Conversation
|
A note about this: Currently if a start interval is not specified, it will get the default start interval of 30s... which is the same as the normal interval. |
|
I'll also note: The only real interest I have in this PR is:
One concern I have for this is the fact that healthcheck configs are in the container |
|
First I'd like to say @cpuguy83 rules. Second are there are any updates on this PR? Any chance for this to be merged in its current form? |
|
Is there any update about this PR? Thank you |
|
@dperny can you check if this requires changes on the SwarmKit side to allow this to be used? |
|
Hey. Is this PR full being considered? |
|
(Needs a rebase, but it appears to be pretty superficial -- just markdown and yaml, so implementation probably doesn't need to change for that 👀) |
9816d6c to
a71a096
Compare
|
I did a quick rebase, and added a second commit with some changes I wanted to suggest; @cpuguy83 PTAL if that last commit looks sane to your (if so I can squash) |
corhere
left a comment
There was a problem hiding this comment.
The logic for selecting between startInterval and probeInterval is inconsistent with the behavior described by the Swagger docs and with how handleProbeResult decides when to start incrementing the failure streak. Which is the intended behavior?
This is useless as this is not an initial delay. We need this [PR](moby/moby#40894), which may land soon...
|
Can we expect this PR to land in 22.06 ? |
|
@bpascard Nope. |
|
Any chances this PR lands in 23.0 ? |
|
@bpascard Nope. 23.0 is the new name for the release formerly known as 22.06. |
|
@cpuguy83 will you have time to rebase and address reviews on this one soon? |
|
😭 Windows
I don't blame Windows too much here, though, since its some custom busybox on Windows. |
This adds an additional interval to be used by healthchecks during the start period. Typically when a container is just starting you want to check if it is ready more quickly than a typical healthcheck might run. Without this users have to balance between running healthchecks to frequently vs taking a very long time to mark a container as healthy for the first time. Signed-off-by: Brian Goff <[email protected]> Signed-off-by: Sebastiaan van Stijn <[email protected]>
d7fba45 to
2216d3c
Compare
|
Updated this to skip the new test on Windows. |
|
All green now. |
|
Swarmkit PR: moby/swarmkit#3142 |
|
Created a tracking issue for all the other changes that need to be completed before the 25.0 release: #45897 |
This adds an additional interval to be used by healthchecks during the
start period.
Typically when a container is just starting you want to check if it is
ready more quickly than a typical healthcheck might run. Without this
users have to balance between running healthchecks to frequently vs
taking a very long time to mark a container as healthy for the first
time.
Replaces #33918
Closes #33410