tomcz / openldap_exporter Goto Github PK

Would you be willing to publish your release to dockerhub or similar?
Instructions are here https://docs.github.com/en/actions/publishing-packages/publishing-docker-images

I am happy to help, but it relies on secrets and items out of the scope of submitting PRs
Alternatively, I'm happy for fork and publish an image.

Hello,

thank you very much for your exporter! I finally installed it on my primary OpenLDAP-server and have integrated it into Prometheus. Does somebody have a working Grafana dashboard and would like to add it here to a contrib folder? That would be really nice.

Thanks in advance.

Hi

Our company can only allow us to install golang package available in pkg.go.dev. I want to use this openldap exporter but i found the only available version in pkg.go.dev is [v0.0.0-20230326010953-24ad55465feb].

REF: https://pkg.go.dev/github.com/tomcz/openldap_exporter

Could you upload the latest stable version (v.2.2.2) to pkg.go.dev?
Thanks in advance.

May it be possible to implement TLS connection to OpenLdap server?

https://prometheus.io/docs/instrumenting/writing_exporters/#scheduling exposing metrics using internal scheduling is anti-pattern in prometheus, if we collect metrics when each scrape came in , these are the advantages:

1. the default up metrics can be used to indicate whether the scrape is a success or a failure.
2. the scrape frequency can be controlled by prometheus rather than the prometheus and exporter
3. no hidden outdated data, if the exporter scrape data in backgroud , there is a chance that prometheus get outdated data, the metrics would

there might be disadvantage: if collection of metrics is expensive, like it takes a minute or more to complete, this could be a problem, in which case, a pushgateway may be more suitable.

Hello,

$ ldapsearch -LLL -x -H ldaps://xxxx -b dc=datac,dc=xxx,dc=com -D uid=xxxxx -w xxxx -s base contextCSN
dn: dc=datac,dc=xxx,dc=com
contextCSN: 20230104171052.895118Z#000000#001#000000
contextCSN: 20230106202836.142061Z#000000#00b#000000

I'm running with -replicationObject "dc=datac,dc=xxx,dc=com" parameter

And I'm getting this output from exporter:

$ curl -s http://localhost:9330/metrics  | grep -v \# | grep replication
openldap_monitor_replication{id="001",type="count"} 0
openldap_monitor_replication{id="001",type="gt"} 1.672855957e+09
openldap_monitor_replication{id="001",type="mod"} 0

The questions are:

1-) I'm interested to get information about #00b# and not #001# is this possible?

2-) Is this graph the expected result?

expr sum(openldap_monitor_replication) by (instance, dn)>0

3-) The "count" or "mod" supposed to increase somehow?

4-) What the "gt" metric means?

Thanks

1. In order to monitor the availability of ldap service , please add latency indicator for querying ldap.
   2.when I run openldap_exporter , there is a warning information, I don't know how to set filter in config.yaml , I find the objectClass=monitorOperation is a const , may I set it in config.yaml with variable?

WARN[0120] query failed component=scraper error="LDAP Result Code 1 "Operations Error": 000020D6: SvcErr: DSID-03100836, problem 5012 (DIR_ERROR), data 0\n\x00" filter="(objectClass=monitorOperation)"

Hello,
How --replicationObject is supposed to work ?

I made some try to monitor "dc=domain,dc=local" and I have the following error message
WARN[0030] query failed component=scraper error="LDAP Result Code 4 "Size Limit Exceeded": " filter="(objectClass=*)"

I understand the errror, but not the expected syntax

Best regards

Good morning,

Would you please consider to add support for web-config.yml?
https://github.com/prometheus/prometheus/blob/main/documentation/examples/web-config.yml

That would also enable basic auth and tls

Here is how it was done in another exporter: hikhvar/mqtt2prometheus@a25cf3f

Thank you !

Hi @tomcz,

I was taking a look at your exporter, and I am really interested in using it.
There is some issues I wanted to clarify before with you.

1. Why do you want to set the query interval at the exporter side?

  -interval duration
        Scrape interval (default 30s)

Prometheus set this at the server-side, otherwise you have one LDAP scrape time locally, and one different by every prometheus scrape. I mean, you should be querying LDAP simultaneously when prometheus scrapes the exporter target.

2. Why did you put your code under ~/code/src/app instead of '~/'?

If you agree with those changes, I can start making some PR.

Thanks!

go install honnef.co/go/tools/cmd/staticcheck@latest
# honnef.co/go/tools/unused
/go/pkg/mod/honnef.co/go/[email protected]/unused/unused.go:419:14: obj.Origin undefined (type *types.Var has no field or method Origin)
/go/pkg/mod/honnef.co/go/[email protected]/unused/unused.go:421:14: obj.Origin undefined (type *types.Func has no field or method Origin)
note: module requires Go 1.19
# honnef.co/go/tools/staticcheck
/go/pkg/mod/honnef.co/go/[email protected]/staticcheck/lint.go:4025:29: righti.Method(i).Origin undefined (type *types.Func has no field or method Origin)
/go/pkg/mod/honnef.co/go/[email protected]/staticcheck/lint.go:4030:36: sel.Obj().(*types.Func).Origin undefined (type *types.Func has no field or method Origin)
note: module requires Go 1.19
make: *** [Makefile:33: lint] Error 2

go install honnef.co/go/tools/cmd/[email protected]
works

Hi!, i'm trying to get metrics about ldap replication, i use:

/usr/bin/openldap_exporter --config exporter.yaml --replicationObject "dc=my,dc=organization"

But my ldap DB it's quite big and i get:

WARN[0270] query failed                                  component=scraper error="LDAP Result Code 4 \"Size Limit Exceeded\": " filter="(objectClass=*)"

Looking into the code i found this:

func (s *Scraper) addReplicationQueries() {
	for _, q := range s.Sync {
		queries = append(queries,
			&query{
				baseDN:       q,
				searchFilter: objectClass("*"),
				searchAttr:   monitorReplicationFilter,
				metric:       monitorReplicationGauge,
				setData:      setReplicationValue,
			},
		)
	}
}

If i change:

searchFilter: objectClass("*"),

with

searchFilter: "(contextCSN=*)",

it works, so i'm trying to make a fork adding some parameters to override that but i don't know much about go.

So the question is: i'm using this exporter right?, or there is any other way to achieve this?

Release v2.1 adds support for unix socket connections, but this is not reflected in the documentation.