Hi,
I know you will be fully aware that quey.io is currently unavailable, this I also assume is a very rare event.
However, they have now been down for 4 hours.
Are there plans to host images via multiple providers?
As far as I can tell Kubernetes does not support backup images locations, I will be raising a suggestion there too.
Thank you
Guy
Since the latest version a mirror is being pushed to GCR https://github.com/kubernetes/ingress-nginx/issues/5573
You can pull the latest from:
gcr.io/k8s-staging-ingress-nginx/nginx-ingress-controller:0.32.0
given Quay.io terrible track record lately, wouldn't it make more sense to just use the gcr.io registry as the default one instead?
Closing. As @robertvanhoesel commented, we have a temporal copy of the image in gcr now.
@jemag No. The project has been using quay.io for four and a half years and this is the second important outage.
I have been working since the last quay.io outage to start pushing to gcr.io as a backup.
The plan is to have all the automation needed for this for the next release.
For some context about this work:
Also, it is important to mention that this can happen to any of the available options.
Never it is a good idea to overreact after an outage.
I appreciate the backup image and I do understand that you have a long working relationship with them.
Perhaps it was overreaction on my part, however 3 outage in a single month (2 of them being multiple hours) seemed a bit concerning. This is not something I have witnessed with other providers.
Either way, it is merely a suggestion. Thank you for providing the context about your work
I fully agree not to overreact, Using most big providers will give a high 9 count, two an even higher count. There will never be a perfect solution.
Thank you for your answers and keep up the great work.