[JBoss JIRA] (JBIDE-26086) SSP server does not send error codes for bad requests
by Rob Stryker (JIRA)
[ https://issues.jboss.org/browse/JBIDE-26086?page=com.atlassian.jira.plugi... ]
Rob Stryker updated JBIDE-26086:
--------------------------------
Comment: was deleted
(was: [~jrichter1] Can you please craft a curl or http request that I can use to replicate and dig deeper on this? As currently written, I'm not able to act on or investigate this issue. )
> SSP server does not send error codes for bad requests
> -----------------------------------------------------
>
> Key: JBIDE-26086
> URL: https://issues.jboss.org/browse/JBIDE-26086
> Project: Tools (JBoss Tools)
> Issue Type: Bug
> Components: stack-server-protocol
> Affects Versions: 4.6.0.AM3
> Reporter: Jan Richter
> Assignee: Rob Stryker
>
> Sending a request the server does not like produces no response, it is simply logged on the server side and the client is not notified of what happened. Something like a code 400 would be appreciated.
> For example, initiating a connection with no content-length header simply logs the following on the server side, but the client is stuck waiting.
> {noformat}java.lang.IllegalStateException: Missing header Content-Length in input "GET / HTTP/1.1
> Sec-WebSocket-Version: 13
> Sec-WebSocket-Key: 4vrbTo0+BlSk5wK7si6SWQ==
> Connection: Upgrade
> Upgrade: websocket
> Sec-WebSocket-Extensions: permessage-deflate; client_max_window_bits
> Host: localhost:27511
> "
> at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.listen(StreamMessageProducer.java:89)
> at org.eclipse.lsp4j.jsonrpc.json.ConcurrentMessageProcessor.run(ConcurrentMessageProcessor.java:95)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748){noformat}
> This is most likely a 'feature' of lsp4j, thoughts?
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
7 years, 10 months
[JBoss JIRA] (JBIDE-26088) arbitrary folder should not be able to be searched for runtimes
by Rob Stryker (JIRA)
[ https://issues.jboss.org/browse/JBIDE-26088?page=com.atlassian.jira.plugi... ]
Rob Stryker commented on JBIDE-26088:
-------------------------------------
branch with wip is here: https://github.com/robstryker/org.jboss.tools.ssp/tree/JBIDE-26088
> arbitrary folder should not be able to be searched for runtimes
> ---------------------------------------------------------------
>
> Key: JBIDE-26088
> URL: https://issues.jboss.org/browse/JBIDE-26088
> Project: Tools (JBoss Tools)
> Issue Type: Bug
> Components: stack-server-protocol
> Reporter: Rob Stryker
>
> Server currently has following interface / json entry point:
> CompletableFuture<List<ServerBean>> findServerBeans(DiscoveryPath path);
> However, any arbitrary system path can be searched this way. It'd be better if any discovery path had to be kept in the model via addDiscoveryPath first, to ensure that the server isn't somehow restricted from searching various paths.
> Of course right now there's no API or configuration file to actually restrict scanning various folders, but I would imagine that eventually such restrictions would be made such that the void addDiscoveryPath(DiscoveryPath path); entrypoint could restrict or reject additions.
> I believe [~dgolovin] disagrees with this approach. Please discuss.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
7 years, 10 months
[JBoss JIRA] (JBIDE-26088) arbitrary folder should not be able to be searched for runtimes
by Rob Stryker (JIRA)
Rob Stryker created JBIDE-26088:
-----------------------------------
Summary: arbitrary folder should not be able to be searched for runtimes
Key: JBIDE-26088
URL: https://issues.jboss.org/browse/JBIDE-26088
Project: Tools (JBoss Tools)
Issue Type: Bug
Components: stack-server-protocol
Reporter: Rob Stryker
Server currently has following interface / json entry point:
CompletableFuture<List<ServerBean>> findServerBeans(DiscoveryPath path);
However, any arbitrary system path can be searched this way. It'd be better if any discovery path had to be kept in the model via addDiscoveryPath first, to ensure that the server isn't somehow restricted from searching various paths.
Of course right now there's no API or configuration file to actually restrict scanning various folders, but I would imagine that eventually such restrictions would be made such that the void addDiscoveryPath(DiscoveryPath path); entrypoint could restrict or reject additions.
I believe [~dgolovin] disagrees with this approach. Please discuss.
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
7 years, 10 months
[JBoss JIRA] (JBIDE-26086) SSP server does not send error codes for bad requests
by Rob Stryker (JIRA)
[ https://issues.jboss.org/browse/JBIDE-26086?page=com.atlassian.jira.plugi... ]
Rob Stryker commented on JBIDE-26086:
-------------------------------------
[~jrichter1] Can you please craft a curl or http request that I can use to replicate and dig deeper on this? As currently written, I'm not able to act on or investigate this issue.
> SSP server does not send error codes for bad requests
> -----------------------------------------------------
>
> Key: JBIDE-26086
> URL: https://issues.jboss.org/browse/JBIDE-26086
> Project: Tools (JBoss Tools)
> Issue Type: Bug
> Components: stack-server-protocol
> Affects Versions: 4.6.0.AM3
> Reporter: Jan Richter
> Assignee: Rob Stryker
>
> Sending a request the server does not like produces no response, it is simply logged on the server side and the client is not notified of what happened. Something like a code 400 would be appreciated.
> For example, initiating a connection with no content-length header simply logs the following on the server side, but the client is stuck waiting.
> {noformat}java.lang.IllegalStateException: Missing header Content-Length in input "GET / HTTP/1.1
> Sec-WebSocket-Version: 13
> Sec-WebSocket-Key: 4vrbTo0+BlSk5wK7si6SWQ==
> Connection: Upgrade
> Upgrade: websocket
> Sec-WebSocket-Extensions: permessage-deflate; client_max_window_bits
> Host: localhost:27511
> "
> at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.listen(StreamMessageProducer.java:89)
> at org.eclipse.lsp4j.jsonrpc.json.ConcurrentMessageProcessor.run(ConcurrentMessageProcessor.java:95)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748){noformat}
> This is most likely a 'feature' of lsp4j, thoughts?
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
7 years, 10 months
[JBoss JIRA] (JBIDE-26084) Explorer: pods are show twice if they're controlled by an rc (no dc, no service)
by Andre Dietisheim (JIRA)
[ https://issues.jboss.org/browse/JBIDE-26084?page=com.atlassian.jira.plugi... ]
Andre Dietisheim updated JBIDE-26084:
-------------------------------------
Description:
steps:
# EXEC: Create a new application via the following custom template [^eap-without-service-nor-deploymentconfig.json]
# EXEC: look into OpenShift explorer and see where the pods for the application are shown
Result:
They're listed twice, once as pods on their own and a 2nd time as children for the replication controller (that the template creates)
!image-2018-06-06-10-33-55-241.png!
Also see the discussion in the PR Also see discussion for the corresponding [PR|https://github.com/jbosstools/jbosstools-openshift/pull/1748] for JBIDE-24979
was:
steps:
# EXEC: Create a new application via the following custom template [^eap-without-service-nor-deploymentconfig.json]
# EXEC: look into OpenShift explorer and see where the pods for the application are shown
Result:
They're listed twice, once as pods on their own and a 2nd time as children for the replication controller (that the template creates)
!image-2018-06-06-10-33-55-241.png!
> Explorer: pods are show twice if they're controlled by an rc (no dc, no service)
> --------------------------------------------------------------------------------
>
> Key: JBIDE-26084
> URL: https://issues.jboss.org/browse/JBIDE-26084
> Project: Tools (JBoss Tools)
> Issue Type: Bug
> Components: openshift
> Affects Versions: 4.6.0.AM2
> Reporter: Andre Dietisheim
> Labels: explorer
> Fix For: 4.6.x
>
> Attachments: eap-without-service-nor-deploymentconfig.json, image-2018-06-06-10-33-55-241.png
>
>
> steps:
> # EXEC: Create a new application via the following custom template [^eap-without-service-nor-deploymentconfig.json]
> # EXEC: look into OpenShift explorer and see where the pods for the application are shown
> Result:
> They're listed twice, once as pods on their own and a 2nd time as children for the replication controller (that the template creates)
> !image-2018-06-06-10-33-55-241.png!
> Also see the discussion in the PR Also see discussion for the corresponding [PR|https://github.com/jbosstools/jbosstools-openshift/pull/1748] for JBIDE-24979
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)
7 years, 10 months