Open asfimport opened 11 years ago
Troels Liebe Bentsen (migrated from Bugzilla): Should note that this is on Windows 8
Sebb (migrated from Bugzilla): Here is a stack dump during the pause:
"Thread Group 1-1" prio=6 tid=0x038f3c00 nid=0xd80 runnable [0x03b7f000] java.lang.Thread.State: RUNNABLE at java.net.Inet4AddressImpl.getHostByAddr(Native Method) at java.net.InetAddress$1.getHostByAddr(Unknown Source) at java.net.InetAddress.getHostFromNameService(Unknown Source) at java.net.InetAddress.getHostName(Unknown Source) at java.net.InetAddress.getHostName(Unknown Source) at sun.security.ssl.SSLSocketImpl.getHost(Unknown Source)
This will need further investigation.
Sebb (migrated from Bugzilla): The stack dump was on WinXP using JMeter 2.8.
@pmouawad (migrated from Bugzilla): Hello, What java version are you using ?
Regards
Troels Liebe Bentsen (migrated from Bugzilla): C:\Users\TroelsLiebeBentsen>java -version java version "1.7.0_11" Java(TM) SE Runtime Environment (build 1.7.0_11-b21) Java HotSpot(TM) 64-Bit Server VM (build 23.6-b04, mixed mode)
Sebb (migrated from Bugzilla): (In reply to comment 3)
The stack dump was on WinXP using JMeter 2.8.
2013/01/18 16:55:19 INFO - jmeter.JMeter: Version 2.8 r1393162 2013/01/18 16:55:19 INFO - jmeter.JMeter: java.version=1.7.0_11 2013/01/18 16:55:19 INFO - jmeter.JMeter: java.vm.name=Java HotSpot(TM) Client VM
Sebb (migrated from Bugzilla): Using:
2013/01/18 17:25:52 INFO - jmeter.JMeter: Version 2.8 r1393162 2013/01/18 17:25:52 INFO - jmeter.JMeter: java.version=1.6.0_38 2013/01/18 17:25:52 INFO - jmeter.JMeter: java.vm.name=Java HotSpot(TM) Client VM 2013/01/18 17:25:52 INFO - jmeter.JMeter: os.name=Windows XP
I get similar results:
"Thread Group 1-1" prio=6 tid=0x03c4c400 nid=0x27c runnable [0x055ff000] java.lang.Thread.State: RUNNABLE at java.net.Inet4AddressImpl.getHostByAddr(Native Method) at java.net.InetAddress$1.getHostByAddr(InetAddress.java:852) at java.net.InetAddress.getHostFromNameService(InetAddress.java:532) at java.net.InetAddress.getHostName(InetAddress.java:475) at java.net.InetAddress.getHostName(InetAddress.java:447) at com.sun.net.ssl.internal.ssl.SSLSocketImpl.getHost(SSLSocketImpl.java:1971)
@pmouawad (migrated from Bugzilla): I don't reproduce on Mac OSx: java version "1.6.0_37" Java(TM) SE Runtime Environment (build 1.6.0_37-b06-434-10M3909) Java HotSpot(TM) 64-Bit Server VM (build 20.12-b01-434, mixed mode)
@sebb do you use the plan as is or do you point
@Troels Liebe Bentsen Is the IP in the test plan public ?
Thanks
Sebb (migrated from Bugzilla): (In reply to comment 8)
I don't reproduce on Mac OSx: java version "1.6.0_37" Java(TM) SE Runtime Environment (build 1.6.0_37-b06-434-10M3909) Java HotSpot(TM) 64-Bit Server VM (build 20.12-b01-434, mixed mode)
@sebb do you use the plan as is or do you point
Yes (except I unchecked "Follow redirects" - not relevant here)
@Troels Liebe Bentsen Is the IP in the test plan public ?
Thanks
@pmouawad (migrated from Bugzilla): I reproduce on Windows 7 but not on Mac OS. Seems to affect Windows Family only.
Could be something related to this: http://www.velocityreviews.com/forums/showpost.php?p=2959030&postcount=8
Sebb (migrated from Bugzilla): However the problem only affects HC4, not the HC3.1 or Java implementations.
So if it is a Windows issue, why is it being triggered by HC4 only?
I think it may be something to do with the way SSL verification is handled in HC4. Even though we use the HC4TrustAllSSLSocketFactory, it still invokes org.apache.http.conn.ssl.AbstractVerifier.verify(AbstractVerifier.java:91) which calls getSession, even though the session is not needed later.
I've not yet had a chance to try recoding to avoid this; if it is the cause then I think we may need to raise an enhancement request against HC4.
@pmouawad (migrated from Bugzilla): I thought that:
Hitesh (migrated from Bugzilla): Created attachment jmeter: Jmeter file for Mac OS x attached
Sebb (migrated from Bugzilla): (In reply to comment 13)
Created attachment 30151 [details] Jmeter file for Mac OS x attached
Seems to be for a different bug report. Was this intended for https://github.com/apache/jmeter/issues/3092 ?
@pmouawad (migrated from Bugzilla): Yes it's related to 54799
Dzmitry Kashlach (migrated from Bugzilla): Failed to reproduce on Xubuntu 12.04 java version "1.6.0_45" Java(TM) SE Runtime Environment (build 1.6.0_45-b06) Java HotSpot(TM) Server VM (build 20.45-b01, mixed mode)
@pmouawad (migrated from Bugzilla): Could this be an explanation:
@pmouawad (migrated from Bugzilla): Anybody could test this bug on Windows again to see if it's still in 3.0 or current nightly ? Thanks
Troels Liebe Bentsen (Bug 54449): When using HTTPClient4 with HTTPS, requests hangs for 4-5 seconds for each request when using IP's without reverse DNS, using the DNS names or adding the IP and DNS name to hosts file removes the delay. Also using HTTPClient3.1 instead seems to solve the issue.
I have added a plan that shows the issue.
Regards Troels
Created attachment http-hang-bug.jmx: Jmeter plan that shows the issue
http-hang-bug.jmx
````xmltrue
Votes in Bugzilla: 2 Severity: minor OS: Windows XP