Re: [Devtools] damn!

2018-07-19 Thread Burr Sutter 
On Thu, Jul 19, 2018 at 2:50 AM Gerard Braad  wrote:

> Burr
>
> >   I turned off wifi on the laptop and ran for 1 hour with most of the
> bit.ly/istio-tutorial demos completely disconnected.
>
> Curious what else you have done to make this work ... as we have a way
> to intercept DNS requests, but it is not foolproof yet (manual config)
>

I have Gasmask on the Mac.  It allows you to easily edit and make new
/etc/hosts files "on-the-fly".

So, I can easily just add/edit/activate different hosts files and since I
had been working on the plane (literally on-the-fly) I have experimented
with it a few times and it worked in this instance when minishift could not
see any network, it would start (I guess the ping 8.8.8.8 is its test to
see if it should try GitHub)

192.168.99.104   customer-tutorial.192.168.99.104.nip.io
192.168.99.104   grafana-istio-system.192.168.99.104.nip.io
192.168.99.104   tracing-istio-system.192.168.99.104.nip.io
192.168.99.104   prometheus-istio-system.192.168.99.104.nip.io
192.168.99.104   servicegraph-istio-system.192.168.99.104.nip.io
192.168.99.104   istio-ingressgateway-istio-system.192.168.99.104.nip.io


> On Thu, Jul 19, 2018 at 3:36 PM Budh Ram Gurung 
> wrote:
> >
> > Hi Burr,
> >
> > On Thu, Jul 19, 2018 at 11:53 AM Lalatendu Mohanty 
> wrote:
> >>
> >>
> >>
> >> On Thu, Jul 19, 2018 at 9:43 AM, Praveen Kumar 
> wrote:
> >>>
> >>> On Thu, Jul 19, 2018 at 12:31 AM Burr Sutter 
> wrote:
> >>> >
> >>> > Checking if requested OpenShift version 'v3.9.0' is valid ...
> >>> >
> >>> >Hit github rate limit: GET
> https://api.github.com/repos/openshift/origin/releases/tags/v3.9.0: 403
> API rate limit exceeded for 50.207.133.82. (But here's the good news:
> Authenticated requests get a higher rate limit. Check out the documentation
> for more details.) [rate reset in 13m35s]
> >>>
> >>> I hope this is coming from minishift side, for rate limit we have doc
> >>> in troubleshooting guide[0], I think it good to tell user this doc
> >>> link or says check the troubleshooting document section.
> >>>
> >>> [0]
> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#github-api-rate-limit-exceeded
> >>>
> >>
> >> Also you skip these startup checks [1] . However these checks are there
> to warn you or fail the minishift start if  certain conditions are not met.
> You can skip the individual checks  too.
> >>
> >> For example if you want to skip the version check then you need to
> skip the check i.e. "minishift config set skip-check-openshift-version
> true" .
> >>
> >> There are some other checks you can see in "minishift config --help |
> grep -i openshift"
> >>
> >>  * skip-check-openshift-version
> >
> >
> > This is the best solution in this case for you :)
> >
> >>
> >>  * warn-check-openshift-version
> >>  * skip-check-openshift-release
> >>  * warn-check-openshift-release
> >>
> >> [1]
> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#minshift-startup-check-failed
> >>
> >> -Lala
> >> ___
> >> Devtools mailing list
> >> Devtools@redhat.com
> >> https://www.redhat.com/mailman/listinfo/devtools
> >
> >
> > Regards,
> > Budh Ram Gurung
> >
> >
> > ___
> > Devtools mailing list
> > Devtools@redhat.com
> > https://www.redhat.com/mailman/listinfo/devtools
>
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] 10 pods per core

2018-07-19 Thread Jean-Francois Maury 
oc get nodes against cdk 3.5 (minishift ?) will show you the node is
limited to 20 pods

On Thu, Jul 19, 2018 at 2:49 PM Burr Sutter  wrote:

> There must still be some limit - and it seems about 10 pods per core (just
> counting the Running, not Completed/Pending/Error).
>
> Failed Scheduling
> 0/1 nodes are available: 1 Insufficient pods.
> 26 times in the last
>
> On Sun, Jul 15, 2018 at 10:27 PM Lalatendu Mohanty 
> wrote:
>
>>
>>
>> On Sun, Jul 15, 2018 at 12:57 AM, Clayton Coleman 
>> wrote:
>>
>>> In 3.9 we removed this default (or maybe 3.10).  If minishift isn’t
>>> explicitly setting it should already be relaxed.
>>>
>>>
>> +1, we had similar observation on 3.9.
>>
>> On Jul 14, 2018, at 2:51 PM, Burr Sutter  wrote:
>>>
>>> Can we relax that setting to allow 15 or 20 pods per code on minishift?
>>>
>>> In order to run "hello world" with Istio, you need at least 19 pods (not
>>> including build/deploy pods) and using 3 cores (on a 4 core machine) for
>>> the VM running minishift makes everything else (slides, chrome, etc) often
>>> much too slow.
>>>
>>> If you have not run our primary Istio tutorial via minishift, it would
>>> be a good experience for you :-)
>>> bit.ly/istio-tutorial
>>>
>>>
>>>
>>>
>>> ___
>>> Devtools mailing list
>>> Devtools@redhat.com
>>> https://www.redhat.com/mailman/listinfo/devtools
>>>
>>>
>>> ___
>>> Devtools mailing list
>>> Devtools@redhat.com
>>> https://www.redhat.com/mailman/listinfo/devtools
>>>
>>>
>> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools
>


-- 

JEFF MAURY

Red Hat



jma...@redhat.com


@redhatjobs    redhatjobs
   @redhatjobs

___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] 10 pods per core

2018-07-19 Thread Burr Sutter 
On Thu, Jul 19, 2018 at 7:10 AM Jean-Francois Maury 
wrote:

> oc get nodes against cdk 3.5 (minishift ?) will show you the node is
> limited to 20 pods
>

I do not see "20" in that output

oc get nodes

NAMESTATUSROLES AGE   VERSION

localhost   Ready 8hv1.9.1+a0ce1bc657

>
> On Thu, Jul 19, 2018 at 2:49 PM Burr Sutter  wrote:
>
>> There must still be some limit - and it seems about 10 pods per core
>> (just counting the Running, not Completed/Pending/Error).
>>
>> Failed Scheduling
>> 0/1 nodes are available: 1 Insufficient pods.
>> 26 times in the last
>>
>> On Sun, Jul 15, 2018 at 10:27 PM Lalatendu Mohanty 
>> wrote:
>>
>>>
>>>
>>> On Sun, Jul 15, 2018 at 12:57 AM, Clayton Coleman 
>>> wrote:
>>>
 In 3.9 we removed this default (or maybe 3.10).  If minishift isn’t
 explicitly setting it should already be relaxed.


>>> +1, we had similar observation on 3.9.
>>>
>>> On Jul 14, 2018, at 2:51 PM, Burr Sutter  wrote:

 Can we relax that setting to allow 15 or 20 pods per code on minishift?

 In order to run "hello world" with Istio, you need at least 19 pods
 (not including build/deploy pods) and using 3 cores (on a 4 core machine)
 for the VM running minishift makes everything else (slides, chrome, etc)
 often much too slow.

 If you have not run our primary Istio tutorial via minishift, it would
 be a good experience for you :-)
 bit.ly/istio-tutorial




 ___
 Devtools mailing list
 Devtools@redhat.com
 https://www.redhat.com/mailman/listinfo/devtools


 ___
 Devtools mailing list
 Devtools@redhat.com
 https://www.redhat.com/mailman/listinfo/devtools


>>> ___
>> Devtools mailing list
>> Devtools@redhat.com
>> https://www.redhat.com/mailman/listinfo/devtools
>>
>
>
> --
>
> JEFF MAURY
>
> Red Hat
>
> 
>
> jma...@redhat.com
> 
> 
> @redhatjobs    redhatjobs
>    @redhatjobs
> 
>
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] 10 pods per core

2018-07-19 Thread Burr Sutter 
There must still be some limit - and it seems about 10 pods per core (just
counting the Running, not Completed/Pending/Error).

Failed Scheduling
0/1 nodes are available: 1 Insufficient pods.
26 times in the last

On Sun, Jul 15, 2018 at 10:27 PM Lalatendu Mohanty 
wrote:

>
>
> On Sun, Jul 15, 2018 at 12:57 AM, Clayton Coleman 
> wrote:
>
>> In 3.9 we removed this default (or maybe 3.10).  If minishift isn’t
>> explicitly setting it should already be relaxed.
>>
>>
> +1, we had similar observation on 3.9.
>
> On Jul 14, 2018, at 2:51 PM, Burr Sutter  wrote:
>>
>> Can we relax that setting to allow 15 or 20 pods per code on minishift?
>>
>> In order to run "hello world" with Istio, you need at least 19 pods (not
>> including build/deploy pods) and using 3 cores (on a 4 core machine) for
>> the VM running minishift makes everything else (slides, chrome, etc) often
>> much too slow.
>>
>> If you have not run our primary Istio tutorial via minishift, it would be
>> a good experience for you :-)
>> bit.ly/istio-tutorial
>>
>>
>>
>>
>> ___
>> Devtools mailing list
>> Devtools@redhat.com
>> https://www.redhat.com/mailman/listinfo/devtools
>>
>>
>> ___
>> Devtools mailing list
>> Devtools@redhat.com
>> https://www.redhat.com/mailman/listinfo/devtools
>>
>>
>
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] damn!

2018-07-19 Thread Burr Sutter 
On Thu, Jul 19, 2018 at 1:36 AM Budh Ram Gurung  wrote:

> Hi Burr,
>
> On Thu, Jul 19, 2018 at 11:53 AM Lalatendu Mohanty 
> wrote:
>
>>
>>
>> On Thu, Jul 19, 2018 at 9:43 AM, Praveen Kumar 
>> wrote:
>>
>>> On Thu, Jul 19, 2018 at 12:31 AM Burr Sutter  wrote:
>>> >
>>> > Checking if requested OpenShift version 'v3.9.0' is valid ...
>>> >
>>> >Hit github rate limit: GET
>>> https://api.github.com/repos/openshift/origin/releases/tags/v3.9.0: 403
>>> API rate limit exceeded for 50.207.133.82. (But here's the good news:
>>> Authenticated requests get a higher rate limit. Check out the documentation
>>> for more details.) [rate reset in 13m35s]
>>>
>>> I hope this is coming from minishift side, for rate limit we have doc
>>> in troubleshooting guide[0], I think it good to tell user this doc
>>> link or says check the troubleshooting document section.
>>>
>>> [0]
>>> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#github-api-rate-limit-exceeded
>>>
>>>
>> Also you skip these startup checks [1] . However these checks are there
>> to warn you or fail the minishift start if  certain conditions are not met.
>> You can skip the individual checks  too.
>>
>> For example if you want to skip the version check then you need to  skip
>> the check i.e. "minishift config set skip-check-openshift-version true"
>> .
>>
>> There are some other checks you can see in "minishift config --help |
>> grep -i openshift"
>>
>>  * skip-check-openshift-version
>>
>
> This is the best solution in this case for you :)
>

Is there a reason why this check is "fatal", not allowing minishift to
start? It seems like it should not be a fatal error (GitHub rate limiting
therefore die)

>
>
>>  * warn-check-openshift-version
>>  * skip-check-openshift-release
>>  * warn-check-openshift-release
>>
>> [1]
>> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#minshift-startup-check-failed
>>
>> -Lala
>> ___
>> Devtools mailing list
>> Devtools@redhat.com
>> https://www.redhat.com/mailman/listinfo/devtools
>
>
> Regards,
> Budh Ram Gurung
>
>
> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools
>
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] damn!

2018-07-19 Thread Gerard Braad 
Burr

>   I turned off wifi on the laptop and ran for 1 hour with most of the 
> bit.ly/istio-tutorial demos completely disconnected.

Curious what else you have done to make this work ... as we have a way
to intercept DNS requests, but it is not foolproof yet (manual config)
On Thu, Jul 19, 2018 at 3:36 PM Budh Ram Gurung  wrote:
>
> Hi Burr,
>
> On Thu, Jul 19, 2018 at 11:53 AM Lalatendu Mohanty  
> wrote:
>>
>>
>>
>> On Thu, Jul 19, 2018 at 9:43 AM, Praveen Kumar  wrote:
>>>
>>> On Thu, Jul 19, 2018 at 12:31 AM Burr Sutter  wrote:
>>> >
>>> > Checking if requested OpenShift version 'v3.9.0' is valid ...
>>> >
>>> >Hit github rate limit: GET 
>>> > https://api.github.com/repos/openshift/origin/releases/tags/v3.9.0: 403 
>>> > API rate limit exceeded for 50.207.133.82. (But here's the good news: 
>>> > Authenticated requests get a higher rate limit. Check out the 
>>> > documentation for more details.) [rate reset in 13m35s]
>>>
>>> I hope this is coming from minishift side, for rate limit we have doc
>>> in troubleshooting guide[0], I think it good to tell user this doc
>>> link or says check the troubleshooting document section.
>>>
>>> [0] 
>>> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#github-api-rate-limit-exceeded
>>>
>>
>> Also you skip these startup checks [1] . However these checks are there to 
>> warn you or fail the minishift start if  certain conditions are not met. You 
>> can skip the individual checks  too.
>>
>> For example if you want to skip the version check then you need to  skip the 
>> check i.e. "minishift config set skip-check-openshift-version true" .
>>
>> There are some other checks you can see in "minishift config --help | grep 
>> -i openshift"
>>
>>  * skip-check-openshift-version
>
>
> This is the best solution in this case for you :)
>
>>
>>  * warn-check-openshift-version
>>  * skip-check-openshift-release
>>  * warn-check-openshift-release
>>
>> [1] 
>> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#minshift-startup-check-failed
>>
>> -Lala
>> ___
>> Devtools mailing list
>> Devtools@redhat.com
>> https://www.redhat.com/mailman/listinfo/devtools
>
>
> Regards,
> Budh Ram Gurung
>
>
> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools

___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] damn!

2018-07-19 Thread Budh Ram Gurung 
Hi Burr,

On Thu, Jul 19, 2018 at 11:53 AM Lalatendu Mohanty 
wrote:

>
>
> On Thu, Jul 19, 2018 at 9:43 AM, Praveen Kumar  wrote:
>
>> On Thu, Jul 19, 2018 at 12:31 AM Burr Sutter  wrote:
>> >
>> > Checking if requested OpenShift version 'v3.9.0' is valid ...
>> >
>> >Hit github rate limit: GET
>> https://api.github.com/repos/openshift/origin/releases/tags/v3.9.0: 403
>> API rate limit exceeded for 50.207.133.82. (But here's the good news:
>> Authenticated requests get a higher rate limit. Check out the documentation
>> for more details.) [rate reset in 13m35s]
>>
>> I hope this is coming from minishift side, for rate limit we have doc
>> in troubleshooting guide[0], I think it good to tell user this doc
>> link or says check the troubleshooting document section.
>>
>> [0]
>> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#github-api-rate-limit-exceeded
>>
>>
> Also you skip these startup checks [1] . However these checks are there to
> warn you or fail the minishift start if  certain conditions are not met.
> You can skip the individual checks  too.
>
> For example if you want to skip the version check then you need to  skip
> the check i.e. "minishift config set skip-check-openshift-version true" .
>
> There are some other checks you can see in "minishift config --help | grep
> -i openshift"
>
>  * skip-check-openshift-version
>

This is the best solution in this case for you :)


>  * warn-check-openshift-version
>  * skip-check-openshift-release
>  * warn-check-openshift-release
>
> [1]
> https://docs.openshift.org/latest/minishift/troubleshooting/troubleshooting-getting-started.html#minshift-startup-check-failed
>
> -Lala
> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools


Regards,
Budh Ram Gurung
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools

Re: [Devtools] 10 pods per core

2018-07-19 Thread Bob Davis 
This is a good point though - as I gather that a lot of the administrative
workload is being shifted over to the team that owns the app vs. the
traditional IT/admin teams we have traditionally built for. This means
we're going to have more and more developer-admins who know Linux and/or
OpenShift poorly and for whom achieving wizard-like admin chops is not a
goal.

Bob


On Thu, Jul 19, 2018 at 9:49 AM Burr Sutter  wrote:

>
>
> On Thu, Jul 19, 2018 at 7:26 AM Jean-Francois Maury 
> wrote:
>
>> Sorry oc get nodes -o yaml
>>
>
> since I am mostly trying to be a Java & JavaScript developer, I am a poor
> OpenShift and Linux sys admin/operator
>
> so, I am pretty literal :-)
>
>
>>
>
> On Thu, Jul 19, 2018 at 3:21 PM Burr Sutter  wrote:
>>
>>>
>>>
>>> On Thu, Jul 19, 2018 at 7:10 AM Jean-Francois Maury 
>>> wrote:
>>>
 oc get nodes against cdk 3.5 (minishift ?) will show you the node is
 limited to 20 pods

>>>
>>> I do not see "20" in that output
>>>
>>> oc get nodes
>>>
>>> NAMESTATUSROLES AGE   VERSION
>>>
>>> localhost   Ready 8hv1.9.1+a0ce1bc657
>>>

 On Thu, Jul 19, 2018 at 2:49 PM Burr Sutter  wrote:

> There must still be some limit - and it seems about 10 pods per core
> (just counting the Running, not Completed/Pending/Error).
>
> Failed Scheduling
> 0/1 nodes are available: 1 Insufficient pods.
> 26 times in the last
>
> On Sun, Jul 15, 2018 at 10:27 PM Lalatendu Mohanty <
> lmoha...@redhat.com> wrote:
>
>>
>>
>> On Sun, Jul 15, 2018 at 12:57 AM, Clayton Coleman <
>> ccole...@redhat.com> wrote:
>>
>>> In 3.9 we removed this default (or maybe 3.10).  If minishift isn’t
>>> explicitly setting it should already be relaxed.
>>>
>>>
>> +1, we had similar observation on 3.9.
>>
>> On Jul 14, 2018, at 2:51 PM, Burr Sutter  wrote:
>>>
>>> Can we relax that setting to allow 15 or 20 pods per code on
>>> minishift?
>>>
>>> In order to run "hello world" with Istio, you need at least 19 pods
>>> (not including build/deploy pods) and using 3 cores (on a 4 core 
>>> machine)
>>> for the VM running minishift makes everything else (slides, chrome, etc)
>>> often much too slow.
>>>
>>> If you have not run our primary Istio tutorial via minishift, it
>>> would be a good experience for you :-)
>>> bit.ly/istio-tutorial
>>>
>>>
>>>
>>>
>>> ___
>>> Devtools mailing list
>>> Devtools@redhat.com
>>> https://www.redhat.com/mailman/listinfo/devtools
>>>
>>>
>>> ___
>>> Devtools mailing list
>>> Devtools@redhat.com
>>> https://www.redhat.com/mailman/listinfo/devtools
>>>
>>>
>> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools
>


 --

 JEFF MAURY

 Red Hat

 

 jma...@redhat.com
 
 
 @redhatjobs    redhatjobs
    @redhatjobs
 

>>>
>>
>> --
>>
>> JEFF MAURY
>>
>> Red Hat
>>
>> 
>>
>> jma...@redhat.com
>> 
>> 
>> @redhatjobs    redhatjobs
>>    @redhatjobs
>> 
>>
> ___
> Devtools mailing list
> Devtools@redhat.com
> https://www.redhat.com/mailman/listinfo/devtools
>


-- 

Bob Davis
Senior Product Manager, Red Hat Developers
bobda...@redhat.com | 210-452-8945
developers.redhat.com
___
Devtools mailing list
Devtools@redhat.com
https://www.redhat.com/mailman/listinfo/devtools