Bug 316088 - jem.proxy showing up as invalidly signed
Summary: jem.proxy showing up as invalidly signed
Status: VERIFIED FIXED
Alias: None
Product: WTP Java EE Tools
Classification: WebTools
Component: jst.jem (show other bugs)
Version: 3.1   Edit
Hardware: PC Windows 7
: P3 blocker (vote)
Target Milestone: 3.2 RC4   Edit
Assignee: David Williams CLA
QA Contact: Chuck Bridgham CLA
URL:
Whiteboard: PMC_approved
Keywords:
Depends on:
Blocks:
 
Reported: 2010-06-08 00:57 EDT by David Williams CLA
Modified: 2010-06-10 12:31 EDT (History)
1 user (show)

See Also:
david_williams: pmc_approved+
david_williams: pmc_approved? (raghunathan.srinivasan)
david_williams: pmc_approved? (naci.dai)
deboer: pmc_approved+
neil.hauge: pmc_approved+
david_williams: pmc_approved? (kaloyan)

Attachments
patch to add eclipse.inf file. (416 bytes, patch)
2010-06-08 00:57 EDT, David Williams CLA 		no flags Details | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description David Williams CLA 2010-06-08 00:57:39 EDT 
Created attachment 171367 [details]
patch to add eclipse.inf file. 

I did a preliminary contribution to helios RC4 based on our most recent S build, and there was one error discovered: 


The following error occured when building Helios:

org.eclipse.core.runtime.CoreException: Unable to unpack artifact osgi.bundle,org.eclipse.jem.proxy,2.0.300.v200906292000 in repository file:/shared/helios/aggregation/final/aggregate: Invalid content:org/eclipse/jem/internal/proxy/remote/REMExpression$TryCatch.class
Caused by: org.eclipse.osgi.signedcontent.InvalidContentException: The file "org/eclipse/jem/internal/proxy/remote/REMExpression$TryCatch.class" in the jar "/tmp/signatureFile7932.jar" has been tampered!


I'm not sure why this has happened here, and now. This version is ours, right: 2.0.300.v200906292000? No one else using/contributing one? 

But, these do pop up from time to time, and the response is always "well, don't pack it ... there are just some code patterns that do not work well with the pack/unpack algorithm." 

If we had more time, I'm sure we'd enjoy investigating byte codes to the nth degree ... but, since no time, I suggest we add the "no pack" directives to the eclipse.inf file and avoid any risk. 

This could be related to the p2izing build change, since I'm not positive I am now correcting the right effort level (E4) which was always a recommended work around for some pack200 bug. 

Like I said, easiest just not to pack. Its pretty small anyway, if I recall. 
 
Is there a nested jar in this jar? Not sure, but in the patch I excluded whole jar, and nested jars ... just in case.
Comment 1 David Williams CLA 2010-06-08 01:13:07 EDT 
I've released based on my own recognizance (and neil's :)  (would not effect code in any way) so I can restart the midnight build, and be able to test result in the morning. In case we need yet some other fix. (which would be unlikely).
Comment 2 David Williams CLA 2010-06-08 01:14:01 EDT 
released (sorry carl, chuck for no time for review)
Comment 3 David Williams CLA 2010-06-08 01:50:12 EDT 
Hm, this is bad. I reverted to S-3.2.0RC3-20100529234347 and the error still occurred. that same build had been succeeding for several runs over past few days. Not good not good :( 
(that is, not sure what's causing it ... only thing I can think of now, is if someone else is incorrect contributing it :( 

Same message ... Unable to unpack artifact osgi.bundle,org.eclipse.jem.proxy,2.0.300.v200906292000
Comment 4 David Williams CLA 2010-06-08 02:00:41 EDT 
(In reply to comment #3)
> Hm, this is bad. I reverted to S-3.2.0RC3-20100529234347 and the error still
> occurred. ...

Oh, maybe not. someone else had triggered a build before I did ... I just assumed it had my revert in it, but apparently not. 

All is well ... relatively speaking.
Comment 5 David Williams CLA 2010-06-08 15:48:33 EDT 
I case I got anyone worried ... I have verified this via preliminary contribution to helios common site. Worked fine. Aren't byte codes and class file formats fun? :)