Open ghost opened 10 years ago
I've been having this issue too. I've just noticed this thread where claus suggests that a 5 minute outage can be caused by the container re-provisioning itself - https://github.com/fabric8io/fabric8/issues/1017. My guess is CXF is modifying the jetty settings and that's causing a re-provision ... although that doesn't explain why this only happens in openshift.
Hi, We have noticed an issue with the fuse openshift cartridge, where having created a gear, we are then trying to deploy a cxf based webservice, we lose the fabric commands for around four minutes where upon they recover.
I have isolated the steps to repeat:
Checking the logs there appear to be problems where fabric cannot connect to its internal GIT repo, which seems to use jetty? This is closely followed by a number of key bundles being uninstalled one of which includes the fabric commands. This happens on our corporate openshift enterprise environment, also happens on redhat's online service. Does not happen on straight jboss-fuse 6.1 installed locally.
Please advise if we should be using a different profile rather than feature-cxf? Ultimately we can live with this, since the services appear to redeploy after about 4mins and then the commands are returned and the service is deployed. Ideally however we would like to be able to use the fabric commands from the console directly.
thanks
Martin