General questions about Managed Service for GitLab
-
How can I migrate data from GitLab to Managed Service for GitLab?
-
How do I update the Managed Service for GitLab instance software?
-
Why cannot I push changes to my Managed Service for GitLab repository?
-
What should I do if I get a HTTP 500 or HTTP 502 error when opening an instance?
-
How do I configure an alert for when disk space usage reaches a certain percentage?
-
Can I change instance type or disk size after the instance has been created?
-
What should I do if I cannot connect to the system hook on
localhost?
What are the advantages of Managed Service for GitLab over a custom GitLab Community Edition installation?
The main advantage of Managed Service for GitLab is lower installation and administration costs for GitLab. In addition, Managed Service for GitLab adds features not found in Community Edition
How can I migrate data from GitLab to Managed Service for GitLab?
You can migrate data from a custom GitLab installation to Managed Service for GitLab. Learn how to do this in this guide. Before you begin, see how to run migration.
You can migrate projects from GitLab.com by means of manual exports and imports.
How do I update the Managed Service for GitLab instance software?
Managed Service for GitLab instances automatically update their GitLab software as its new released versions are adapted to the Yandex Cloud environment. You cannot update the software version on your own.
If you need a sooner update, contact support
- Instance ID.
- Required version of GitLab.
- Desired date and time of the update.
- Reason for the update.
Alert
During this operation, both the Managed Service for GitLab instance under update and any data in it will be inaccessible.
Can I integrate GitLab authentication providers?
Yes, you can. To do this, configure OmniAuth.
Can I use Yandex ID or Yandex 360 for authentication?
Yes, you can. To do this, add a provider of the Yandex ID type to OmniAuth and configure its settings.
Is there GitLab integration with Yandex Tracker?
Yes, there is. For the integration settings, see Integration with Yandex Tracker
Why cannot I push changes to my Managed Service for GitLab repository?
Error messages:
You are not allowed to push code to this project.
You are not allowed to push code to protected branches on this project.
To push changes to a Managed Service for GitLab repository, assign
- To protected branches, e.g.,
master, assign theMaintainerorOwnerrole. - To unprotected branches, assign the
Developer,Maintainer, orOwnerrole.
Users with the Guest and Reporter roles cannot push changes.
Learn more about roles in this GitLab guide
What should I do if I get a HTTP 413 error during git push?
When sending changes over HTTP or HTTPS, you may get this error:
error: RPC failed;
HTTP 413 curl 22
The requested URL returned error: 413
A single git push run cannot send more than 250 MB of data over HTTP or HTTPS. You cannot update nginx settings for an individual instance.
Push the changes via SSH or split them into multiple pushes. For information on how to set up SSH access, see Getting started with Managed Service for GitLab.
What should I do if I get a HTTP 500 or HTTP 502 error when opening an instance?
The instance may run out of disk space. You can increase the instance disk space on your own.
See also Cleaning up full instance disk space.
To track disk usage, set up monitoring and alerts. Container images and build artifacts often take up disk space. To remove them regularly, configure cleanup policies.
How can I clear pipeline logs to free up disk space?
Logs cannot be deleted separately. However, you delete them by deleting irrelevant pipelines.
Where can I monitor my disk space usage?
You can monitor your disk space usage:
- In the management console using the instance state monitoring tools.
- In Yandex Monitoring, where you also can set up alerts for specified metrics.
How do I configure an alert for when disk space usage reaches a certain percentage?
See Setting up alerts in Monitoring for Managed Service for GitLab.
Why are backups not created?
If backup creation fails (the Failed status), set up a dedicated security group and assign it to the GitLab instance.
Can I change instance type or disk size after the instance has been created?
Yes, you can upgrade your instance to a higher-performance type and increase its disk size. However, you cannot reduce instance disk size or downgrade the instance to a lower-performance type. For more information, see Updating instance settings.
What should I do if I cannot connect to the system hook on localhost?
If you cannot connect to the system hook, use the 127.0.0.1 IP address instead of localhost:
- In the system hook settings (Admin area → System Hooks), change the URL value to
http://127.0.0.1:24080/default. - In the GitLab settings that allow sending messages to the local network (Admin area → Settings → Network → Expand outbound requests, the CIDR input field), add
http://127.0.0.1:24080to the list of IP addresses and domain names.
What must I do if I get the EOF fatal error on a worker?
Complete error message:
EOF fatal: early EOF fatal: fetch-pack: invalid index-pack output
You can fix the error only in a runner deployed manually on the VM. Do it by applying these settings:
sysctl -w net.core.rmem_max=26214400
sysctl -w net.core.rmem_default=6250000
sysctl -w net.core.wmem_max=26214400
sysctl -w net.core.wmem_default=6250000
sysctl -w net.ipv4.tcp_rmem='4096 6250000 26214400'
sysctl -w net.ipv4.tcp_wmem='4096 6250000 26214400'
What should I do if my SSH certificate expires?
Usually, the SSL certificate on Managed Service for GitLab instances is renewed automatically. If this does not happen, restart the instance.
If restarting does not resolve the issue, contact support