Don't allocate a pseudo-TTY #169

pitkley · 2017-12-10T12:40:20Z

This fixes issue #52 as far as I can tell.

Since documentation on the Docker -t flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross.

From my arguably limited tests, I have confirmed that echo hello | cross run works. Furthermore, this enables me to run cross build in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: the input device is not a TTY.

This fixes issue cross-rs#52 as far as I can tell. Since documentation on the `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross.

japaric · 2017-12-16T14:11:51Z

Thanks, @pitkley. LGTM.

Let's see what homu thinks.

@homunkulus r+

homunkulus · 2017-12-16T14:11:52Z

📌 Commit 974d8c6 has been approved by japaric

homunkulus · 2017-12-16T14:11:59Z

⌛ Testing commit 974d8c6 with merge 617b9bd...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2017-12-16T16:42:05Z

💔 Test failed - status-travis

pitkley · 2018-01-12T09:36:25Z

Sorry that I forgot about this PR. Unfortunately, I can't make anything out of the failed architectures. Are those failures unrelated/flaky, or is this something caused by this PR?

homunkulus · 2018-01-12T20:52:54Z

⌛ Testing commit 974d8c6 with merge 3163527...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-01-12T20:54:00Z

💔 Test failed - status-travis

homunkulus · 2018-01-13T16:12:51Z

⌛ Testing commit 974d8c6 with merge 1090151...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-01-13T16:13:45Z

💔 Test failed - status-travis

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-01-23T20:11:32Z

⌛ Testing commit 974d8c6 with merge 81a7d40...

homunkulus · 2018-01-23T20:13:56Z

💔 Test failed - status-travis

homunkulus · 2018-01-24T19:31:11Z

⌛ Testing commit 974d8c6 with merge bb6c62f...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-01-24T19:48:12Z

⌛ Testing commit 974d8c6 with merge c9f65fc...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-01-24T19:49:57Z

💔 Test failed - status-travis

myagley · 2018-02-15T00:50:52Z

Hello. Is there any update on the status of this PR? Thanks!

homunkulus · 2018-02-17T16:03:00Z

⌛ Testing commit 974d8c6 with merge f85aca9...

Don't allocate a pseudo-TTY This fixes issue #52 as far as I can tell. Since documentation on the Docker `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross. From my arguably limited tests, I have [confirmed](https://gist.github.com/pitkley/c581f612225688937cc8b7f3a7deff9a) that `echo hello | cross run` works. Furthermore, this enables me to run `cross build` in a Gitlab CI job, which otherwise also failed with the error mentioned in #52: `the input device is not a TTY`.

homunkulus · 2018-02-17T16:05:40Z

💔 Test failed - status-travis

myagley · 2018-04-30T22:27:35Z

Is it possible to please get this merged? I can't tell from the test output what is causing the failures?

darobs · 2018-05-14T18:02:26Z

It would be really nice to run cross in a CI job, is it possible to get this PR merged?

Dylan-DPC-zz · 2018-05-14T18:19:47Z

@homunkulus try

darobs · 2018-05-18T00:11:15Z

Thank you @Dylan-DPC - Any insight as to what's happening during the test?

Dylan-DPC-zz · 2018-10-12T14:06:40Z

bors: try

malbarbo · 2018-10-12T14:34:34Z

We lost color output if we don't allocate a tty. See for example https://travis-ci.org/rust-embedded/cross/jobs/440652090#L862

bors · 2018-10-12T15:36:12Z

try

Build failed

continuous-integration/travis-ci/push

borsboom · 2018-11-17T00:01:16Z

This also hit me trying to build using cross using Azure Pipelines, for the same reason as #52 (comment).

Could this PR be modified to use atty to detect whether it's running on a TTY, and use that to decide whether to pass through docker -t?

ndusart · 2018-12-11T11:45:20Z

Any chance this PR could be merged ?

It's the only blocking step before being able to use cross in our CI jobs...

Disasm · 2019-02-20T18:07:32Z

bors try

bors · 2019-02-20T19:30:28Z

try

Build failed

continuous-integration/travis-ci/push

Disasm · 2019-03-03T08:36:21Z

@rust-embedded/tools Could someone help with this?

dignifiedquire · 2019-03-26T20:11:55Z

any update on this, this is blocking me from using cross on CI

walfie · 2019-05-10T18:31:03Z

I'm also looking into using Azure Pipelines for CI, and it seems like there's several projects that have hacky workarounds that are waiting for this fix, e.g.,

would be great to have these issues resolved..

roblourens · 2019-06-29T03:14:31Z

Sorry to generate more noise on this, but what can we do to help this PR get merged? It's not clear to me whether the CI issues are legit? I would also really like to use this with Azure Pipelines.

Alternatively does anyone know how to get Azure Pipelines to allocate a TTY for a script?

xoac · 2019-08-23T20:37:48Z

@roblourens Look here for my workaround https://github.com/xoac/rust-azure-pipelines/blob/983921f9df3bdb43ef86bac8c11f0caa073ecfde/ci/install-cross-rust.yml

reitermarkus · 2019-09-01T01:18:53Z

bors try

bors · 2019-09-01T02:09:49Z

try

Build failed

continuous-integration/travis-ci/push

Don't allocate a pseudo-TTY

974d8c6

This fixes issue cross-rs#52 as far as I can tell. Since documentation on the `-t` flag is rather sparse, I am not sure if this has any negative implications on existing usage of cross.

bors bot added a commit that referenced this pull request Oct 12, 2018

Try #169:

75a8bd0

bors bot added a commit that referenced this pull request Feb 20, 2019

Try #169:

ae48126

Disasm added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-tools labels Feb 20, 2019

Disasm requested a review from a team February 20, 2019 18:11

Disasm requested a review from ryankurte March 3, 2019 08:39

taiki-e mentioned this pull request Jul 26, 2019

Set up CI with Azure Pipelines tokio-rs/bytes#275

Merged

bors bot added a commit that referenced this pull request Sep 1, 2019

Try #169:

2bfef65

reitermarkus mentioned this pull request Sep 1, 2019

Avoid TTY if not already running as TTY. #294

Merged

bors bot closed this in 13500f7 Sep 1, 2019

pitkley deleted the docker-no-pseudo-tty branch September 1, 2019 14:10

Don't allocate a pseudo-TTY #169

Don't allocate a pseudo-TTY #169

Conversation

pitkley commented Dec 10, 2017

japaric commented Dec 16, 2017

homunkulus commented Dec 16, 2017

homunkulus commented Dec 16, 2017

homunkulus commented Dec 16, 2017

pitkley commented Jan 12, 2018

homunkulus commented Jan 12, 2018

homunkulus commented Jan 12, 2018

homunkulus commented Jan 13, 2018

homunkulus commented Jan 13, 2018

homunkulus commented Jan 23, 2018

homunkulus commented Jan 23, 2018

homunkulus commented Jan 24, 2018

homunkulus commented Jan 24, 2018

homunkulus commented Jan 24, 2018

myagley commented Feb 15, 2018

homunkulus commented Feb 17, 2018

homunkulus commented Feb 17, 2018

myagley commented Apr 30, 2018

darobs commented May 14, 2018

Dylan-DPC-zz commented May 14, 2018

darobs commented May 18, 2018

Dylan-DPC-zz commented Oct 12, 2018

malbarbo commented Oct 12, 2018

bors bot commented Oct 12, 2018

try

Build failed

borsboom commented Nov 17, 2018

ndusart commented Dec 11, 2018 • edited Loading

Disasm commented Feb 20, 2019

bors bot commented Feb 20, 2019

try

Build failed

Disasm commented Mar 3, 2019 • edited Loading

dignifiedquire commented Mar 26, 2019

walfie commented May 10, 2019

roblourens commented Jun 29, 2019

xoac commented Aug 23, 2019

reitermarkus commented Sep 1, 2019

bors bot commented Sep 1, 2019

try

Build failed

ndusart commented Dec 11, 2018 •

edited

Loading

Disasm commented Mar 3, 2019 •

edited

Loading