-
Notifications
You must be signed in to change notification settings - Fork 475
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.
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
add Precision Time Protocol support #4
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall, it looks very good to me and very aligned to the prototype we've already done. Good job!
time/ptp.md
Outdated
#### PTP4L configuration file | ||
|
||
Default ptp4l configuration file (ptp4l.conf) will be used when starting ptp4l | ||
service, user doesn't need to specify ptp4l.conf in `ptp4lOpts`. The content |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As I understand, the daemonset will use the default configuration from ptp4l.conf and the user can overwrite some of the options via ptp4lOpts. It would be worth to mention where this options will be stored (like a configMap or in /etc/sysconfig/ptp4l).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Correct!
updated the doc to add:
- ptp4l.conf is installed in ptp daemon image by default under
/etc/ptp4l.conf
-f ptp4l.conf
will be automatically appended toptp4lOpts
by PTP daemon- some of
ptp4l.conf
configs can be overwritten by specifying certain options inptp4lOpts
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
time/ptp.md
Outdated
In case of failure in linuxptp configuration, OpenShift nodes will be left as | ||
no time source to sync, resulting in potential time disorder among nodes. | ||
This can be resolved by providing PTP and NTP redundancy via timemaster service | ||
which automatically rolls back to use default NTP time source if configured. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks like we could mitigate the risk of failure by using timemaster (which is right) but we don't support timemaster service configuration. Just to make it clear.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
updated to explicitly mention :
- timemaster service can mitigate the redundancy issue between PTP and NTP
- timemaster configuration will not be supported in initial proposal.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
time/ptp.md
Outdated
|
||
#### Tech Preview -> GA | ||
|
||
- Have an opeartor to manage PTP CRDs |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Typo: operator
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Eagle eyes! Thanks ! fixed
## Infrastructure Needed [optional] | ||
|
||
This requires a github repo be created under openshift org to hold PTP | ||
DaemonSet code. The name of this repo is `ptp-daemon`. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As mentioned above, there will be the need of an operator managing the PTP daemonset. Then, there will be two different repos? one with the ptp-daemon and one with the ptp-operator?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, the idea is to first implement a PTP daemon in 4.3 as Tech Preview, and have an operator to manage this daemon and exposes all CRDs mentioned in proposal. Operator will need a separate repo under openshift github org (this is not strictly required, as you can see, SR-IOV Operator uses one repo to hold code for both daemon and operator).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, that's why I mentioned it. I think the trend in OpenShift is to name the repos according to the operator, even if they contains the code of the actual application that is being "operated".
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If the long term direction is to deliver this via an operator I think we should just go ahead and create the repo as if it's an operator.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
After looking at this some more it seems like there are more examples of separate repos for operator and operands, ie: image-registry, kubernetes-apiserver. So it seems there's no norm established here so lets just go with separate repos as initially proposed.
## Infrastructure Needed [optional] | ||
|
||
This requires a github repo be created under openshift org to hold PTP | ||
DaemonSet code. The name of this repo is `ptp-daemon`. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If the long term direction is to deliver this via an operator I think we should just go ahead and create the repo as if it's an operator.
@sdodson thanks for the review! |
@zshi-redhat lets go ahead with separate repos |
@sdodson cool, thanks, will use separate repos for ptp-daemon and ptp-operator |
@sdodson could you approve this PR, I think we are ready to merge. |
/approve |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: fepan, sdodson, zshi-redhat 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 |
I realize this already merged a while back and likely has had some refinement but I do want to ask a few questions:
|
OLM, it's an optional operator installed via openshift marketplace
Currently, no interaction with MCO. All ptp services are configured via this Opreator. |
Perfect
Got it 👍
OK, for this please make sure to interact with the MCO team when you get to the point of designing this addition (cc'ing @runcom as an FYI as he leads that team).
Awesome 😃 |
Signed-off-by: Nir <niry@redhat.com>
ConfigMap: make it not exclusive for ipfix (it could be helpful in th…
os: un-RAID ESP; use GRUB RAID module
No description provided.