Where communities thrive

  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
Henri Cook
Hi all. After a GCP auto-upgrade at the weekend our instance of Mattermost was down for about 24 hours. It's back online now but none of my gitlab-mattermost-integration messages are arriving, does anyone know what might be wrong? A stuck message queue somewhere I can jiggle or something? I can definitely talk to mattermost from the Gitlab machine and send webhooks (with curl) - so it doesn't appear to be network level
Anyone can explain how to insert a "service token" with gitlab integration? I cannot paste the token I retrieved from k8s.. What am I doing wrong?
Hassnain Raza Alvi
Can anyone help me in fixing issue with TLS certificates
I setup helm chart according to this and provided certificate chain and private key but still it want me to connect to let's enccrypt. I have installed it on premises so internet access is not allowed from this machine
Can anyone tell me what i am doing wrong here ?
Here are error logs:
Error generating CA certificate chain for Secret "default/gitlab-cert": Get http://cert.int-x3.letsencrypt.org/: dial tcp: lookup cert.int-x3.letsencrypt.org on no such host
Hassnain Raza Alvi
is there anyway of setting internediate certificate myself instead of ingress controlller building it on runtime?
Schroedingers Cat
Hi, since three weeks or so, I'm not receiving a mail when exporting a project from gitlab.com - any idea what's going on?
Yemin Sajid
I am unable to push files to the Gitlab maven registry. It gives HTTP 500 error when it tries to push the package.aar.md5 file. Any clue where I should look into?
1 reply
Schroedingers Cat
Florian Langer
Hi! Does anyone know if it's possible to restrict artifacts downloads from a matrix to another matrix? So not every job in the next matrix downloads all artifacts from the matrix before? Thanks in advance!
Julian Bueno
Hi everybody, I am trying to include my pipeline from another project using include: remote: 'https://gitlab.com/awesome-project/raw/master/.before-script-template.yml' in using gitlab.com and the both repos are private
I cannot export a project from gitlab.com (never receive the mail with the link, also checked the spam) - are the server admins aware of this issue?
Dominic Caudell
Hi. Say one has been hacked, and some confidential source code posted to a public repository, what would be the appropriate channel to have it taken down?
Bartomiej Dr籀髒d髒
Hello all! In a User view in the AdminArea, there is a Created By field. Any idea how can I retrieve this information with RestAPI?
Matthew Lymer
Hello everyone, I'm trying to use the "advanced search" functionality to find a hostname within my various repos, I try the term s3.amazonaws.com (where I know for sure my code has references to it) and I get no search results, however, if I try amazonaws only, I get results (but too many because I want to find only s3 specific ones). Does anyone know why periods breaks my search, I tried the obvious candidates of "s3.amazonaws.com" and s3\.amazonaws\.com but none of the alterations return anything
url: "https://s3.amazonaws.com/some-secret-bucket/some-secret-project_{{ version }}_amd64.deb"
is the example line I'm intending to find
santhakumar g
hi team need some clarification on subscription and purchasing plan and contact supoort for newly started gitlabci accounts
how to contact sales team
Francisco Javier Guzm獺n Jim矇nez

Hi all! I have recently started a Kubernetes cluster and as some of you might now, Gitlab offers 200$ free credit on top of the 300$ Google offers for new users on this. So, I clicked on the Apply for credit button and applied for it but I have just got a refusal because I have not applied with a valid corporate email.

Is this expected? I am building my application which I hope in future becomes a company, otherwise I would not be right now begging free triers, etc. By the look of it it is only for established companies, does anyone know anything about this?

Thank you in advance and regards.

Henrik Christian Grove
@santhakumar_g_gitlab : https://about.gitlab.com/sales/
Hey guys, how do I get support for gitlab? We pushed a semi-large object (ip database) to our repo and now we can't pull updates. On gitlabs side, it seems that they are killing the pull process: error: unpack-objects died of signal 9
Christos Botsikas
Hi I've merged a pull request using squash option in Giltab but for some reason it ended up deleting all files except the one that was created on this PR... Any idea what could have gone wrong?
The original PR had 2 commits, with only 1 change, the new file. The squashed commit has 1 new file created and 1654 files deleted :/
I'm using gitlab.com

Hey all.
Ive attached a k8s cluster to a group in our on-prem gitlab, and deployed, helm and a gitlab runner. I had problems with proxies getting helm etc to work, but managed to get the PodPreset working to inject those settings.
Now however, the runner errors out:

WARNING: Checking for jobs... failed                runner=uagZ6Z7E status=couldn't execute POST against https://gitlab.mgsops.net/api/v4/jobs/request: Post https://gitlab.mgsops.net/api/v4/jobs/request: x509: certificate signed by unknown authority
WARNING: Checking for jobs... failed                runner=uagZ6Z7E status=couldn't execute POST against https://gitlab.mgsops.net/api/v4/jobs/request: Post https://gitlab.mgsops.net/api/v4/jobs/request: x509: certificate signed by unknown authority

when it tries to register with the gitlab server. Because corporate self-signed CA certs. I've deployed runners on docker, and on k8s before, but this way I have no control over the initial invocation or config.toml to inject the required custom CA settings. How?

can I find the pages URL (if it is active) or domains linked to a project using the API somehow?
Hi All, I am starting a new project with three members(Contributors). I am the owner/maintainer of this project and others have developers' role. At this moment, we are directly committing our code in the master branch. I found that the last commit and the respective files are only viewable in GitLab project. It is NOT showing all the files from everyone's commit. How to solve this problem so that everyone can see the files from others also?
Anyone please help me on this issue.
In AutoDevOps - if I commit a dotnet core project - it is successfully built because the Dockerfile is discovered. But then the test and code_quality jobs fail - terminating the whole thing - because there is no herokuish buildback found.
OK, so I can disable that by setting TEST_DISABLED to 1 in variables
But now, we get to the staging deployment:
 $ auto-deploy deploy
70 secret/staging-secret replaced
71 No helm values file found at '.gitlab/auto-deploy-values.yaml'
72 Deploying new stable release...
73 Release "staging" does not exist. Installing it now.
74 Error: validation failed: unable to recognize "": no matches for kind "Deployment" in version "extensions/v1beta1"
78 ERROR: Job failed: command terminated with exit code 1
not sure what I am looking at here. anyone got any ideas?
Bad gateway on gitlab.com? Going on for several minutes, now.
Edit: Back, but extremely slow page build up.
Bad gateway is back.
Seems to be back.
Schmitt Christian
hello, I'm trying to figure out the metrics dashboard and with our "manual" deployment I could've already gotten the nginx-ingress dashboard to work, unfortunatly I could not get system metrics to work, is there any way to debug that? I.e. we created a custom helm chart that uses the CI_ENVIRONMENT_SLUG, however it always prints "No data to display", but "Response metrics (NGINX Ingress)" does show stuff..
Schmitt Christian
would be cool if somebody could give me some hints on that
AutoDevops triggers build fails at staging. I set the CI variable TRACE to 1 so that auto-deploy would log its output, and got this:
 + echo 'Deploying new stable release...'
2302 + helm upgrade --install --wait --set service.enabled=true --set gitlab.app=examples-auto-devops-dotnet-api-demo --set gitlab.env=staging --set gitlab.envName=staging --set gitlab.envURL=http://examples-auto-devops-dotnet-api-demo-staging.k3.ops.net --set releaseOverride=staging --set image.repository=registry.gitlab.ops.net/examples/auto-devops-dotnet-api-demo/master --set image.tag=2019c8ab51e45999cd18cc5013d6a66ef9ae906e --set 'image.secrets[0].name=gitlab-registry-examples-auto-devops-dotnet-api-demo' --set application.track=stable --set application.database_url=postgres://user:testing-password@staging-postgres:5432/staging --set application.secretName=staging-secret --set application.secretChecksum=e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 --set service.commonName=le-10631.k3.ops.net --set service.url=http:/examples-auto-devops-dotnet-api-demo-staging.k3.ops.net --set service.additionalHosts= --set replicaCount=1 --set postgresql.enabled=true --set postgresql.managed= --set postgresql.managedClassSelector= --set postgresql.nameOverride=postgres --set postgresql.postgresUser=user --set postgresql.postgresPassword=testing-password --set postgresql.postgresDatabase=staging --set postgresql.imageTag=9.6.2 --set application.initializeCommand= --set application.migrateCommand= --namespace=auto-devops-dotnet-api-demo-10631-staging staging chart/
2303 Release "staging" does not exist. Installing it now.
2304 Error: validation failed: unable to recognize "": no matches for kind "Deployment" in version "extensions/v1beta1"

Hi guys,
Is self-hosting gitlab CE possible on windows? I have been browsing around but can't find any information regarding this subject.

Thanks in advance,

@JeremyThemans I recall reading in the gitlab omnibus docs that hosting on windows was not supported. I'm not sure if that was because i was reading the docker specific docs at the time.
Thanks. I will have another look and hopefully find some information.
Ok I found it.
For anyone with the same question: https://docs.gitlab.com/ee/install/requirements.html#microsoft-windows
TL;DR it doesn't support windows.
On the subject of runners on kubernetes :- how do I get to the logs of the runner-helpers?
When I look in my gitlab-managed-apps namespace all I see is the runner-gitlab-runner-somehash pod, al matching deployment and replicaset. Where are the ephemeral runner-helper pods/logs to be found?
M獺t矇 T籀th

Hello, I encountered a problem
using graphiQL of my gitlab (https://git.a38.hu/-/graphql-explorer) I get the following error message:
"message": "Field 'projects' doesn't accept argument 'ids'"

but it should be accepting ids as filters, and it does work on other gitlab servers (https://gitlab.com/-/graphql-explorer, https://gitlab.techpm.guru/-/graphql-explorer)

can anybody please help with this?

Peter Anker

We are running our builds on GitLab Shared Runners with gitlab-org-docker builds, but builds are now getting stuck. We have made sure there are enough "build minutes" purchased. Is this an issue with GitLab or an issue with our account? Several of our repos and pipelines seem affected.

Thor Larholm
All the other runners out of the 15 from Gitlab are still online, it's specifically those with the gitlab-org-docker tag
shared runners offline
OMFG. How much babysitting does AutoDevOps need to actually do some kind of simple POC?

So i've run a latest gitlab omnibus ce on docker (Swarm). Ive added an instance level k3s cluster. Ive installed the gitlab runner and prometheus applications via the integration. I created a test project "demo/autodevops-demo" which is a dotnetcore app that I added a Dockerfile to. I then turned on the registry and restarted gitlab because "build" was erroring out as CI_REGISTRY_IMAGE wasn't defined and no part of the AutoDevOps scripts botheres to check. I then had to add the variable TEST_DISABLED=1 as a CI/CD variable, because despite the documents saying you can create projects with Dockerfiles, AutoDevOps always triggers a test step which needs a herokuish buildpack. I then added CODE_QUALITY_DISABLED=1 because that container stalls, well, for at least 30minutes and Im not sure why. So it finally builds, pushes the dockerfile to the registry, and with the default AutoDevOps strategy (CD to production) ends up failing at step 3: production - with this error:

+ kubectl get namespace auto-devops-demo-2-production
Unable to connect to the server: authenticationrequired
+ kubectl create namespace auto-devops-demo-2-production
Unable to connect to the server: authenticationrequired
Uploading artifacts for failed job
Uploading artifacts...
WARNING: environment_url.txt: no matching files    
WARNING: tiller.log: no matching files             
ERROR: No files to upload                          
Cleaning up file based variables
ERROR: Job failed: command terminated with exit code 1

ideas? what /how does it need to authenticate with the managed k8s cluster and fail?

Henrik Christian Grove
Is it possible to get a list of packages to migrate before configuring object storage?