- Set up and test Service Ping locally
- Test Prometheus-based Service Ping
- Service Ping Payload drop
- Generate Service Ping
Troubleshooting Service Ping
Set up and test Service Ping locally
To set up Service Ping locally, you must:
Set up local repositories
- Clone and start GitLab.
- Clone and start Versions Application.
Make sure you run
docker-compose up
to start a PostgreSQL and Redis instance. - Point GitLab to the Versions Application endpoint instead of the default endpoint:
- Open service_ping/submit_service.rb locally and modify
STAGING_BASE_URL
. - Set it to the local Versions Application URL:
http://localhost:3000
.
- Open service_ping/submit_service.rb locally and modify
Test local setup
-
Using the
gitlab
Rails console, manually trigger Service Ping:GitlabServicePingWorker.new.perform('triggered_from_cron' => false)
-
Use the
versions
Rails console to check the Service Ping was successfully received, parsed, and stored in the Versions database:UsageData.last
Test Prometheus-based Service Ping
If the data submitted includes metrics queried from Prometheus you want to inspect and verify, you must:
- Ensure that a Prometheus server is running locally.
- Ensure the respective GitLab components are exporting metrics to the Prometheus server.
If you do not need to test data coming from Prometheus, no further action is necessary. Service Ping should degrade gracefully in the absence of a running Prometheus server.
Three kinds of components may export data to Prometheus, and are included in Service Ping:
-
node_exporter
: Exports node metrics from the host machine. -
gitlab-exporter
: Exports process metrics from various GitLab components. - Other various GitLab services, such as Sidekiq and the Rails server, which export their own metrics.
Test with an Omnibus container
This is the recommended approach to test Prometheus-based Service Ping.
To verify your change, build a new Omnibus image from your code branch using CI/CD, download the image, and run a local container instance:
- From your merge request, select the
qa
stage, then trigger thee2e:package-and-test
job. This job triggers an Omnibus build in a downstream pipeline of theomnibus-gitlab-mirror
project. - In the downstream pipeline, wait for the
gitlab-docker
job to finish. - Open the job logs and locate the full container name including the version. It takes the following form:
registry.gitlab.com/gitlab-org/build/omnibus-gitlab-mirror/gitlab-ee:<VERSION>
. - On your local machine, make sure you are signed in to the GitLab Docker registry. You can find the instructions for this in Authenticate to the GitLab Container Registry.
- Once signed in, download the new image by using
docker pull registry.gitlab.com/gitlab-org/build/omnibus-gitlab-mirror/gitlab-ee:<VERSION>
- For more information about working with and running Omnibus GitLab containers in Docker, refer to GitLab Docker images documentation.
Test with GitLab development toolkits
This is the less recommended approach, because it comes with a number of difficulties when emulating a real GitLab deployment.
The GDK is not set up to run a Prometheus server or node_exporter
alongside other GitLab components. If you would
like to do so, Monitoring the GDK with Prometheus is a good start.
The GCK has limited support for testing Prometheus based Service Ping. By default, it comes with a fully configured Prometheus service that is set up to scrape a number of components. However, it has the following limitations:
- It does not run a
gitlab-exporter
instance, so severalprocess_*
metrics from services such as Gitaly may be missing. - While it runs a
node_exporter
,docker-compose
services emulate hosts, meaning that it usually reports itself as not associated with any of the other running services. That is not how node metrics are reported in a production setup, wherenode_exporter
always runs as a process alongside other GitLab components on any given node. For Service Ping, none of the node data would therefore appear to be associated to any of the services running, because they all appear to be running on different hosts. To alleviate this problem, thenode_exporter
in GCK was arbitrarily “assigned” to theweb
service, meaning only for this servicenode_*
metrics appears in Service Ping.
Service Ping Payload drop
Symptoms
You will be alerted by the Data team and their Monte Carlo alerting.
Locating the problem
First you need to identify at which stage in Service Ping data pipeline the drop is occurring.
Start at Service Ping Health Dashboard on Sisense.
You can use this query as an example, to start detecting when the drop started.
Troubleshoot the GitLab application layer
We conducted an investigation into an unexpected drop in Service ping Payload events volume.
GitLab team members can view more information in this confidential issue:
https://gitlab.com/gitlab-data/analytics/-/issues/11071
Troubleshoot VersionApp layer
Check if the export jobs are successful.
Check Service Ping errors in the Service Ping Health Dashboard.
Troubleshoot Google Storage layer
Check if the files are present in Google Storage.
Troubleshoot the data warehouse layer
Reach out to the Data team to ask about current state of data warehouse. On their handbook page there is a section with contact details.
Cannot disable Service Ping with the configuration file
The method to disable Service Ping with the GitLab configuration file does not work in GitLab versions 9.3.0 to 13.12.3. To disable it, you must use the Admin Area in the GitLab UI instead. For more information, see this issue.
GitLab functionality and application settings cannot override or circumvent restrictions at the network layer. If Service Ping is blocked by your firewall, you are not impacted by this bug.
Check if you are affected
You can check if you were affected by this bug by using the Admin Area or by checking the configuration file of your GitLab instance:
-
Using the Admin Area:
- On the left sidebar, select Search or go to.
- Select Admin Area.
- On the left sidebar, select Settings > Metrics and profiling.
- Expand Usage statistics.
-
Are you able to check or uncheck the checkbox to disable Service Ping?
- If yes, your GitLab instance is not affected by this bug.
- If you can’t check or uncheck the checkbox, you are affected by this bug. See the steps on how to fix this.
-
Checking your GitLab instance configuration file:
To check whether you’re impacted by this bug, check your instance configuration settings. The configuration file in which Service Ping can be disabled depends on your installation and deployment method, but is typically one of the following:
-
/etc/gitlab/gitlab.rb
for Linux package installations and Docker. -
charts.yaml
for GitLab Helm and cloud-native Kubernetes deployments. -
gitlab.yml
for self-compiled installations.
To check the relevant configuration file for strings that indicate whether Service Ping is disabled, you can use
grep
:# Linux package grep "usage_ping_enabled'\] = false" /etc/gitlab/gitlab.rb # Kubernetes charts grep "enableUsagePing: false" values.yaml # From source grep "usage_ping_enabled'\] = false" gitlab/config.yml
If you see any output after running the relevant command, your GitLab instance may be affected by the bug. Otherwise, your instance is not affected.
-
How to fix the “Cannot disable Service Ping” bug
To work around this bug, you have two options:
- Update to GitLab 13.12.4 or newer to fix this bug.
-
If you can’t update to GitLab 13.12.4 or newer, enable Service Ping in the configuration file, then disable Service Ping in the UI. For example, if you’re using the Linux package:
-
Edit
/etc/gitlab/gitlab.rb
:gitlab_rails['usage_ping_enabled'] = true
-
Reconfigure GitLab:
sudo gitlab-ctl reconfigure
- On the left sidebar, select Search or go to.
- Select Admin Area.
- On the left sidebar, select Settings > Metrics and profiling.
- Expand Usage statistics.
- Clear the Enable Service Ping checkbox.
- Select Save Changes.
-
Generate Service Ping
Generate or get the cached Service Ping in rails console
Use the following method in the rails console.
Gitlab::Usage::ServicePingReport.for(output: :all_metrics_values, cached: true)
Generate a fresh new Service Ping
Use the following method in the rails console.
This also refreshes the cached Service Ping displayed in the Admin Area.
Gitlab::Usage::ServicePingReport.for(output: :all_metrics_values)
Generate and print
Generates Service Ping data in JSON format.
gitlab-rake gitlab:usage_data:generate
Generates Service Ping data in YAML format:
gitlab-rake gitlab:usage_data:dump_sql_in_yaml
Generate and send Service Ping
Prints the metrics saved in conversational_development_index_metrics
.
gitlab-rake gitlab:usage_data:generate_and_send