CHE-513 Better handle exceptions when ws agent pings Che server #376
Comments
|
Fixed with #368 |
ghost
closed this as
Katka92
pushed a commit
to Katka92/che
that referenced
this issue
Jun 27, 2018
…ipse-che#417) * Adapt to the last upstream changes * Make Bayesian plugin work with multi-tenancy (issue eclipse-che#376) by: - creating a service endpoint started on wsmaster along with the Bayesian Agent, to provide the required up-to-date token (from the user that started the workspace) from the machine token of the workspace agent - retrieving this token endpoint when starting the language server on the workspace agent. Signed-off-by: David Festal <dfestal@redhat.com>
This issue was closed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
This is an autogenerated issue.
Original: https://jira.codenvycorp.com/browse/CHE-513
Reporter: Eugene Ivantsov
Description: See #319 to get a stack trace.
There's no diagnostic info there at all, like connection refused, connection timeout, malformed url, bad request etc.
This checker should handle exceptions in such a way so that it is well understood why the connection failed. We should also stop ws-agent then, since it produces tons of logs.
It will be also helpful to print /etc/hosts file since this information will help in diagnosis and possible solutions.
We should also print the following error:
"The workspace agent has attempted to start, but it is unable to ping the Che server at apiEndpoint
The workspace agent has been forcefully stopped. This error happens when the agent cannot resolve the location of the Che server. This error can usually be fixed with additional configuration settings in /conf/che.properties. The Che server will stop this workspace after a short timeout. You can get help by posting your config, stacktrace and workspace /etc/hosts below as a GitHub issue"
The text was updated successfully, but these errors were encountered: