libvirt: bump memory of machines to 16G in the provider config#5230
libvirt: bump memory of machines to 16G in the provider config#5230Prashanth684 wants to merge 1 commit intoopenshift:masterfrom
Conversation
While the terraform defaults for the libvirt master memory size was set to 16G through this PR:openshift#5069, the sizes were still fixed to 8G because they were being overwritten by the value set in the provider. Change the provider value to 16G to address this issue. Note: This will also set the worker memory size to 16G which should be fine, but can also be overriden by changing the machineset manifest.
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
@praveenkumar this will also set the worker memory to 16G , is that ok? or do we want to keep that at 8G? |
|
/retest |
1 similar comment
|
/retest |
| Kind: "LibvirtMachineProviderConfig", | ||
| }, | ||
| DomainMemory: 8192, | ||
| DomainMemory: 16384, |
There was a problem hiding this comment.
@Prashanth684 we are using https://cloud.google.com/compute/docs/general-purpose-machines#n1_machines (n1-standard-16) and if we are increased the resource that much it means we are not able to accommodate cluster in this machine (I think standard e2e-libvirt runs 3master-3worker mode).
There was a problem hiding this comment.
hmm..ok yeah n1-standard-15 has only 60G..so should we reduce worker nodes to 2 and keep it at 8192 and increase master memory to 12G then ?
There was a problem hiding this comment.
@Prashanth684 I will take a look tomorrow how it all link together and update you here, first we need to manual create a gcp instance (same resource) and try to start the cluster and see if it able to execute test as expected.
There was a problem hiding this comment.
Did you get a chance to take a look at this?
There was a problem hiding this comment.
i did kick off a libvirt e2e and as I am monitoring, i see that Praveen is right. the standard machines have 64G memory which is not sufficient for this change. But even otherwise, i think we need to look into bumping the gcp instance type to something beefier because the libvirt master defaults need to be 16G anyway for consistent CI runs
There was a problem hiding this comment.
@Prashanth684 the gcp request you mentioned should be accomplished in #5841
|
/retest |
|
/test e2e-libvirt |
|
@Prashanth684: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
|
@Prashanth684 It appears that we are not pursuing this change anymore. If that is the case, could you please close this PR? |
5 participants
While the terraform defaults for the libvirt master memory size was set to 16G through this PR:#5069, the sizes were still fixed to 8G because they were being overwritten by the value set in the provider. Change the provider value to 16G to address this issue.
Note: This will also set the worker memory size to 16G which should be fine, but can also be overriden by changing the machineset manifest.