Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Cannot get to compile on the PI #3406

Closed
Mccpie01 opened this issue Jan 27, 2019 · 13 comments
Closed

Cannot get to compile on the PI #3406

Mccpie01 opened this issue Jan 27, 2019 · 13 comments

Comments

@Mccpie01
Copy link

I am trying to compile for the ARMS (Raspberry Pi/ Rock Pi). But i get the error with that make.

sudo make launch

GO15VENDOREXPERIMENT=1 go install -ldflags "-X github.com/pachyderm/pachyderm/src/server/vendor/github.com/pachyderm/pachyderm/src/client/version.AdditionalVersion=-82a7869060f00a48b266fffde9cdf87452740370" ./src/server/cmd/pachctl pachctl deploy local --dry-run | kubectl apply -f - /bin/sh: 1: pachctl: Exec format error error: no objects passed to apply Makefile:357: recipe for target 'launch' failed make: *** [launch] Error 1
My best Guss is that it did not see the ARM architecture. Can it be added?

What is the goal / desired outcome? To make it work on ARMs (Raspberry pi/Rockpi)

If there is a way to accomplish this today via a workaround, what does that require? recognize that it’s on arm

(Optional) What is your proposal for a feature to solve this? Add support to recognize that it’s on arm

Environment?:

  • Kubernetes version (use kubectl version):
    Client Version: version.Info{Major:"1", Minor:"13", GitVersion:"v1.13.2", GitCommit:"cff46ab41ff0bb44d8584413b598ad8360ec1def", GitTreeState:"clean", BuildDate:"2019-01-10T23:35:51Z", GoVersion:"go1.11.4", Compiler:"gc", Platform:"linux/arm64"}
    Server Version: version.Info{Major:"1", Minor:"13", GitVersion:"v1.13.1", GitCommit:"eec55b9ba98609a46fee712359c7b5b365bdd920", GitTreeState:"clean", BuildDate:"2018-12-13T10:31:33Z", GoVersion:"go1.11.2", Compiler:"gc", Platform:"linux/arm64"}

  • Pachyderm CLI version (use pachctl version): won't compile/Install

  • OS (e.g. from /etc/os-release):
    NAME="Ubuntu"
    VERSION="18.04.1 LTS (Bionic Beaver)"
    ID=ubuntu
    ID_LIKE=debian
    PRETTY_NAME="Ubuntu 18.04.1 LTS"
    VERSION_ID="18.04"
@gabrielgrant
Copy link
Contributor

gabrielgrant commented Jan 28, 2019 •
edited
Loading

To clarify, are you trying to run pachctl on the RPi, or run the whole of pachyderm (including kubernetes) there?

Looks like we had an issue for pachctl support in #1734 which should be fairly straightforward

Support for deploying all of pachyderm onto arm, however, is probably going to be a bit more tricky

@gabrielgrant gabrielgrant mentioned this issue Jan 28, 2019
Closed
@Mccpie01
Copy link
Author

I want to run pachyderm on my Pi. I cannot get it to compile.

I saw that you tried it in the past. Were you ever able to compile it on the Pi?

@gabrielgrant
Copy link
Contributor

gabrielgrant commented Jan 28, 2019 •
edited
Loading

I have not personally re-visited running on RPi, but I think the best way to approach this would be to break down the steps:

  1. build the command line tool (make install)
  2. Make sure that works (run pachctl version --client-only)
  3. run pachctl deploy local
  4. check the deploy with kubectl get all
  5. ensure you can talk to the cluster (pachctl version)
  6. ensure all functionality works, by following the intro tutorial: http://docs.pachyderm.io/en/latest/getting_started/beginner_tutorial.html

@Mccpie01
Copy link
Author

I see this in the make file
launch: install check-kubectl $(eval STARTTIME := $(shell date +%s)) pachctl deploy local --dry-run | kubectl $(KUBECTLFLAGS) apply -f - # wait for the pachyderm to come up until timeout 1s ./etc/kube/check_ready.sh app=pachd; do sleep 1; done @echo "pachd launch took $$(($$(date +%s) - $(STARTTIME))) seconds"

line 356 of https://github.com/pachyderm/pachyderm/blob/master/Makefile

@gabrielgrant
Copy link
Contributor

yes, sorry, i was totally wrong about that. just updated my comment to hopefully be more helpful

@gabrielgrant
Copy link
Contributor

to be clear, are you trying to run the entire kubernetes cluster on the RPi? or just the pachctl client, talking to a remote kubernetes cluster?

@Mccpie01
Copy link
Author

I found the issue. It installs in my local user and not in /usr/local/bin
and in the user local bin is a x86 bin. (Probable for my first steps in Linux )
So i have to close the issue.

Thanks for pointing me in the right direction

@Mccpie01
Copy link
Author

``

@gabrielgrant
Copy link
Contributor

@Mccpie01 to clarify, did you manage to get it to compile for arm?

@Mccpie01
Copy link
Author

Mccpie01 commented Jan 29, 2019 via email

@Mccpie01
Copy link
Author

Mccpie01 commented Jan 29, 2019 via email

@gabrielgrant
Copy link
Contributor

ok, to get the whole of pach running on RPi will require rebuilding the docker images for ARM, in addition to recompiling the command line client

Have you got kubernetes running on RPi/ARM already? It seems that isn't something that is officially supported by k8s out of the box, though it does seem to be possible: https://github.com/luxas/kubernetes-on-arm

@Mccpie01
Copy link
Author

Mccpie01 commented Jan 30, 2019 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@gabrielgrant @Mccpie01