feat(pytorch): Add init container config to avoid DNS lookup failure #1493
Conversation
|
/hold |
google-oss-prow
bot
requested review from
ChanYiLin,
jinchihe and
terrytangyuan
Signed-off-by: Ce Gao <ce.gao@outlook.com>
Signed-off-by: Ce Gao <ce.gao@outlook.com>
Signed-off-by: Ce Gao <ce.gao@outlook.com>
Pull Request Test Coverage Report for Build 1519323487
💛 - Coveralls |
Signed-off-by: Ce Gao <ce.gao@outlook.com>
|
/hold cancel |
| PyTorchInitContainerImageDefault = "alpine:3.10" | ||
| // PyTorchInitContainerTemplateFileDefault is the default template file for | ||
| // the pytorch init container. | ||
| PyTorchInitContainerTemplateFileDefault = "/etc/config/initContainer.yaml" |
There was a problem hiding this comment.
I saw PyTorchInitContainerImage in the original pytorch-operator, but not PyTorchInitContainerTemplateFileDefault. Do we really need this one?
There was a problem hiding this comment.
The implementation in pytorch operator hard coded a path. Here we use a CLI flag to make it configurable.
There was a problem hiding this comment.
LGTM. I would suggest @Jeffwan or other pytorch-controller contributor to take a look before merge.
There was a problem hiding this comment.
Does user have any requirements to custom init container? Could we make it completely transparent to users?
There was a problem hiding this comment.
I think they may update the resource requests. WDYT
/cc @johnugeorge
There was a problem hiding this comment.
i'd like to custom the init-container-image bescause of my project worked on a network limitations env,what shou i do?
| PyTorchInitContainerImageDefault = "alpine:3.10" | ||
| // PyTorchInitContainerTemplateFileDefault is the default template file for | ||
| // the pytorch init container. | ||
| PyTorchInitContainerTemplateFileDefault = "/etc/config/initContainer.yaml" |
There was a problem hiding this comment.
Does user have any requirements to custom init container? Could we make it completely transparent to users?
|
It looks good to me. Feel free to unhold it when it's ready to merge. /lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: gaocegege, Jeffwan The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/unhold |
What this PR does / why we need it:
This PR is to add the init container, which checks if the master is already running.
Which issue(s) this PR fixes (optional, in
Fixes #<issue number>, #<issue number>, ...format, will close the issue(s) when PR gets merged):Fixes #1482