Hi @kmon,
The brownout should be over, we had an issue with our internal tooling related to this brownout which is being address at present.
Thanks,
Brent
Hi @kmon,
The brownout should be over, we had an issue with our internal tooling related to this brownout which is being address at present.
Thanks,
Brent
Hi @Neeraj,
It looks like you are upgrading from ubuntu-2004 to ubuntu-2204 I might suggest upgrading to a newer version of ubuntu-2004, ubuntu-2004:current
and seeing if that works before upgrading to 2204.
Thanks,
Brent
Hi Team,
Still same error after updating to above image .
Here the link to build : pipelines/bitbucket/labdex/diaceutics-sbo-test-automation-framework/6246/workflows/72af180b-de76-4417-bccb-32b51f4d873b/jobs/10340
Is the brownout over for today and (according to the original post) not scheduled again until May?
There is one more window today at 20:00 UTC â 00:00 UTC.
Hi @Neeraj,
Do you mind submitting a support ticket for this and one of our internal folks will try and help you.
Thanks,
Brent
Where to submit the support ticket,link please ?
As some others have mentioned, brownouts are not a very good way to get peoplesâ attention, please use other strategies in the future (banners, emails, loud warnings in logs, etc.). Our team was unable to deploy code to production for a while as we scrambled to get a fix in place for this outage. None of us got any emails about the brownouts (or even the deprecation). We had seen the warnings about images being deprecated in the UI, but delayed upgrading as we thought the deprecation was September 30 and it wasnât made obvious that there were going to be brownouts.
Brownout?!? Willingly terminating our ability to push code to production?!? Clearly you do not value the nature of your vendor relationship with your customers.
More than not valuing their users time, they are deliberately wasting time and potentially causing and worsening production issues by purposefully breaking CI pipelines. My team was not able to deploy to production for a FULL day, and that was while we were having production issues.
As others here mentioned, a warning inside a job is completely invisible.
Terrible choice.
Please release the AWS ECR orb.
Any users of this orb cannot run pipelines due to a reference in the current version to a browned-out image. CircleCI owns the code and release of this orb, so itâs pretty rough that even first-party orbs are being hit by this.
The ORB code is posted to github with a link on the ORBâs main page
The image used by the ORB can be changed by using the executor parameter of build_and_push_image which allows the default executor to be changed
The latest version of the orb - 9.0.2 should allow you to not use a deprecated version.
My team sets versions of node but we donât specify linux versions, how can we set the linux version without making changes to what version of node were running? Trying to set the machine to current
, default
or a specific version, seems to always get overridden with what node is pulling in so I always get the same error image 'ubuntu-2004:2022.04.1' is unavailable
Hi @michaeldgoodrum,
I am guessing you might be using an orb which is overriding the machine image here. I would suggest looking at which orbs you are using and see if they are overriding this. If you are still having problems, I would suggest opening a support ticket and one of our internal folks can help you.
Thanks,
Brent
Hi Brent,
9.0.2 does not have the fix commit, we are stuck waiting for a new version. The commit was only added three days ago and 9.0.2 was released in January.
Thanks,
Andy
Can you detail in a new post how you are using the node support? There are 2 options that I am aware of
The problem is that the node based images are built based on which release of node you want to use, rather than a named release of Ubuntu, so old node versions use old Ubuntu versions.
The next problem is that both solutions use CircleCIâs OS docker images. If the cimg/node container was built using a Ubuntu image that has been withdrawn due to this Brownout it means that the cimg/node container can not be loaded as it will be trying to pull in the Ubuntu image during the deployment process.
Please note, I am not a staff member so I can only speculate on issues like this. I believe that the staff members are mainly focused on resolving issues caused by this test and I have no more information than what has been posted in the forum and on the status page.
Hi @Andy-Bell,
The latest change only updates the pipeline for circleci publishing of the orb. I will publish a new version but it is not needed.
Thanks,
Brent