AWS: Change master to m3.medium by default #21160
Conversation
|
Labelling this PR as size/S |
k8s-github-robot
added
the
size/S
|
GCE e2e test build/test passed for commit 129c4243ce7398f840bc5dcda1c194162bf0f4a4. |
|
@justinsb LGTM, modulo docs. Can you add the doc update to this PR? |
m3.large for > 150 nodes. t2.micro often runs out of memory. The t2 class has very difficult-to-understand behaviour when it runs out of CPU. The m3.medium is reasonably affordable, and avoids these problems. Fix kubernetes#21151 Issue kubernetes#18975
justinsb
force-pushed
the
aws_bigger_master
branch
from
129c424 to
a16372c
Compare
justinsb
changed the title
|
Added docs.... possibly too many words, but there's just so much to say :-) . |
|
Labelling this PR as size/M |
k8s-github-robot
added
size/M
|
GCE e2e test build/test passed for commit a16372c. |
|
We have had some reports that using a t2 is actually OK - that we had a bug that was the root of the problems here (e.g. #18975 (comment)). I still think we are better to recommend a configuration with more breathing room, and let people that want to save money knowingly & deliberately opt-in to the smaller configuration. @brendandburns ? |
|
Moving to v1.2 so we make a decision either way... |
brendandburns
added
lgtm
|
@justinsb sgtm. lgtm. |
|
@k8s-bot test this Tests are more than 48 hours old. Re-running tests. |
|
GCE e2e build/test failed for commit a16372c. |
|
Is it OK to mark this e2e-not-required? Feels a bit like cheating, but it might at least free up some e2e bandwidth, and it isn't like e2e is testing this... |
|
Automatic merge from submit-queue |
Auto commit by PR queue bot
m3.large for > 150 nodes.
t2.micro often runs out of memory. The t2 class has very
difficult-to-understand behaviour when it runs out of CPU. The
m3.medium is reasonably affordable, and avoids these problems.
WIP because docs
Fix #21151
Issue #18975