Re: One command to check if all etcd serverd are up and if all masters are connected to this etcd nodes?

2017-01-13 Thread Jay Vyas 
etcdhealth probably is sufficient but in any case

I was recently wondering myself what etcdhealth measured + does it
gaurantee the nodes are consistent, synced, no split brain?

We used to do something like this:

curl -L -X PUT http://node1:2379/v2/keys/message -d value="ABCD"
&& curl -L http://node2:2379/v2/keys/message | grep -q ABCD
&& curl -L http://node3:2379/v2/keys/message | grep -q ABCD

To guarantee that writes were working + consistent across the cluster.

jay vyas

On Fri, Jan 13, 2017 at 8:04 AM, Aleksandar Lazic 
wrote:

> Hi.
>
> In case the peer* stuff doesn't work I have used this command.
> Just as plan B.
>
> ###
> /usr/bin/etcdctl --peers \
>   $( awk -F= '/ETCD_LISTEN_CLIENT_URLS=/ {print $2}' < /etc/etcd/etcd.conf
> ) \
>   --debug \
>   --ca-file /etc/origin/master/master.etcd-ca.crt \
>   --key-file /etc/origin/master/master.etcd-client.key \
>   --cert-file /etc/origin/master/master.etcd-client.crt \
>   cluster-health
> ###
>
> BR Aleks
>
>
>
>  On Fri, 13 Jan 2017 12:57:24 +0100 *Diego Castro
> >* wrote 
>
> You can check cluster health with etcdctl command, just run the following
> command on one of yours etcd servers:
>
> $ cd /etc/etcd/
>
> $ etcdctl  --cert-file peer.crt --key-file peer.key --ca-file ca.crt -C
> https://:2379  cluster-health
>
>
> ---
> Diego Castro / The CloudFather
> GetupCloud.com - Eliminamos a Gravidade
>
> 2017-01-13 7:24 GMT-03:00 Stéphane Klein :
>
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
> Hi,
>
> I'm wondering if there are a command to check etcd servers status?
> Check if all etcd nodes are up.
>
> ```
> etcdClientInfo:
>   ca: master.etcd-ca.crt
>   certFile: master.etcd-client.crt
>   keyFile: master.etcd-client.key
>   urls:
> - https://etcd-1.priv.example.com:2379
> - https://etcd-2.priv.example.com:2379
> - https://etcd-3.priv.example.com:2379
> ```
>
> Check if this configuration is valid on all OpenShift masters.
>
> Something like `oc get nodes`.
>
> I try `oc status` but I haven't information about etcd configure.
>
> Best regards,
> Stéphane
> --
> Stéphane Klein 
> blog: http://stephane-klein.info
> cv : http://cv.stephane-klein.info
> Twitter: http://twitter.com/klein_stephane
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
> ---
> Mit freundlichen Grüßen
> Aleksandar Lazic - ME2Digital e. U.
> https://me2digital.online/
> UID-Nr.: ATU71765716
> IBAN: AT27 1420 0200 1096 9086
> Firmenbuch: 462678 i
>
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
>
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: S2I and git-lfs

2017-01-13 Thread Ben Parees 
On Thu, Jan 12, 2017 at 8:36 PM, Clayton Coleman 
wrote:

> Is git-lfs packaged in epel ben?
>

​does not appear to be, no.  There is a rhel package, but it's not produced
by us:
https://github.com/git-lfs/git-lfs/wiki/Installation#rhelcentos
​



>
> On Jan 12, 2017, at 4:10 PM, Philippe Lafoucrière <
> philippe.lafoucri...@tech-angels.com> wrote:
>
>
> On Thu, Jan 12, 2017 at 3:54 PM, Ben Parees  wrote:
>
>> ​that qualifies as client-side logic (the plugin is executing the logic
>> on the client side, seamlessly or otherwise)
>
>
> Ok, thanks for the explanation.
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
>


-- 
Ben Parees | OpenShift
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: Run a HTTP/2 service on Openshift

2017-01-13 Thread Aleksandar Lazic 


Hi Karl.



you can use the passthrough mode and terminate the gRPC service with https 
directly.



https://docs.openshift.org/latest/architecture/core_concepts/routes.html#secured-routes

=Passthrough Termination



This is the TCP mode you searching for ;-).



Hth

Aleks



 On Fri, 13 Jan 2017 11:22:19 +0100 Karl Gerhard karl_g...@gmx.at 
wrote 




Hi, 



ist it possible to run gRPC on Openshift? 

As far as I understand gRPC uses HTTP/2 and the Openshift Router/Haproxy 
doesn't support HTTP/2 yet. Only way I can think of to make this work is 
modifying the Haproxy config to use "mode tcp" instead of "mode http" which is 
the default. But the Haproxy config is automatically generated by the Go Binary 
in the Pod so this solution would probably involve a lot of fiddling around and 
end up being not very pretty. 



Has anyone succeeded in getting gRPC or any HTTP/2 service to work on Openshift 
and could share his/her experience? 



Regards 

Karl 



___ 

users mailing list 

users@lists.openshift.redhat.com 

http://lists.openshift.redhat.com/openshiftmm/listinfo/users 





--- 

Mit freundlichen Grüßen

Aleksandar Lazic - ME2Digital e. U.

https://me2digital.online/

UID-Nr.: ATU71765716

IBAN: AT27 1420 0200 1096 9086
Firmenbuch: 462678 i





___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: One command to check if all etcd serverd are up and if all masters are connected to this etcd nodes?

2017-01-13 Thread Diego Castro 
You can check cluster health with etcdctl command, just run the following
command on one of yours etcd servers:

$ cd /etc/etcd/

$ etcdctl  --cert-file peer.crt --key-file peer.key --ca-file ca.crt -C
https://:2379  cluster-health


---
Diego Castro / The CloudFather
GetupCloud.com - Eliminamos a Gravidade

2017-01-13 7:24 GMT-03:00 Stéphane Klein :

> Hi,
>
> I'm wondering if there are a command to check etcd servers status?
> Check if all etcd nodes are up.
>
> ```
> etcdClientInfo:
>   ca: master.etcd-ca.crt
>   certFile: master.etcd-client.crt
>   keyFile: master.etcd-client.key
>   urls:
> - https://etcd-1.priv.example.com:2379
> - https://etcd-2.priv.example.com:2379
> - https://etcd-3.priv.example.com:2379
> ```
>
> Check if this configuration is valid on all OpenShift masters.
>
> Something like `oc get nodes`.
>
> I try `oc status` but I haven't information about etcd configure.
>
> Best regards,
> Stéphane
> --
> Stéphane Klein 
> blog: http://stephane-klein.info
> cv : http://cv.stephane-klein.info
> Twitter: http://twitter.com/klein_stephane
>
> ___
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
>
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: How can I scale up the number of etcd host ?

2017-01-13 Thread Stéphane Klein 
2017-01-10 20:17 GMT+01:00 Scott Dodson :

> openshift-ansible doesn't currently provide this, there's an issue
> requesting it https://github.com/openshift/openshift-ansible/issues/1772
> which links to a blog post describing how to do it, though I've not
> validated that myself. The only hard part is the certificate
> management, otherwise scaling procedures should mirror those
> documented by etcd upstream.
>

This is my experimentation:

* I add 2 news etcd host (2 etcd hosts are useless, see
https://coreos.com/etcd/docs/latest/v2/admin_guide.html#optimal-cluster-size
)
* Next, I have executed this Ansible playbooks:

*
https://github.com/openshift/openshift-ansible/blob/master/playbooks/byo/openshift-cluster/redeploy-certificates.yml
*
https://github.com/openshift/openshift-ansible/blob/master/playbooks/byo/openshift-cluster/config.yml
  with --skip-tags=hosted option

* Finally I restart origin-master

Troubles:

* if I shutdown etcd1, I need to restart master node to see data in
OpenShift Console or launch build…

This behavior is normal? Voluntary? or I have something badly configured?

Best regards,
Stéphane
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Run a HTTP/2 service on Openshift

2017-01-13 Thread Karl Gerhard 

Hi,

ist it possible to run gRPC on Openshift?
As far as I understand gRPC uses HTTP/2 and the Openshift Router/Haproxy doesn't support HTTP/2 
yet. Only way I can think of to make this work is modifying the Haproxy config to use "mode 
tcp" instead of "mode http" which is the default. But the Haproxy config is 
automatically generated by the Go Binary in the Pod so this solution would probably involve a lot 
of fiddling around and end up being not very pretty.

Has anyone succeeded in getting gRPC or any HTTP/2 service to work on Openshift 
and could share his/her experience?

Regards
Karl

___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Re: How can I scale up the number of etcd host ?

2017-01-13 Thread Stéphane Klein 
2017-01-12 17:12 GMT+01:00 Alex Wauck :

> Are you using the built-in OpenShift etcd on that one node, or are you
> using real etcd?
>

I use registry.access.redhat.com/rhel7/etc standard docker OpenShift image.

Best regards,
Stéphane
___
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users