should not require to update the Fluentd configuration and the applications will not (yet) be running in a docker environment.
Question:
How to send "logs" from a bunch of programs to an fluentd instance, without the need to perform curl calls for every log entry that your application was originally writing to std-out?
When a UDP or TCP connection' is necessary for the application to run, it seems to become less easy to debug, and any dependency of your program that returns std-out will be required to be parsed, just to get it's logging passed through.
Thoughts:
Alternatively, a question could be, how to accept a 'connection' object which can either point to a file or to a TCP connection? So that switching between the std-out or a TCP destination is a matter of changing a single value?
I like the 'tail' input plugin, which could be what I am looking for, but then:
the original log file never appears to stop growing (will the trail position value reset when it is simply removed? I couldn't find this behaviour), and
it seems that it requires to reconfigure fluentd for every new program that you start on that server (if it logs in another file), I would highly prefer to keep that configuration on the program side...
I build an EFK stack with a docker logdriver set to fluentd, which does not seem to have an optimal solid solution either, but without docker, I already get kind of stuck with setting up a basic configuration (not referring to fluent.conf here).
TL;DR
std-out -> fluentd: Redirect the program output, when launching your program, to a file. On linux, use logrotate, you will love it.
Windows: use fluent-bit.
App side config: use single (or predictable) log locations, and the
fluentd/fluent-bit 'in_tail' plugin.
logging general:
It's recommended to always write application output to a file, if the std-out must be written to a file, pipe it's output at program startup. For more flexibility for the fluentd configuration, pipe them to separate files (just like 'Apache' does):
My_program.exe Do some crazy stuf > my_out_file.txt 2> my_error_file.txt
This opens the option for fluentd to read from this/these file(s).
Windows:
For Windows systems, use fluent-bit, it likely solves the issue for aggregating the Windows OS program logs. Support for Windows has just been implemented recently.
fluent-bit supports:
the 'tail' plugin, which records the 'inode' value (unique, renaming insensitive, file pointer) and the 'index' (called 'pos' for the full-blown 'fluent' application) value in a sqllite3 database and deals with un-processable data, which is allocated to a certain key ('log' by default)
Works on Windows machines, but note that it cannot buffer to disk, so be sure a lost connection, or another issue with the output, is reestablished or fixed in time so that you will not be running into OOM issues.
Appl. side config:
The tail plugin can monitor a folder, this makes it practically possible to keep the configuration on the side of your program. Just make sure you write your logs of your different applications to a predictable directory.
Fluent-bit setup/config:
For Linux, just use fluentd (unless > 100000 messages per second are required, which is where fluent-bit becomes your only choice).
For Windows, install Fluent-bit, and make it run as a deamon (almost funny sollution).
There are 2 execution methods:
Providing configuration directly via the commandline
Using a config file (example included in zip), and referring to it with the -c flag.
Directly from commandline
Some example executions (without making use of the option to work with a configuration file) can be found here:
PS .\bin\fluent-bit.exe -i winlog -p "channels=Setup,Windows PowerShell" -p "db=./test.db" -o stdout -m '*'
-i declares the input method. Currently, only a few plugins have been implemented, see the man page below.
PS fluent-bit.exe --help
Available Options
-b --storage_path=PATH specify a storage buffering path
-c --config=FILE specify an optional configuration file
-f, --flush=SECONDS flush timeout in seconds (default: 5)
-F --filter=FILTER set a filter
-i, --input=INPUT set an input
-m, --match=MATCH set plugin match, same as '-p match=abc'
-o, --output=OUTPUT set an output
-p, --prop="A=B" set plugin configuration property
-R, --parser=FILE specify a parser configuration file
-e, --plugin=FILE load an external plugin (shared lib)
-l, --log_file=FILE write log info to a file
-t, --tag=TAG set plugin tag, same as '-p tag=abc'
-T, --sp-task=SQL define a stream processor task
-v, --verbose increase logging verbosity (default: info)
-s, --coro_stack_size Set coroutines stack size in bytes (default: 98302)
-q, --quiet quiet mode
-S, --sosreport support report for Enterprise customers
-V, --version show version number
-h, --help print this help
Inputs
tail Tail files
dummy Generate dummy data
statsd StatsD input plugin
winlog Windows Event Log
tcp TCP
forward Fluentd in-forward
random Random
Outputs
counter Records counter
datadog Send events to DataDog HTTP Event Collector
es Elasticsearch
file Generate log file
forward Forward (Fluentd protocol)
http HTTP Output
influxdb InfluxDB Time Series
null Throws away events
slack Send events to a Slack channel
splunk Send events to Splunk HTTP Event Collector
stackdriver Send events to Google Stackdriver Logging
stdout Prints events to STDOUT
tcp TCP Output
flowcounter FlowCounter
Filters
aws Add AWS Metadata
expect Validate expected keys and values
record_modifier modify record
rewrite_tag Rewrite records tags
throttle Throttle messages using sliding window algorithm
grep grep events by specified field values
kubernetes Filter to append Kubernetes metadata
parser Parse events
nest nest events by specified field values
modify modify records by applying rules
lua Lua Scripting Filter
stdout Filter events to STDOUT
I am new to prometheus/alertmanager.
I have created a cron job which executes shell script every minute. This shell script generates "test.prom" file (with a gauge metric in it) in the same directory which is assigned to --textfile.collector.directory argument (to node-exporter). I verified (using curl http://localhost:9100/metrics) that the node-exporter exposes that custom metric correctly.
When I tried to run a query against that custom metric in prometheus dashboard, it does not show up any results (it says no data found).
I could not figure out why the query against the metric exposed via node-exporter textfile collector fails. Any clues what I missed ? Also please let me know how to check and ensure that prometheus scraped my custom metric 'test_metric` ?
My query in prometheus dashboard is test_metric != 0 (in prometheus dashboard) which did not give any results. But I exposed test_metric via node-exporter textfile.
Any help is appreciated !!
BTW, the node-exporter is running as docker container in Kubernetes environment.
I had a similar situation, but it was not a configuration problem.
Instead, my data included timestamps:
# HELP network_connectivity_rtt Round Trip Time to each node
# TYPE network_connectivity_rtt gauge
network_connectivity_rtt{host="home"} 53.87 1541426242
network_connectivity_rtt{host="hop_1"} 58.8 1541426242
network_connectivity_rtt{host="hop_2"} 21.93 1541426242
network_connectivity_rtt{host="hop_3"} 71.69 1541426242
PNE was picking them up without any problem once I reloaded it. As prometheus is running under systemd, I had to check the logs like this:
journalctl --system -u prometheus.service --follow
There I read this line:
msg="Error on ingesting samples that are too old or are too far into the future"
Once I removed the timestamps, values started appearing. This lead me to read more in detail about the timestamps, and I found out they have to be in miliseconds. So this format now is ok:
# HELP network_connectivity_rtt Round Trip Time to each node
# TYPE network_connectivity_rtt gauge
network_connectivity_rtt{host="home"} 50.47 1541429581376
network_connectivity_rtt{host="hop_1"} 3.38 1541429581376
network_connectivity_rtt{host="hop_2"} 11.2 1541429581376
network_connectivity_rtt{host="hop_3"} 20.72 1541429581376
I hope it helps someone else.
Its my bad. I did not included scrape instructions for node-exporter in prometheus.yaml file. It worked after including them.
This issue is happening because of stale metrics.
Lets say you have written you metric in file at 13.00
by default after 5min prometheus will consider you metric stale and it might disappear from there at the time you are making query.
We are collecting the logs of our applications. Since we containerize our applications, the way to collect logs needs a little bit changes.
We log via the Docker Logging Driver:
Application output the logs to container’s stdout and stderr
Using json-file logging driver, docker output logs to json file on
the host machine
Service on the host machine forwards the log files.
But the logs from Docker has additional information which unnecessary and make the forward step complicated because we need to remove those additional information before forward.
For example, the log from Docker is as below, but all we want is the value of log field. Is there a way to customize log format and only output the information wanted by override some Docker's configurations?
{
“log”: "{“level”: “info”,“message”: “data is correct”,“timestamp”: “2017-08-01T11:35:30.375Z”}\r\n",
“stream”: “stdout”,
“time”: “2017-08-03T07: 58: 02.387253289Z”
}
I don't know of any way to customize the output of the json-file docker log plugin. However docker supports the gelf plugin which allows you to send logs to logstash. Using logstash you can output logs in many different ways (by using output plugins) and at the same time customize the format.
For instance to output logs to a file (without any other metadata) you can use something like the following:
output {
file {
path => "/path/to/logfile"
codec => line { format => "%{message}"}
}
}
If you don't want to add complexity to your logging logic, you can keep using the json-file driver and use an utility such as jq to parse the file and extract only the relevant information. For instance with jq you can do: jq -r .log </path/to/logfile>
This will read each line of the specified file as a json object and output only the log field.
My Fluent Bit Docker container is adding a timestamp with the local time to the logs that received via STDIN; otherwise all the logs received via rsyslog or journald seem to have a UTC time format.
I have a basic EFK stack where I am running Fluent Bit containers as remote collectors which are forwarding all the logs to a FluentD central collector, which is pushing everything into Elasticsearch.
I've added a filter to the Fluent Bit config file where I have experimented with many ways to modify the timestamp, to no avail. It seems like I am overthinking it; it should be much easier to modify the timestamp.
These are all the ways I've tried to modify the timestamp with the fluent-bit.conf filter
[FILTER]
Name record_modifier
Match_Regex ^(?!log.*).*$ ## only match the input received via stdin
Tag log.stdout ## tag to mark input received via stdin
Add sourcetype timestamp ## tried to add timestamp from lua script
Parser docker ## tried to use docker parser for timestamp
Time_key utc ## tried to add timestamp as a key
script test.lua ## sample lua script from fluentbit docs
call cb_print ## call a function from within lua script
What is the de facto method to make all the timestamps uniform to UTC? Any help or suggestion is appreciated.
The way it works is that the docker parser extracts the content of 'log' and respect the timestamp defined by docker.
One quick workaround would be to modify your parsers.conf and make sure the docker parser does not resolve the timestamp, on that way Fluent Bit will assign the current time in UTC for you.