[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim updated JBIDE-10484: ------------------------------------- Attachment: could-not-configure.png -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim edited comment on JBIDE-10484 at 12/14/11 9:39 AM: -------------------------------------------------------------------- The OpenShift java client library throws a Bad request exception with the following detail message: {code} {"debug":"","data":null,"api_c":["placeholder"],"result":"A jenkins application named 'jenkins' in namespace 'jbosstools' already exists. You can only have 1 jenkins application per account.","messages":"","broker":"1.1.1","broker_c":["namespace","rhlogin","ssh","app_uuid","debug","alter","cartridge","cart_type","action","app_name","api"],"exit_code":115,"api":"1.1.1"} {code} So detail informations, that would explain what happened, are present but not displayed yet. We have to fix that so that the things get more obvious to the user. We can still further enhance the user experience on top of that (ex. by automatically selecting the mongodb cartridge if the user selects the rockmongo cartridge. ex. by explicitly telling the user that he may not embed jenkins if he already has 4 apps since jenkins would need 2 application slots). was (Author: adietish): The OpenShift java client library throws a Bad request exception with the following detail message: {quote} {"debug":"","data":null,"api_c":["placeholder"],"result":"A jenkins application named 'jenkins' in namespace 'jbosstools' already exists. You can only have 1 jenkins application per account.","messages":"","broker":"1.1.1","broker_c":["namespace","rhlogin","ssh","app_uuid","debug","alter","cartridge","cart_type","action","app_name","api"],"exit_code":115,"api":"1.1.1"} {quote} So detail informations, that would explain what happened, are present but not displayed yet. We have to fix that so that the things get more obvious to the user. We can still further enhance the user experience on top of that (ex. by automatically selecting the mongodb cartridge if the user selects the rockmongo cartridge. ex. by explicitly telling the user that he may not embed jenkins if he already has 4 apps since jenkins would need 2 application slots). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim edited comment on JBIDE-10484 at 2/17/12 12:20 PM: -------------------------------------------------------------------- The problem resides in the openshift-java-client. See when for instance the error occurs when creating an application and you reached the quota limit: the server responds with: {quote} {"broker":"1.1.1","data":"","broker_c":["namespace","rhlogin","ssh","app_uuid","debug","alter","cartridge","cart_type","action","app_name","api"],"exit_code":1,"api_c":["placeholder"],"api":"1.1.2","messages":"OpenShift authentication will be performing maintenance on Saturday Feb 18th from 9pm til 3am.\nApplications and sites will not go down, but you will not be able to do anything that requires logging in.\n","debug":"","result":"adietish@redhat.com has already reached the application limit of 5"} {quote} The client library will throw a BadRequestException which occurs here - catch(HttpClientException): OpenShiftService: {code} private String sendRequest(final String request, final String url, final String password, final String authKey, final String authIV, final String errorMessage) throws OpenShiftException { try { String requestMessage = new OpenShiftEnvelopeFactory(password, authKey, authIV, request).createString(); String response = createHttpClient(id, url, this.doSSLChecks).post(requestMessage); return JsonSanitizer.sanitize(response); } catch (MalformedURLException e) { throw new OpenShiftException(e, errorMessage); } catch (UnauthorizedException e) { throw new InvalidCredentialsOpenShiftException(url, e); } catch (NotFoundException e) { throw new NotFoundOpenShiftException(url, e); } catch (HttpClientException e) { throw new OpenShiftEndpointException(url, e, errorMessage); } } {code} The error message is given by the calling method here: OpenShiftService: {code} protected IApplication requestApplicationAction(final ApplicationRequest applicationRequest, final IUser user) throws OpenShiftException { String url = applicationRequest.getUrlString(getServiceUrl()); String request = new ApplicationRequestJsonMarshaller().marshall(applicationRequest); String response = sendRequest(request, url, user.getPassword(), user.getAuthKey(), user.getAuthIV(), MessageFormat.format("Could not {0} application \"{1}\" at \"{2}\"", applicationRequest.getAction().getCommand(), applicationRequest.getName(), url)); OpenShiftResponse<IApplication> openshiftResponse = new ApplicationResponseUnmarshaller(applicationRequest.getName(), applicationRequest.getCartridge(), user, this).unmarshall(response); return openshiftResponse.getOpenShiftObject(); } {code} The reason for the failure is present in the HttpClientException that is nested into the OpenShiftEndpointException. The former would contain the full response message though which is not handy for reporting. We will have to make sure the response is unmarshalled so that one could extract the reason from OpenShiftResponse#getResult. was (Author: adietish): The problem resides in the openshift-java-client. See when for instance the error occurs when creating an application and you reached the quota limit: the server responds with: {quote} {"broker":"1.1.1","data":"","broker_c":["namespace","rhlogin","ssh","app_uuid","debug","alter","cartridge","cart_type","action","app_name","api"],"exit_code":1,"api_c":["placeholder"],"api":"1.1.2","messages":"OpenShift authentication will be performing maintenance on Saturday Feb 18th from 9pm til 3am.\nApplications and sites will not go down, but you will not be able to do anything that requires logging in.\n","debug":"","result":"adietish@redhat.com has already reached the application limit of 5"} {quote} The client library will throw a BadRequestException which occurs here - catch(HttpClientException): OpenShiftService: {code} private String sendRequest(final String request, final String url, final String password, final String authKey, final String authIV, final String errorMessage) throws OpenShiftException { try { String requestMessage = new OpenShiftEnvelopeFactory(password, authKey, authIV, request).createString(); String response = createHttpClient(id, url, this.doSSLChecks).post(requestMessage); return JsonSanitizer.sanitize(response); } catch (MalformedURLException e) { throw new OpenShiftException(e, errorMessage); } catch (UnauthorizedException e) { throw new InvalidCredentialsOpenShiftException(url, e); } catch (NotFoundException e) { throw new NotFoundOpenShiftException(url, e); } catch (HttpClientException e) { throw new OpenShiftEndpointException(url, e, errorMessage); } } {code} The error message is given by the calling method here: OpenShiftService: {code} protected IApplication requestApplicationAction(final ApplicationRequest applicationRequest, final IUser user) throws OpenShiftException { String url = applicationRequest.getUrlString(getServiceUrl()); String request = new ApplicationRequestJsonMarshaller().marshall(applicationRequest); String response = sendRequest(request, url, user.getPassword(), user.getAuthKey(), user.getAuthIV(), MessageFormat.format("Could not {0} application \"{1}\" at \"{2}\"", applicationRequest.getAction().getCommand(), applicationRequest.getName(), url)); OpenShiftResponse<IApplication> openshiftResponse = new ApplicationResponseUnmarshaller(applicationRequest.getName(), applicationRequest.getCartridge(), user, this).unmarshall(response); return openshiftResponse.getOpenShiftObject(); } {code} The reason for the failure would be actually present in OpenShiftResponse#getResult. We'd have to make sure the exception that is thrown includes the reason. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim commented on JBIDE-10484: ------------------------------------------ made the following changes in the client library to report error details: https://github.com/bdecoste/openshift-java-client/commit/ce54ca23e9b8cce6... https://github.com/bdecoste/openshift-java-client/commit/12662af4324e28d5... -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim updated JBIDE-10484: ------------------------------------- Attachment: report-details-2.patch -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Max Rydahl Andersen updated JBIDE-10484: ---------------------------------------- Assignee: Andre Dietisheim (was: Max Rydahl Andersen) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Andre Dietisheim commented on JBIDE-10484: ------------------------------------------ the TODO is intentional since the wizard is not aware of unexpected changes in the user instance. We currently have no property change support in the user class since this is a client library class. The real fix would be to inject/weave it in but we worked around it so far and will do that later on. In the result the wizard is in an invalid state if a timeout execption while waiting for an app to become available happens: the app was created but the wizard ran into a timeout, so the application name text field should now validate false since the app was created, the name's alreday sued. We'd either have to tell the wizard to refresh its user instance or inject the property support and listen for changes. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBIDE-10484?page=com.atlassian.jira.plugi... ] Stefan Bunciak closed JBIDE-10484. ---------------------------------- Verified on JBoss Tools 3.3.0.Beta1e -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira