IBM Verify

Hello,
I have set up a junction to my openshift platform.
I have a strange behaviour, the junction results down but if I call the junction the application behind work fine and at that moment the junction is up, only to come back down after some time.

before invoking the junction 

After invoking the junction 

what do i need to configure to have the junction always up and running?

Hello,
I have set up a junction to my openshift platform.
I have a strange behaviour, the junction results down but if I call the junction the application behind work fine and at that moment the junction is up, only to come back down after some time.

before invoking the junction 

After invoking the junction 

what do i need to configure to have the junction always up and running?

Hi Scott,

Thank you for your answer,
I have an application exposed by RH Openshift via a route like http://myapp.myopenshit.com/myapp.

I created a trasparent junction like this:
Server task mywebseal create -t tcp -h myapp.myopenshit.com -p 80 -v myapp.myopenshit.com -x /myapp

If I try to see the status of this new junction via the command:
Server task mywebseal show /myapp
I can see that the resource is online but the server status is not running as you can see from the following image.

If I then call my junction from my portal (https://my.websealdomain.it/myapp) it works fine and the server status goes from 'not running' to 'running' and then returns to a 'not running' status after some time.

If I may add one more piece of information if I try to connect to the same endpoint configured on the junction it responds correctly.

What do i have to configure on  webseal side or on RH Openshift side to make the junction always running?

where is my error?

Regards,

Fabio

Hello,
I have set up a junction to my openshift platform.
I have a strange behaviour, the junction results down but if I call the junction the application behind work fine and at that moment the junction is up, only to come back down after some time.

before invoking the junction 

After invoking the junction 

what do i need to configure to have the junction always up and running?

Hi Scott,

Thank you for your answer,
I have an application exposed by RH Openshift via a route like http://myapp.myopenshit.com/myapp.

I created a trasparent junction like this:
Server task mywebseal create -t tcp -h myapp.myopenshit.com -p 80 -v myapp.myopenshit.com -x /myapp

If I try to see the status of this new junction via the command:
Server task mywebseal show /myapp
I can see that the resource is online but the server status is not running as you can see from the following image.

If I then call my junction from my portal (https://my.websealdomain.it/myapp) it works fine and the server status goes from 'not running' to 'running' and then returns to a 'not running' status after some time.

If I may add one more piece of information if I try to connect to the same endpoint configured on the junction it responds correctly.

What do i have to configure on  webseal side or on RH Openshift side to make the junction always running?

where is my error?

Regards,

Fabio

Hello,
I have set up a junction to my openshift platform.
I have a strange behaviour, the junction results down but if I call the junction the application behind work fine and at that moment the junction is up, only to come back down after some time.

before invoking the junction 

After invoking the junction 

what do i need to configure to have the junction always up and running?

Hi Fabio,

I just helped a customer with this exact same issue.  Check the ping-uri in the WRP config file.  For example,

[junction]
ping-method = HEAD
ping-uri = /
ping-response-code-rules = +2??

We used the curl command built-in to the admin CLI to run the above to the junction server.  For example,

isva> tools
isva:tools> curl -v -s -k -X HEAD https://backendserver:port

We saw that a 500 was returned.  The backend did not like the HEAD command or accessing using /.  We added a one-off health check like,

[junction:/jct]
ping-method = GET
ping-uri = /appname
ping-response-code-rules = +2??

This resolved the issue and the server always shows running as it should be.

Given your problem description this sounds like the exact same issue and you'll need to figure out a proper health check.

Hi Scott,

Thank you for your answer,
I have an application exposed by RH Openshift via a route like http://myapp.myopenshit.com/myapp.

I created a trasparent junction like this:
Server task mywebseal create -t tcp -h myapp.myopenshit.com -p 80 -v myapp.myopenshit.com -x /myapp

If I try to see the status of this new junction via the command:
Server task mywebseal show /myapp
I can see that the resource is online but the server status is not running as you can see from the following image.

If I then call my junction from my portal (https://my.websealdomain.it/myapp) it works fine and the server status goes from 'not running' to 'running' and then returns to a 'not running' status after some time.

If I may add one more piece of information if I try to connect to the same endpoint configured on the junction it responds correctly.

What do i have to configure on  webseal side or on RH Openshift side to make the junction always running?

where is my error?

Regards,

Fabio

Hello,
I have set up a junction to my openshift platform.
I have a strange behaviour, the junction results down but if I call the junction the application behind work fine and at that moment the junction is up, only to come back down after some time.

before invoking the junction 

After invoking the junction 

what do i need to configure to have the junction always up and running?

Thank you for your answer Nick.
I tried applying your solution on one of my webseals, but now it no longer starts even though I removed the configuration you recommended.
In the message log of the webseal I have this error:
HPDCO0156E Configuration item missing (junction, basicauth-dummy-passwd).

any ideas?

Hi Fabio,

I just helped a customer with this exact same issue.  Check the ping-uri in the WRP config file.  For example,

[junction]
ping-method = HEAD
ping-uri = /
ping-response-code-rules = +2??

We used the curl command built-in to the admin CLI to run the above to the junction server.  For example,

isva> tools
isva:tools> curl -v -s -k -X HEAD https://backendserver:port

We saw that a 500 was returned.  The backend did not like the HEAD command or accessing using /.  We added a one-off health check like,

[junction:/jct]
ping-method = GET
ping-uri = /appname
ping-response-code-rules = +2??

This resolved the issue and the server always shows running as it should be.

Given your problem description this sounds like the exact same issue and you'll need to figure out a proper health check.

Hi Scott,

Thank you for your answer,
I have an application exposed by RH Openshift via a route like http://myapp.myopenshit.com/myapp.

I created a trasparent junction like this:
Server task mywebseal create -t tcp -h myapp.myopenshit.com -p 80 -v myapp.myopenshit.com -x /myapp

If I try to see the status of this new junction via the command:
Server task mywebseal show /myapp
I can see that the resource is online but the server status is not running as you can see from the following image.

If I then call my junction from my portal (https://my.websealdomain.it/myapp) it works fine and the server status goes from 'not running' to 'running' and then returns to a 'not running' status after some time.

If I may add one more piece of information if I try to connect to the same endpoint configured on the junction it responds correctly.