Skip to content

Commit 2b3c163

Browse files
committed
fix(e2e): size tenant workers by resources, not by u1.large
Sizing the workers with u1.large doubled their memory along with the vCPUs, and four 8Gi workers do not fit on the sandbox beside the rest of the suite: the second worker of each cluster stayed Pending with Insufficient memory on all three nodes, so the node group could never reach two Ready nodes and node-join failed for a reason that has nothing to do with the CPU ceiling it was meant to test. Setting resources.cpu and resources.memory keeps the two vCPUs at the memory the workers had before. The chart drops the instancetype from the VM when a group sets both, and the values schema still requires the field, so it stays in the values as the sizing fallback it now is. Assisted-By: Claude <[email protected]> Signed-off-by: Aleksei Sviridkin <[email protected]>
1 parent 71c0fc1 commit 2b3c163

1 file changed

Lines changed: 12 additions & 2 deletions

File tree

‎hack/e2e-chainsaw/_lib/run-kubernetes.sh‎

Lines changed: 12 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2651,7 +2651,10 @@ ${ouroboros_addon}
26512651
md0:
26522652
diskSize: 20Gi
26532653
gpus: []
2654-
instanceType: u1.large
2654+
# Sizing comes from resources below; this is here because the values
2655+
# schema requires the field, and the chart drops the instancetype from
2656+
# the VM whenever a group sets both resources.cpu and resources.memory.
2657+
instanceType: u1.medium
26552658
# The failure this suite keeps hitting is a worker that stalls in the
26562659
# guest before Talos apid answers, which leaves nothing to read on the
26572660
# management side and nothing for talosctl to connect to. Turning this on
@@ -2660,7 +2663,14 @@ ${ouroboros_addon}
26602663
logSerialConsole: true
26612664
maxReplicas: 10
26622665
minReplicas: 2
2663-
resources: {}
2666+
# Two vCPUs at the memory the workers had before. Sized by resources
2667+
# rather than by instanceType: u1.large doubles memory along with the
2668+
# vCPUs, and four 8Gi workers do not fit beside the rest of the suite,
2669+
# so the second worker of each cluster stayed Pending on Insufficient
2670+
# memory and the node group never reached two Ready nodes.
2671+
resources:
2672+
cpu: 2
2673+
memory: 4Gi
26642674
roles:
26652675
- ingress-nginx
26662676
storageClass: replicated

0 commit comments

Comments
 (0)