Thanks Stephane. The pressure on the service is slightly lower since Friday, there is no emergency. My call was for those who have obvious oversized or unused slaves; I donât think Pharo is in this case⦠;-) Thank you anyway. - Bruno
Le 17 nov. 2017 à 22:30, Stephane Ducasse <stepharo.self@gmail.com> a écrit :
Hi Bruno
We will have a look. Now the failures/erasure of the configurations this summer did not really help us. We still not fully recovered. Once Pharo will be managed on git. I think that many of our community jobs will be migrated to travis and other services.
Stef
On Thu, Nov 16, 2017 at 1:02 PM, Bruno Carrez <bruno.carrez@inria.fr> wrote:
Dear CI users,
As some of you may have noticed, the CI service is in capacity troubles. Most VM deployment fail due to a lack of resources.
A lot of VMs use a high-end hardware configuration (XL or XXL), but you must be aware that cloudstack is a mutalised service; so, it is important to use the right configuration for your real needs. It is not acceptable to have unused or oversized slaves on one side, and stucked users on the other side that cannot use the service.
A lot of users are affected by this issue and are unable use the platform at all !
So I highly recommend to check if you can reduce your footprint on the service:
- by downgrading your oversized slaves, - or deleting your unused slaves
If you donât know how to decrease the slave configuration, issue a support ticket and we will be glad to help you.
Thanks in advance, in the name of all the ci users.
- The CI team