Cannot deploy ApexClass on Save

[Stu Forbes]

Hi,

I've just updated to version 1.5.2, and since updating, can no longer deploy to Salesforce. The progress bar appears at the bottom of the editor, but completes far quicker than usual, and the file is not deployed. However, there is no error message displayed in IntelliJ

I've attached a log file snippet. It appears that a NullPointerException is thrown on save.

Regards, Stu Forbes

[Scott Wells]

Stu, sorry for the issue. I'm looking into this now and wondered if you might try something for me. Could you rebuild caches & indices using Settings>Illuminated Cloud (connectionName)>Rebuild, restart IDEA, and see if the problem persists? Let me know...

[Stu Forbes]

Exceptions when attempting to rebuild cache and indices

[Stu Forbes]

Hi Scott, I tried rebuilding the cache and indices, but that too throws an exception, and IntelliJ becomes unresponsive. I've attached the log snippet from this operation also.

Thanks, Stu

[Scott Wells]

Stu, the exceptions that you showed (except for one that I'm addressing now) are normal. When you ask IntelliJ IDEA to rebuild indices, it logs things like:

2015-09-06 14:38:06,011 [ 667097] INFO - il.indexing.FileBasedIndexImpl - Rebuild requested for index CaseInsensitiveFilenameIndex java.lang.Throwable: java.lang.Throwable: Rebuilding CaseInsensitiveFilenameIndex at com.intellij.util.indexing.FileBasedIndexImpl.scheduleRebuild(FileBasedIndexImpl.java:1214)

The one that's not normal and that I'm addressing for the next patch is:

2015-09-06 14:38:22,822 [ 683908] ERROR - il.indexing.FileBasedIndexImpl - Problems during index shutdown java.lang.Throwable: Do not log ProcessCanceledException at com.intellij.idea.IdeaLogger.error(IdeaLogger.java:124) at com.intellij.openapi.diagnostic.Logger.error(Logger.java:130) at com.intellij.util.indexing.FileBasedIndexImpl.performShutdown(FileBasedIndexImpl.java:678)

You should see that either idea.exe or idea64.exe is still running. Kill that process and start it up again. You should see things index at startup. Can you see if the same issue occurs after doing that?

If it does, please send me any relevant info from the log after the restart. If necessary, you can return to 1.5.1.0 using this link:

https://plugins.jetbrains.com/plugin/7831?pr=&showAllUpdates=true

Just download the build for the appropriate version of IntelliJ and use Settings>Plugins>Install plugin from disk... and I'll keep trying to isolate what might be going on there.

I sincerely apologize for the inconvenience!

[Stu Forbes]

Thanks for getting back to me. I have rebuilt the cache and indices now, but am still not able to deploy my code.

Strangely though, I no longer see the NullPointerException in the log file. I have tried restarting IntelliJ, and they are still not showing up.

I'll attach a log snippet of the log file after a restart. FYI I'm running IntelliJ on OSX.

Many thanks for providing the link to version 1.5.1.0. If the issue hasn't been resolved by tomorrow morning, I will revert to that.

Regards, Stu

[Stu Forbes]

Log snippet from IntelliJ startup

[Scott Wells]

That startup log looks clean which is good. Can you also verify that the class(es) you're trying to deploy are properly selected for your module? If not, they won't be deployed when you save to the local file system.

[Stu Forbes]

Hi Scott, I uninstalled the plugin last night, and manually installed version 1.5.1.0, and the files started deploying to SF again. This morning, out of curiosity, I updated the plugin back to 1.5.2.0, and bizarrely, this is now working too. I know this isn't ideal, as these kinds of problems can be a nightmare to solve.

I'll keep an eye out for any problems, and will update you if I find anything. For now, shall I mark this bug 'resolved'?

Thanks, Stu

[Scott Wells]

Yeah, let's go ahead and resolve it. Obviously there's something here, but as you pointed out, this one is difficult enough to debug remotely when it's happening!

I'm going to be working a set of enhancments for deployment and retrieval soon. I'll make sure to supplement the debug logging in that area so that if this did start to occur again, we could enable some key logging and get a better idea of what's happening.

Thanks for working through it!