Stabilize describer outputs #25251

timoreimann · 2016-05-06T08:33:16Z

All describers should exhibit a stable output behavior; that is, the output should only change if the state of the underlying object changes. Specifically, we should make sure that all iterations over maps (which Go randomizes) are always carried out in a stable, deterministic fashion.

TODO list (and related object(s)):

(Original issue descriptions:

kubectl describe pod <some pod> enumerates container resources for various aspects, namely CPU and memory for QoS tiers, limits, and requests at this moment. When invoking the command multiple times, one can see that the order is not stable: Sometimes, CPU is listed before memory, and other times it's just vice versa. It's not stable across resource types either (i.e., QoS tiers may show CPU first while limits doesn't).

For certain use cases (such as when passing the command to watch -1), a stable order would be helpful to minimize distraction from such "irrelevant" output changes.

Happy to work on a PR if folks agree.)

The text was updated successfully, but these errors were encountered:

roberthbailey · 2016-05-08T07:00:33Z

/cc @jlowdermilk

bgrant0607 · 2016-05-09T16:46:57Z

This is due to random map order. Agree that stable order would be useful.
cc @kubernetes/kubectl

smarterclayton · 2016-05-09T16:50:49Z

Agreed - all describers should be stable order and we should probably make sure we educate reviewers to look for that.

bgrant0607 · 2016-05-09T17:05:06Z

Among other things, all maps need to be sorted.

timoreimann · 2016-05-11T14:31:58Z

Unless someone has already started to work on this, I'd like to take a stab.

sylwekb · 2016-05-20T14:57:02Z

@timoreimann Have you started working on this? If not I would like to take it over

timoreimann · 2016-05-20T16:24:12Z

@Gitfred I did actually. I plan to push out a PR over the weekend.

…ectl-describe Automatic merge from submit-queue Stabilize map order in kubectl describe Refs #25251. Add `SortedResourceNames()` methods to map type aliases in order to achieve stable output order for `kubectl` descriptors. This affects QoS classes, resource limits, and resource requests. A few remarks: 1. I couldn't find map usages for described fields other than the ones mentioned above. Then again, I failed to identify those programmatically/systematically. Pointers given, I'd be happy to cover any gaps within this PR or along additional ones. 1. It's somewhat difficult to deterministically test a function that brings reliable ordering to Go maps due to its randomizing nature. None of the possibilities I came up with (rely a "probabilistic testing" against repeatedly created maps, add complexity through additional interfaces) seemed very appealing to me, so I went with testing my `sort.Interface` implementation and the changed logic in `kubectl.describeContainers()`. 1. It's apparently not possible to implement a single function that sorts any map's keys generically in Go without producing lots of boilerplate: a `map[<key type>]interface{}` is different from any other map type and thus requires explicit iteration on the caller site to convert back and forth. Unfortunately, this makes it hard to completely avoid code/test duplication. Please let me know what you think.

timoreimann · 2016-05-29T12:12:43Z

@roberthbailey @bgrant0607 @smarterclayton @jlowdermilk @derekwaynecarr fixed order output for pod descriptions. Do you know of any other describers that use maps as well? A quick glance on my end didn't yield anything but maybe I missed some.

bgrant0607 · 2016-05-31T15:13:59Z

@timoreimann Maps in the API are: labels, annotations, resources, selectors, Secret and ConfigMap data, Flex volume source options

timoreimann · 2016-05-31T15:37:17Z

@bgrant0607 thanks. I'll go over the list and see what still needs to be done. Please let me know if you have any preference towards either tracking everything as part of this issue or a new one (that I can create) instead.

derekwaynecarr · 2016-05-31T20:31:14Z

I think this one issue is fine to track. The current work handled resources. If you want to put a task list on the issue to track that is fine too.

timoreimann · 2016-06-01T00:26:42Z

@bgrant0607 @derekwaynecarr I updated the issue title and description to better reflect the broader scope of the ticket.

Will continue to work on the individual bits and pieces.

errordeveloper · 2016-11-28T22:24:40Z

I fixed two bugs that relate to this (#32229, #35112), I think we need a better contract.

I am looking at pkg/kubectl/describe.go and, and I feel like it can be made more elegant with good use of text/template, or maybe be just a few extra structs could work better. It's just a massive set of special cases right now, with formatting characters everywhere (which seem fairly consistent, but you can never tell 100%).

To illustrate, the structs could look something like these:

type Section struct {
     Name string
}

type LineDescriptions struct {
     Tabs int32
     Data map[string]string
}

type Table struct {
     Head []string
     Data [][]string
}

Let's also consider more convertable output formats, I am thinking of markdown or something similar (of course we don't want YAML, we already have that), it'd help a lot for folks who want to work on IDE integrations and other similar things (I've started more specific discussion in #37576).

smarterclayton · 2016-11-28T22:29:05Z

Since we're going to move this to the server side, it would be better not to dig into this until we settle on that design.

…

On Mon, Nov 28, 2016 at 5:24 PM, Ilya Dmitrichenko ***@***.*** > wrote: I fixed two bugs that relate to this (#32229 <#32229>, #35112 <#35112>), I think we need a better contract. I am looking at pkg/kubectl/describe.go and, and I feel like it can be made more elegant with good use of text/template, or maybe be just a few extra structs could work better. It's just a massive set of special cases right now, with formatting characters everywhere (which seem fairly consistent, but you can never tell 100%). To illustrate, the structs could look something like these: type Section struct { Name string } type LineDescriptions struct { Tabs int32 Data map[string]string } type Table struct { Head []string Data [][]string } Let's also consider more convertable output formats, I am thinking of markdown or something similar (of course we don't want YAML, we already have that), it'd help a lot for folks who want to work on IDE integrations and other similar things (I've started more specific discussion in #37576 <#37576>). — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#25251 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABG_pxTkJiRQ2S2aIJp-FrR6GFhvDnoCks5rC1S3gaJpZM4IYr4_> .

adohe-zz · 2016-11-30T12:59:18Z

@smarterclayton you mean this proposal #29472?

smarterclayton · 2016-11-30T16:43:07Z

That was the rough draft yes - it needs more detail on describe.

…

On Wed, Nov 30, 2016 at 7:59 AM, TonyAdo ***@***.***> wrote: @smarterclayton <https://github.com/smarterclayton> you mean this proposal #29472 <#29472>? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#25251 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABG_p2OXfoRyVIkYCAI1Y7WQTylK-U8lks5rDXMsgaJpZM4IYr4_> .

fejta-bot · 2017-12-19T01:17:39Z

Issues go stale after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

Prevent issues from auto-closing with an /lifecycle frozen comment.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or @fejta.
/lifecycle stale

timoreimann · 2017-12-19T16:25:25Z

Just picked up this issue again.

@smarterclayton is it okay to continue working on the describer output? or is that move to the server side still ongoing?

smarterclayton · 2017-12-19T21:05:05Z

We’re going to discuss it again the first week in January. Probably best to wait until then

fejta-bot · 2019-02-22T01:54:18Z

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

timoreimann · 2019-03-07T00:44:53Z

/remove-lifecycle rotten

fejta-bot · 2019-06-05T00:56:15Z

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

fejta-bot · 2019-07-05T01:43:10Z

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

errordeveloper · 2019-07-06T08:35:45Z

/remove-lifecycle rotten

…

On Fri, 5 Jul 2019, 2:43 am fejta-bot, ***@***.***> wrote: Stale issues rot after 30d of inactivity. Mark the issue as fresh with /remove-lifecycle rotten. Rotten issues close after an additional 30d of inactivity. If this issue is safe to close now please do so with /close. Send feedback to sig-testing, kubernetes/test-infra and/or fejta <https://github.com/fejta>. /lifecycle rotten — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#25251?email_source=notifications&email_token=AAB5MS3QW4XDBCPUHW5RFQLP52RLNA5CNFSM4CDCXY72YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZIKULQ#issuecomment-508602926>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAB5MS3RJSE2I6OFHPVPOELP52RLNANCNFSM4CDCXY7Q> .

fejta-bot · 2019-10-04T08:38:35Z

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

McCoyAle · 2019-10-11T00:56:52Z

Hey, @timoreimann @errordeveloper This one has been open for some time. Is there anything that is needed?

timoreimann · 2019-10-13T20:31:25Z

@McCoyAle tbh, I haven't worked on this one so long that I'm not sure whether it's still relevant. Do you possibly know? If not and you happen to have a moment to verify whether output ordering is still an issue, I'd love to hear about it.

timoreimann · 2019-10-13T20:31:36Z

for now:

/remove-lifecycle rotten

timoreimann · 2019-10-13T20:31:47Z

/remove-lifecycle rotten

timoreimann · 2019-10-13T20:32:05Z

/remove-lifecycle stale

fejta-bot · 2020-01-11T21:32:01Z

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

fejta-bot · 2020-02-10T22:13:48Z

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

fejta-bot · 2020-03-11T22:57:09Z

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

k8s-ci-robot · 2020-03-11T22:57:17Z

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

Instructions 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.

Bug 1854312: 4.4: UPSTREAM: 92014: CSI: Modify VolumeAttachment check to use Informer/cache Origin-commit: 65d455ba14320a3e1270c2d304a0007de3045177

roberthbailey added sig/api-machinery Categorizes an issue or PR as relevant to SIG API Machinery. team/ux labels May 8, 2016

bgrant0607 added priority/backlog Higher priority than priority/awaiting-more-evidence. area/kubectl help-wanted labels May 9, 2016

timoreimann mentioned this issue May 22, 2016

Stabilize map order in kubectl describe #26046

Merged

timoreimann changed the title ~~Stabilize resource order in 'kubectl describe pod' output~~ Stabilize describer outputs Jun 1, 2016

bgrant0607 removed the help-wanted label Aug 30, 2016

fabianofranz mentioned this issue Dec 14, 2016

kubectl umbrella issue for printers and describers #38768

Closed

60 tasks

k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Dec 19, 2017

k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Jan 23, 2019

k8s-ci-robot added lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. and removed lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. labels Feb 22, 2019

k8s-ci-robot removed the lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. label Mar 7, 2019

k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Jun 5, 2019

k8s-ci-robot added lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. and removed lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. labels Jul 5, 2019

k8s-ci-robot removed the lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. label Jul 6, 2019

k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Oct 4, 2019

k8s-ci-robot removed the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Oct 13, 2019

k8s-ci-robot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Jan 11, 2020

k8s-ci-robot added lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. and removed lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. labels Feb 10, 2020

k8s-ci-robot closed this as completed Mar 11, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Stabilize describer outputs #25251

Stabilize describer outputs #25251

timoreimann commented May 6, 2016 •

edited

Loading

roberthbailey commented May 8, 2016

bgrant0607 commented May 9, 2016

smarterclayton commented May 9, 2016

bgrant0607 commented May 9, 2016

timoreimann commented May 11, 2016

sylwekb commented May 20, 2016

timoreimann commented May 20, 2016

timoreimann commented May 29, 2016

bgrant0607 commented May 31, 2016

timoreimann commented May 31, 2016 •

edited

Loading

derekwaynecarr commented May 31, 2016

timoreimann commented Jun 1, 2016

errordeveloper commented Nov 28, 2016 •

edited

Loading

smarterclayton commented Nov 28, 2016 via email

adohe-zz commented Nov 30, 2016

smarterclayton commented Nov 30, 2016 via email

fejta-bot commented Dec 19, 2017

timoreimann commented Dec 19, 2017

smarterclayton commented Dec 19, 2017

fejta-bot commented Feb 22, 2019

timoreimann commented Mar 7, 2019

fejta-bot commented Jun 5, 2019

fejta-bot commented Jul 5, 2019

errordeveloper commented Jul 6, 2019 via email

fejta-bot commented Oct 4, 2019

McCoyAle commented Oct 11, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

fejta-bot commented Jan 11, 2020

fejta-bot commented Feb 10, 2020

fejta-bot commented Mar 11, 2020

k8s-ci-robot commented Mar 11, 2020

Stabilize describer outputs #25251

Stabilize describer outputs #25251

Comments

timoreimann commented May 6, 2016 • edited Loading

roberthbailey commented May 8, 2016

bgrant0607 commented May 9, 2016

smarterclayton commented May 9, 2016

bgrant0607 commented May 9, 2016

timoreimann commented May 11, 2016

sylwekb commented May 20, 2016

timoreimann commented May 20, 2016

timoreimann commented May 29, 2016

bgrant0607 commented May 31, 2016

timoreimann commented May 31, 2016 • edited Loading

derekwaynecarr commented May 31, 2016

timoreimann commented Jun 1, 2016

errordeveloper commented Nov 28, 2016 • edited Loading

smarterclayton commented Nov 28, 2016 via email

adohe-zz commented Nov 30, 2016

smarterclayton commented Nov 30, 2016 via email

fejta-bot commented Dec 19, 2017

timoreimann commented Dec 19, 2017

smarterclayton commented Dec 19, 2017

fejta-bot commented Feb 22, 2019

timoreimann commented Mar 7, 2019

fejta-bot commented Jun 5, 2019

fejta-bot commented Jul 5, 2019

errordeveloper commented Jul 6, 2019 via email

fejta-bot commented Oct 4, 2019

McCoyAle commented Oct 11, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

timoreimann commented Oct 13, 2019

fejta-bot commented Jan 11, 2020

fejta-bot commented Feb 10, 2020

fejta-bot commented Mar 11, 2020

k8s-ci-robot commented Mar 11, 2020

timoreimann commented May 6, 2016 •

edited

Loading

timoreimann commented May 31, 2016 •

edited

Loading

errordeveloper commented Nov 28, 2016 •

edited

Loading