Bug 1433262

Created attachment 1263974 [details]
Archive created with "zip"

Created attachment 1263975 [details]
Archive created with Windows internal zip tool

Taking this one.

Replicated on both OpenJDK 7 & 8:

$ /usr/lib/jvm/icedtea-7/bin/java -cp . TestZip 8192m.zip
Failed  to inspect packed file
java.util.zip.ZipException: invalid CEN header (bad signature)
	at java.util.zip.ZipFile.open(Native Method)
	at java.util.zip.ZipFile.<init>(ZipFile.java:221)
	at java.util.zip.ZipFile.<init>(ZipFile.java:151)
	at java.util.zip.ZipFile.<init>(ZipFile.java:165)
	at TestZip.main(TestZip.java:16)

$ /usr/lib/jvm/icedtea-8/bin/java -cp . TestZip 8192m.zip
Failed  to inspect packed file
java.util.zip.ZipException: invalid CEN header (bad signature)
	at java.util.zip.ZipFile.open(Native Method)
	at java.util.zip.ZipFile.<init>(ZipFile.java:225)
	at java.util.zip.ZipFile.<init>(ZipFile.java:155)
	at java.util.zip.ZipFile.<init>(ZipFile.java:169)
	at TestZip.main(TestZip.java:16)

though, just for amusement value, it does pass on GNU Classpath ;-)

$ cacao -cp . TestZip 8192m.zip
 -> 8192m
Succeeded  to inspect packed file

I'll look into what's going on with the OpenJDK code.

(In reply to Andrew John Hughes from comment #10)
> Replicated on both OpenJDK 7 & 8:
> 
> $ /usr/lib/jvm/icedtea-7/bin/java -cp . TestZip 8192m.zip
> Failed  to inspect packed file

[ ... ]

> 
> $ cacao -cp . TestZip 8192m.zip
>  -> 8192m
> Succeeded  to inspect packed file
> 
> I'll look into what's going on with the OpenJDK code.

Did you find the cause Andrew?

It passes on OpenJDK 11:

$ /usr/lib/openjdk-11/bin/java -cp . TestZip 8192m.zip
 -> 8192m
Succeeded  to inspect packed file

So we just need to find the right fix to backport.

Looks like it might be https://bugs.openjdk.java.net/browse/JDK-8186464 but need to confirm.

Hi Andrew,

It passes on OpenJDK 11 (tested on RHEL8) only for the ZIP file created using reproducer.
However, with the Windows ZIP file in attachment to the BZ, it continues to fail:

-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------
# /usr/lib/jvm/java-11-openjdk-11.0.1.13-6.el8.x86_64/bin/java -cp . TestZip 8192m.windows.zip 
Failed  to inspect packed file
java.util.zip.ZipException: invalid CEN header (bad compression method: 9)
	at java.base/java.util.zip.ZipFile$Source.zerror(ZipFile.java:1529)
	at java.base/java.util.zip.ZipFile$Source.initCEN(ZipFile.java:1497)
	at java.base/java.util.zip.ZipFile$Source.<init>(ZipFile.java:1268)
	at java.base/java.util.zip.ZipFile$Source.get(ZipFile.java:1231)
	at java.base/java.util.zip.ZipFile$CleanableResource.<init>(ZipFile.java:726)
	at java.base/java.util.zip.ZipFile$CleanableResource.get(ZipFile.java:843)
	at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:246)
	at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:176)
	at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:190)
	at TestZip.main(TestZip.java:16)
-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

See also clone for RHEL8 / openjdk-11: BZ1660159

reproducer added to our internal suite

Moving to RHEL 7.8, but we'll aim to get it in 8u232 in October if it is that fix.

(In reply to Andrew John Hughes from comment #16)
> Moving to RHEL 7.8, but we'll aim to get it in 8u232 in October if it is
> that fix.

Not sure if you are aware about similar issue related to Zip 64 support broken on Java 11:
https://access.redhat.com/support/cases/#/case/02408490 (aka https://bugs.openjdk.java.net/browse/JDK-8226530).

Just in case you touch same code, make sure it will not re-introduce other bug from Java 11 on Java 8.

Moving to rhel-7.9.0 as rhel-7.8 now requires exception+ or blocker+. We aim to get this upstream first anyway, so it should appear on all supported releases in due course.

Drop from 7.8 RPL.

JDK-8186464 looks the likely fix for this, as stated above, but as it stands, it depends on a reimplementation of the zip code (https://bugs.openjdk.java.net/browse/JDK-8145260). We're currently looking at the feasibility of backporting this.

Will backport JDK-8186464 without JDK-8145260 by porting it to the native implementation.

Andrew, the attached test program still fails with both attached examples on 1.8.0_262-b10 build (java-1.8.0-openjdk-1.8.0.262.b10-0.el7_8.x86_64).

How that matches with "fixed in java-1.8.0-openjdk-1.8.0.262.b05-0.4.ea.el7" ?

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (java-1.8.0-openjdk bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2020:3920

(In reply to Andrey Loskutov from comment #28)
> Andrew, the attached test program still fails with both attached examples on
> 1.8.0_262-b10 build (java-1.8.0-openjdk-1.8.0.262.b10-0.el7_8.x86_64).
> 
> How that matches with "fixed in java-1.8.0-openjdk-1.8.0.262.b05-0.4.ea.el7"
> ?

The fix is in RHEL 7.9, not RHEL 7.8.

(In reply to Andrew John Hughes from comment #30)
> (In reply to Andrey Loskutov from comment #28)
> > Andrew, the attached test program still fails with both attached examples on
> > 1.8.0_262-b10 build (java-1.8.0-openjdk-1.8.0.262.b10-0.el7_8.x86_64).
> > 
> > How that matches with "fixed in java-1.8.0-openjdk-1.8.0.262.b05-0.4.ea.el7"
> > ?
> 
> The fix is in RHEL 7.9, not RHEL 7.8.

Could you please update "fixed in" version? 
Which exact JDK 8 rpm version contains the fix?

The fixed in version is correct. That's the first RHEL 7.9 build the fix went into.

I've updated the public target release to make it clear which RHEL version has the fix.

Also should be clear from the advisory: https://access.redhat.com/errata/RHBA-2020:3920