Community
Participate
Working Groups
1. Clone repo git.eclipse.org/gitroot/platform/eclipse.platform.ui.git using "master" initial branch 2. Import all projects 3. Close some projects, say, "org.eclipse.ui.carbon", "org.eclipse.ui.cocoa" 4. Team -> Reset, select Remote Tracking -> origin/R3_6_maintenance; "Hard" 5. NPE exception resulting is what seems to be mixed result (some files on master, some on maintenance branch): An internal error occurred during: "Resetting to refs/remotes/origin/R3_6_maintenance". java.lang.NullPointerException java.lang.NullPointerException at org.eclipse.jgit.dircache.DirCacheCheckout.processEntry(DirCacheCheckout.java:358) at org.eclipse.jgit.dircache.DirCacheCheckout.prescanOneTree(DirCacheCheckout.java:282) at org.eclipse.jgit.dircache.DirCacheCheckout.checkout(DirCacheCheckout.java:382) at org.eclipse.egit.core.op.ResetOperation.checkoutIndex(ResetOperation.java:250) at org.eclipse.egit.core.op.ResetOperation.reset(ResetOperation.java:138) at org.eclipse.egit.core.op.ResetOperation.access$0(ResetOperation.java:115) at org.eclipse.egit.core.op.ResetOperation$1.run(ResetOperation.java:105) at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:2344) at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:2326) at org.eclipse.egit.core.op.ResetOperation.execute(ResetOperation.java:109) at org.eclipse.egit.ui.internal.job.JobUtil$1.run(JobUtil.java:50) at org.eclipse.core.internal.jobs.Worker.run(Worker.java:54)
I have a similar trace resetting CDT master java.lang.NullPointerException at org.eclipse.jgit.dircache.DirCacheCheckout.processEntry(DirCacheCheckout.java:358) at org.eclipse.jgit.dircache.DirCacheCheckout.prescanOneTree(DirCacheCheckout.java:282) at org.eclipse.jgit.dircache.DirCacheCheckout.doCheckout(DirCacheCheckout.java:392) at org.eclipse.jgit.dircache.DirCacheCheckout.checkout(DirCacheCheckout.java:379) at org.eclipse.egit.core.op.ResetOperation.checkoutIndex(ResetOperation.java:250) at org.eclipse.egit.core.op.ResetOperation.reset(ResetOperation.java:138) at org.eclipse.egit.core.op.ResetOperation.access$0(ResetOperation.java:115) at org.eclipse.egit.core.op.ResetOperation$1.run(ResetOperation.java:105) at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:2344) at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:2326) at org.eclipse.egit.core.op.ResetOperation.execute(ResetOperation.java:109) at org.eclipse.egit.ui.internal.job.JobUtil$1.run(JobUtil.java:50) at org.eclipse.core.internal.jobs.Worker.run(Worker.java:54)
Just got hit by this. Dupe of bug 347574?
Got it also while performing a rebase of a local branch. I'm on egit 1.1
I tried this with the current nightly version and couldn't reproduce this problem, could you also retry with latest nightly ?
I also ran into this. I could not find a workaround except deleting and cloning the whole large repo again. This is with 1.2.0.201110021920.
It looks like the trigger for this bug is a file deletion from which one cannot recover.
(In reply to comment #6) > It looks like the trigger for this bug is a file deletion from which one cannot > recover. This seems similar to the steps described by comment 347574.
(In reply to comment #5) > I also ran into this. I could not find a workaround except deleting and cloning > the whole large repo again. This is with 1.2.0.201110021920. I managed to recover by performing a hard reset from command line, then deleting and re-importing repo from Egit. I did lose some changes in my local branch though.
The NPE is clearly caused by a bug: i.getDirCacheEntry() is null if DirCacheBuildIterator i is positioned on a tree -> this causes NPE
Pushed http://egit.eclipse.org/r/4345 for review to prevent the NPE. I still couldn't reproduce the issue hence I am not completely sure if this will fix the root cause or just cure the symptom. I tried on Mac so far, maybe we only hit this on Windows so we will try to reproduce on Windows as a next step.
See http://egit.eclipse.org/r/#change,4357
I think this is a duplicate of 347574
This was identified as the same issue as in 347574. *** This bug has been marked as a duplicate of bug 347574 ***