4 ms·

Disclosure: I work at GitLab. Have you looked at using our free hosted SaaS GitLab.com with unlimited repos and collaborators? Thanks for using CE!

by wredej 11y ago

tedmiston 11y ago
Thank you! Somehow I completely overlooked that page. Can you elaborate on the monthly uptime in reference to the quote on the GitLab.com page? > Right now GitLab.com is really slow and frequently down.
- cmrx64 11y ago
  IME there's only an hour or so of downtime each month.
- ksec 11y ago
  Like I said to GitLab numerous time, GitLab should split ( or go back to ) their original Gitlab.org and Gitlab.com Their current SaaS is simply not up to standard and it will be hard for others or business to trust and run on it. They should instead run GitlAb SaaS like a proper business.
  - sytse 11y ago
    We're very aware that the current performance of GitLab.com is not acceptable. This is caused by its fast growth and we're working very hard to improve it in the coming months. Yesterday we migrated to all SSD storage, which is an important step.
- frik 11y ago
  https://twitter.com/gitlabstatus https://twitter.com/gitlabstatus on https://status.gitlab.com https://status.gitlab.com mentions Azure cloud several times. "The rsync to *Azure* SSD got interrupted last night when we had an outage. Restarting it now." - 18th Dec "This means service will partially restore over the next few minutes and then go away again once the *Azure* restart command succeeds." "Looks like we are headed for the ‘double restart’ scenario. The NFS server came back on its own but *Azure* is still busy restarting it." "*Azure* restart of the NFS server is in progress" "Depending on whether the machine is stuck or already rebooting. No way to tell with *Azure*." "*Azure* CLI restart of the NFS server finished, that is usually a good sign." "We are experimenting with copying data out of our current *Azure* storage account. Unfortunately the copying affects http://gitlab.com" - 1st Dez "Azure incident quote: Network Infrastructure and Storage - East US 2 - Partial Service Interruption [East US 2]" - 11. Nov "Seems to be a major @azure outage..." - 11th Nov "At 18:00UTC we will start migrating PostgreSQL to Azure Premium Storage gitlab.com/gitlab-com/ope… , expect 15 minutes downtime." - 10. Nov "trying another Azure restart of the NFS server " - 6th Nov "In the last 24h, we had two reboots of our Redis server seemingly caused by Azure (no kernel panics) and VHD read errors on the NFS server" - 5th Nov "Azure are sponsoring us so we are saving a lot on our hosting bill. We are documenting the move on an upcoming blog post." - 16th Oct There have appearently been really big troubles, starting in November! Is Azure cloud that bad or is it a problem on your side? What about moving to AWS/Google/Rackspace?
  - sytse 11y ago
    We moving to metal on Softlayer. See https://gitlab.com/gitlab-com/operations/issues/14 https://gitlab.com/gitlab-com/operations/issues/14
    - frik 11y ago
      Thanks for the info. All the best with the transition to IBM Softlayer.
      - sytse 11y ago
        Thanks!
dominotw 11y ago
Hey I am browsing some of the CI examples and this is the first one I saw https://gitlab.com/ayufan/python-getting-started/builds/179387 https://gitlab.com/ayufan/python-getting-started/builds/1793... I can't seem to tell how to find logs.
- sytse 11y ago
  The logs are normally on this page. In this case there is no log, probably due to an error (this build is 4 months old). For a better example see https://gitlab.com/gitlab-org/gitlab-ce/builds/485077 https://gitlab.com/gitlab-org/gitlab-ce/builds/485077
  - dominotw 11y ago
    Ah ok. It was unclear from the UI that old logs are archived. The example you linked is much better thanks.
    - sytse 11y ago
      You're very welcome.

Disclosure: I work at GitLab. Have you looked at using our free hosted SaaS GitLab.com with unlimited repos and collaborators? Thanks for using CE!

by wredej 11y ago

Disclosure: I work at GitLab. Have you looked at using our free hosted SaaS GitLab.com with unlimited repos and collaborators? Thanks for using CE!

tedmiston 11y ago

Thank you! Somehow I completely overlooked that page. Can you elaborate on the monthly uptime in reference to the quote on the GitLab.com page? > Right now GitLab.com is really slow and frequently down.

cmrx64 11y ago
IME there's only an hour or so of downtime each month.
ksec 11y ago
Like I said to GitLab numerous time, GitLab should split ( or go back to ) their original Gitlab.org and Gitlab.com Their current SaaS is simply not up to standard and it will be hard for others or business to trust and run on it. They should instead run GitlAb SaaS like a proper business.
- sytse 11y ago
  We're very aware that the current performance of GitLab.com is not acceptable. This is caused by its fast growth and we're working very hard to improve it in the coming months. Yesterday we migrated to all SSD storage, which is an important step.
frik 11y ago
https://twitter.com/gitlabstatus https://twitter.com/gitlabstatus on https://status.gitlab.com https://status.gitlab.com mentions Azure cloud several times. "The rsync to *Azure* SSD got interrupted last night when we had an outage. Restarting it now." - 18th Dec "This means service will partially restore over the next few minutes and then go away again once the *Azure* restart command succeeds." "Looks like we are headed for the ‘double restart’ scenario. The NFS server came back on its own but *Azure* is still busy restarting it." "*Azure* restart of the NFS server is in progress" "Depending on whether the machine is stuck or already rebooting. No way to tell with *Azure*." "*Azure* CLI restart of the NFS server finished, that is usually a good sign." "We are experimenting with copying data out of our current *Azure* storage account. Unfortunately the copying affects http://gitlab.com" - 1st Dez "Azure incident quote: Network Infrastructure and Storage - East US 2 - Partial Service Interruption [East US 2]" - 11. Nov "Seems to be a major @azure outage..." - 11th Nov "At 18:00UTC we will start migrating PostgreSQL to Azure Premium Storage gitlab.com/gitlab-com/ope… , expect 15 minutes downtime." - 10. Nov "trying another Azure restart of the NFS server " - 6th Nov "In the last 24h, we had two reboots of our Redis server seemingly caused by Azure (no kernel panics) and VHD read errors on the NFS server" - 5th Nov "Azure are sponsoring us so we are saving a lot on our hosting bill. We are documenting the move on an upcoming blog post." - 16th Oct There have appearently been really big troubles, starting in November! Is Azure cloud that bad or is it a problem on your side? What about moving to AWS/Google/Rackspace?
- sytse 11y ago
  We moving to metal on Softlayer. See https://gitlab.com/gitlab-com/operations/issues/14 https://gitlab.com/gitlab-com/operations/issues/14
  - frik 11y ago
    Thanks for the info. All the best with the transition to IBM Softlayer.
    - sytse 11y ago
      Thanks!

dominotw 11y ago

Hey I am browsing some of the CI examples and this is the first one I saw https://gitlab.com/ayufan/python-getting-started/builds/179387 https://gitlab.com/ayufan/python-getting-started/builds/1793... I can't seem to tell how to find logs.

sytse 11y ago
The logs are normally on this page. In this case there is no log, probably due to an error (this build is 4 months old). For a better example see https://gitlab.com/gitlab-org/gitlab-ce/builds/485077 https://gitlab.com/gitlab-org/gitlab-ce/builds/485077
- dominotw 11y ago
  Ah ok. It was unclear from the UI that old logs are archived. The example you linked is much better thanks.
  - sytse 11y ago
    You're very welcome.