RFR(XS):JDK-8219228: java/util/Base64/TestEncodingDecodingLength.java failing on 8GB test machine.
Hello! I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe. Could someone please review this minimal change? Bug: https://bugs.openjdk.java.net/browse/JDK-8219228 Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/ Thanks and best regards, Arno
Hi Arno, this looks good and trivial. I'll sponsor this one. Maybe one of the authors/reviewers of the original fix for bug https://bugs.openjdk.java.net/browse/JDK-8218265 (Nishit, Roger, Naoto) should approve this. Thanks Christoph
-----Original Message----- From: core-libs-dev <core-libs-dev-bounces@openjdk.java.net> On Behalf Of Zeller, Arno Sent: Montag, 18. Februar 2019 13:49 To: core-libs-dev <core-libs-dev@openjdk.java.net> Subject: [CAUTION] RFR(XS):JDK-8219228: java/util/Base64/TestEncodingDecodingLength.java failing on 8GB test machine.
Hello!
I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe.
Could someone please review this minimal change?
Bug: https://bugs.openjdk.java.net/browse/JDK-8219228
Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/
Thanks and best regards, Arno
Hi Arno, Although I don't know if turning off or no swap is expected for a test environment, but tried reproducing the issue locally on a 8Gb Ubuntu linux VM with swap turned off, the test case was passing until I started some other app (like browser) in parallel, in which case it failed with error = "Not enough space". Since it is a memory intensive test, I think it is better and safe to increase "os.maxMemory" check Change looks good to me, I am not an openJDK reviewer though. Regards, Nishit Jain On 18-02-2019 18:19, Zeller, Arno wrote:
Hello!
I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe.
Could someone please review this minimal change?
Bug: https://bugs.openjdk.java.net/browse/JDK-8219228
Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/
Thanks and best regards, Arno
Hi Nishit, thanks for testing! Best regards, Arno
-----Original Message----- From: Nishit Jain <nishit.jain@oracle.com> Sent: Dienstag, 19. Februar 2019 10:05 To: Zeller, Arno <arno.zeller@sap.com>; core-libs-dev <core-libs- dev@openjdk.java.net> Subject: Re: RFR(XS):JDK-8219228: java/util/Base64/TestEncodingDecodingLength.java failing on 8GB test machine.
Hi Arno,
Although I don't know if turning off or no swap is expected for a test environment, but tried reproducing the issue locally on a 8Gb Ubuntu linux VM with swap turned off, the test case was passing until I started some other app (like browser) in parallel, in which case it failed with error = "Not enough space". Since it is a memory intensive test, I think it is better and safe to increase "os.maxMemory" check
Change looks good to me, I am not an openJDK reviewer though.
Regards, Nishit Jain On 18-02-2019 18:19, Zeller, Arno wrote:
Hello!
I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe.
Could someone please review this minimal change?
Bug: https://bugs.openjdk.java.net/browse/JDK-8219228
Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/
Thanks and best regards, Arno
Looks fine, Reviewed. On 02/19/2019 04:05 AM, Nishit Jain wrote:
Hi Arno,
Although I don't know if turning off or no swap is expected for a test environment, but tried reproducing the issue locally on a 8Gb Ubuntu linux VM with swap turned off, the test case was passing until I started some other app (like browser) in parallel, in which case it failed with error = "Not enough space". Since it is a memory intensive test, I think it is better and safe to increase "os.maxMemory" check
Change looks good to me, I am not an openJDK reviewer though.
Regards, Nishit Jain On 18-02-2019 18:19, Zeller, Arno wrote:
Hello!
I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe.
Could someone please review this minimal change?
Bug: https://bugs.openjdk.java.net/browse/JDK-8219228
Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/
Thanks and best regards, Arno
+1 Naoto On 2/19/19 7:10 AM, Roger Riggs wrote:
Looks fine, Reviewed.
On 02/19/2019 04:05 AM, Nishit Jain wrote:
Hi Arno,
Although I don't know if turning off or no swap is expected for a test environment, but tried reproducing the issue locally on a 8Gb Ubuntu linux VM with swap turned off, the test case was passing until I started some other app (like browser) in parallel, in which case it failed with error = "Not enough space". Since it is a memory intensive test, I think it is better and safe to increase "os.maxMemory" check
Change looks good to me, I am not an openJDK reviewer though.
Regards, Nishit Jain On 18-02-2019 18:19, Zeller, Arno wrote:
Hello!
I found that the test java/util/Base64/TestEncodingDecodingLength.java fails on a machine with 8GB memory after JDK-8218265. The test starts a VM with -Xmx8GB but the VM needs some more memory than only the heap and on machines with just 8GB of memory (and no swap configured) the test will always fail because the VM cannot get enough native memory. Therefore I suggest to increase "@requires os.maxMemory" to >= 10GB to be safe.
Could someone please review this minimal change?
Bug: https://bugs.openjdk.java.net/browse/JDK-8219228
Webrev: http://cr.openjdk.java.net/~azeller/webrevs/8219228/
Thanks and best regards, Arno
participants (5)
-
Langer, Christoph
-
Naoto Sato
-
Nishit Jain
-
Roger Riggs
-
Zeller, Arno