date:20120206

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

2012-02-06 Thread Scott Ferguson


On 02/06/2012 10:54 AM, Bill Au wrote:

Scott,

What would trigger the JVM bug having to do with the IP addresses?  I 
want to make sure that I am not affected before I upgrade.


What version of JVM does that affect?


http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7078386

Looks like it's fixed in 1.6 b21 from the bug report.

Also, according to the bug report it's specific to Linux and certain 
IPv6 configurations.


It's not related to the 4.0.25 issue. Since Resin calls it once at 
startup, it would crash immediately.


-- Scott


Bill

On Mon, Feb 6, 2012 at 11:34 AM, Scott Ferguson > wrote:


Can you try the snapshot
http://caucho.com/download/resin-pro-4_0-snap.tar.gz?

The IP bug was a red-herring. There is a timing issue with
threading/clustering.

That snapshot is not suitable for production (several known
regression failures), but it can be used to verify the startup bug.

-- Scott


On 02/03/2012 01:45 AM, smallufo wrote:

Hi
This is my ip addresses :

$ ifconfig
eth0  Link encap:Ethernet  HWaddr 00:50:08:02:FA:8E
  inet addr:192.168.2.2  Bcast:192.168.2.255
 Mask:255.255.255.0
  inet6 addr: fe80::250:8ff:fe02:fa8e/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:856152 errors:0 dropped:0 overruns:0 frame:0
  TX packets:460283 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:122552727 (116.8 MiB)  TX bytes:284641238
(271.4 MiB)
  Interrupt:233 Base address:0xc000

eth1  Link encap:Ethernet  HWaddr 00:60:6E:60:07:4B
  inet addr:192.168.1.104  Bcast:192.168.1.255
 Mask:255.255.255.0
  inet6 addr: fe80::260:6eff:fe60:74b/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:11431817 errors:0 dropped:0 overruns:0 frame:0
  TX packets:13056593 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:3686915772 (3.4 GiB)  TX bytes:1152022111 (1.0
GiB)

loLink encap:Local Loopback
  inet addr:127.0.0.1  Mask:255.0.0.0
  inet6 addr: ::1/128 Scope:Host
  UP LOOPBACK RUNNING  MTU:16436  Metric:1
  RX packets:201560031 errors:0 dropped:0 overruns:0 frame:0
  TX packets:201560031 errors:0 dropped:0 overruns:0
carrier:0
  collisions:0 txqueuelen:0
  RX bytes:1693844049 (1.5 GiB)  TX bytes:1693844049 (1.5
GiB)

All are local addresses , no VM installed.


2012/2/3 Scott Ferguson mailto:f...@caucho.com>>

On 02/02/2012 06:53 PM, smallufo wrote:

I cannot operate my server when I start resin 4.0.25
The load is so high that the server almost hangs there...
Even a "ls" is so sluggish...
It happens even in a clean install (no other webapps
installed , just default resin.xml settings)



It might be related to a JVM bug having to do with the IP
addresses. What do you IP addresses look like? Are there any
virtual ones.

-- Scott




$ uname -a
Linux dev 2.6.18-164.6.1.el5 #1 SMP Tue Nov 3 16:18:27 EST
2009 i686 i686 i386 GNU/Linux

/var/log/resin $ head watchdog-manager.log
[2012/02/03 00:08:14.368] {main}
[2012/02/03 00:08:14.369] {main} Resin-4.0.25 (built Tue, 24
Jan 2012 11:09:58 PST)
[2012/02/03 00:08:14.370] {main}
[2012/02/03 00:08:14.370] {main} Linux 2.6.18-164.6.1.el5 i386
[2012/02/03 00:08:14.370] {main} Java(TM) SE Runtime
Environment 1.6.0_22-b04, UTF-8, zh
[2012/02/03 00:08:14.371] {main} Java HotSpot(TM) Server VM
17.1-b03, 32, mixed mode, Sun Microsystems Inc.




2012/2/3 Scott Ferguson mailto:f...@caucho.com>>

On 02/02/2012 08:12 AM, smallufo wrote:

When I execute 'resinctl start'  , I notice my server
almost hangs 
And the server load is very high :

$ w
 00:07:51 up 46 days,  6:07,  3 users,  load average:
526.31, 298.59, 121.04
USER TTY  FROM  LOGIN@   IDLE  
JCPU   PCPU WHAT
smallufo pts/1destiny  Wed166:09  
2.91s  1.92s top

smallufo pts/2destiny  Wed221.00s
 2.67s  0.01s w
smallufo pts/3destiny  Wed233:36   7:07
  6:56  /usr/java/jdk/bin/java -jar
/usr/local/share/resin-4.0.25/lib/resin.jar
-root-directory /hom

The load peaks to 500 or even higher !
After 5 or more minutes , the load goes down...


Can you get a thread dump?

Or better yet, if you have resin-pro, get a "resinctl

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

2012-02-06 Thread Bill Au

Scott,

What would trigger the JVM bug having to do with the IP addresses?  I want
to make sure that I am not affected before I upgrade.

What version of JVM does that affect?

Bill

On Mon, Feb 6, 2012 at 11:34 AM, Scott Ferguson  wrote:

> **
> Can you try the snapshot
> http://caucho.com/download/resin-pro-4_0-snap.tar.gz?
>
> The IP bug was a red-herring. There is a timing issue with
> threading/clustering.
>
> That snapshot is not suitable for production (several known regression
> failures), but it can be used to verify the startup bug.
>
> -- Scott
>
>
> On 02/03/2012 01:45 AM, smallufo wrote:
>
> Hi
> This is my ip addresses :
>
>  $ ifconfig
> eth0  Link encap:Ethernet  HWaddr 00:50:08:02:FA:8E
>   inet addr:192.168.2.2  Bcast:192.168.2.255  Mask:255.255.255.0
>   inet6 addr: fe80::250:8ff:fe02:fa8e/64 Scope:Link
>   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>   RX packets:856152 errors:0 dropped:0 overruns:0 frame:0
>TX packets:460283 errors:0 dropped:0 overruns:0 carrier:0
>   collisions:0 txqueuelen:1000
>   RX bytes:122552727 (116.8 MiB)  TX bytes:284641238 (271.4 MiB)
>   Interrupt:233 Base address:0xc000
>
>  eth1  Link encap:Ethernet  HWaddr 00:60:6E:60:07:4B
>   inet addr:192.168.1.104  Bcast:192.168.1.255  Mask:255.255.255.0
>   inet6 addr: fe80::260:6eff:fe60:74b/64 Scope:Link
>   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>   RX packets:11431817 errors:0 dropped:0 overruns:0 frame:0
>   TX packets:13056593 errors:0 dropped:0 overruns:0 carrier:0
>   collisions:0 txqueuelen:1000
>   RX bytes:3686915772 (3.4 GiB)  TX bytes:1152022111 (1.0 GiB)
>
>  loLink encap:Local Loopback
>   inet addr:127.0.0.1  Mask:255.0.0.0
>   inet6 addr: ::1/128 Scope:Host
>   UP LOOPBACK RUNNING  MTU:16436  Metric:1
>   RX packets:201560031 errors:0 dropped:0 overruns:0 frame:0
>   TX packets:201560031 errors:0 dropped:0 overruns:0 carrier:0
>   collisions:0 txqueuelen:0
>   RX bytes:1693844049 (1.5 GiB)  TX bytes:1693844049 (1.5 GiB)
>
>  All are local addresses , no VM installed.
>
>
>  2012/2/3 Scott Ferguson 
>
>>  On 02/02/2012 06:53 PM, smallufo wrote:
>>
>> I cannot operate my server when I start resin 4.0.25
>> The load is so high that the server almost hangs there...
>> Even a "ls" is so sluggish...
>> It happens even in a clean install (no other webapps installed , just
>> default resin.xml settings)
>>
>>
>>  It might be related to a JVM bug having to do with the IP addresses.
>> What do you IP addresses look like? Are there any virtual ones.
>>
>> -- Scott
>>
>>
>>
>>  $ uname -a
>> Linux dev 2.6.18-164.6.1.el5 #1 SMP Tue Nov 3 16:18:27 EST 2009 i686 i686
>> i386 GNU/Linux
>>
>>   /var/log/resin $ head watchdog-manager.log
>> [2012/02/03 00:08:14.368] {main}
>> [2012/02/03 00:08:14.369] {main} Resin-4.0.25 (built Tue, 24 Jan 2012
>> 11:09:58 PST)
>> [2012/02/03 00:08:14.370] {main}
>> [2012/02/03 00:08:14.370] {main} Linux 2.6.18-164.6.1.el5 i386
>> [2012/02/03 00:08:14.370] {main} Java(TM) SE Runtime Environment
>> 1.6.0_22-b04, UTF-8, zh
>> [2012/02/03 00:08:14.371] {main} Java HotSpot(TM) Server VM 17.1-b03, 32,
>> mixed mode, Sun Microsystems Inc.
>>
>>
>>
>>
>> 2012/2/3 Scott Ferguson 
>>
>>>   On 02/02/2012 08:12 AM, smallufo wrote:
>>>
>>> When I execute 'resinctl start'  , I notice my server almost hangs 
>>> And the server load is very high :
>>>
>>>  $ w
>>>  00:07:51 up 46 days,  6:07,  3 users,  load average: 526.31, 298.59,
>>> 121.04
>>> USER TTY  FROM  LOGIN@   IDLE   JCPU   PCPU WHAT
>>> smallufo pts/1destiny  Wed166:09   2.91s  1.92s top
>>> smallufo pts/2destiny  Wed221.00s  2.67s  0.01s w
>>> smallufo pts/3destiny  Wed233:36   7:07   6:56
>>>  /usr/java/jdk/bin/java -jar /usr/local/share/resin-4.0.25/lib/resin.jar
>>> -root-directory /hom
>>>
>>>  The load peaks to 500 or even higher !
>>> After 5 or more minutes , the load goes down...
>>>
>>>
>>>  Can you get a thread dump?
>>>
>>> Or better yet, if you have resin-pro, get a "resinctl pdf-report
>>> --profile".
>>>
>>> -- Scott
>>>
>>>
>>>  It does not happen in 4.0.19 !
>>>
>>>
>>> ___
>>> resin-interest mailing 
>>> listresin-interest@caucho.comhttp://maillist.caucho.com/mailman/listinfo/resin-interest
>>>
>>>
>>>
>>> ___
>>> resin-interest mailing list
>>> resin-interest@caucho.com
>>> http://maillist.caucho.com/mailman/listinfo/resin-interest
>>>
>>>
>>
>> ___
>> resin-interest mailing 
>> listresin-interest@caucho.comhttp://maillist.caucho.com/mailman/listinfo/resin-interest
>>
>>
>>
>> ___
>> resin-interest mailing list
>> resin-interest@caucho.com
>> http://maillist.cau

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

2012-02-06 Thread Scott Ferguson

Can you try the snapshot 
http://caucho.com/download/resin-pro-4_0-snap.tar.gz?


The IP bug was a red-herring. There is a timing issue with 
threading/clustering.


That snapshot is not suitable for production (several known regression 
failures), but it can be used to verify the startup bug.


-- Scott

On 02/03/2012 01:45 AM, smallufo wrote:

Hi
This is my ip addresses :

$ ifconfig
eth0  Link encap:Ethernet  HWaddr 00:50:08:02:FA:8E
  inet addr:192.168.2.2  Bcast:192.168.2.255  Mask:255.255.255.0
  inet6 addr: fe80::250:8ff:fe02:fa8e/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:856152 errors:0 dropped:0 overruns:0 frame:0
  TX packets:460283 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:122552727 (116.8 MiB)  TX bytes:284641238 (271.4 MiB)
  Interrupt:233 Base address:0xc000

eth1  Link encap:Ethernet  HWaddr 00:60:6E:60:07:4B
  inet addr:192.168.1.104  Bcast:192.168.1.255  Mask:255.255.255.0
  inet6 addr: fe80::260:6eff:fe60:74b/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:11431817 errors:0 dropped:0 overruns:0 frame:0
  TX packets:13056593 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:3686915772 (3.4 GiB)  TX bytes:1152022111 (1.0 GiB)

loLink encap:Local Loopback
  inet addr:127.0.0.1  Mask:255.0.0.0
  inet6 addr: ::1/128 Scope:Host
  UP LOOPBACK RUNNING  MTU:16436  Metric:1
  RX packets:201560031 errors:0 dropped:0 overruns:0 frame:0
  TX packets:201560031 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:0
  RX bytes:1693844049 (1.5 GiB)  TX bytes:1693844049 (1.5 GiB)

All are local addresses , no VM installed.


2012/2/3 Scott Ferguson mailto:f...@caucho.com>>

On 02/02/2012 06:53 PM, smallufo wrote:

I cannot operate my server when I start resin 4.0.25
The load is so high that the server almost hangs there...
Even a "ls" is so sluggish...
It happens even in a clean install (no other webapps installed ,
just default resin.xml settings)



It might be related to a JVM bug having to do with the IP
addresses. What do you IP addresses look like? Are there any
virtual ones.

-- Scott




$ uname -a
Linux dev 2.6.18-164.6.1.el5 #1 SMP Tue Nov 3 16:18:27 EST 2009
i686 i686 i386 GNU/Linux

/var/log/resin $ head watchdog-manager.log
[2012/02/03 00:08:14.368] {main}
[2012/02/03 00:08:14.369] {main} Resin-4.0.25 (built Tue, 24 Jan
2012 11:09:58 PST)
[2012/02/03 00:08:14.370] {main}
[2012/02/03 00:08:14.370] {main} Linux 2.6.18-164.6.1.el5 i386
[2012/02/03 00:08:14.370] {main} Java(TM) SE Runtime Environment
1.6.0_22-b04, UTF-8, zh
[2012/02/03 00:08:14.371] {main} Java HotSpot(TM) Server VM
17.1-b03, 32, mixed mode, Sun Microsystems Inc.




2012/2/3 Scott Ferguson mailto:f...@caucho.com>>

On 02/02/2012 08:12 AM, smallufo wrote:

When I execute 'resinctl start'  , I notice my server almost
hangs 
And the server load is very high :

$ w
 00:07:51 up 46 days,  6:07,  3 users,  load average:
526.31, 298.59, 121.04
USER TTY  FROM  LOGIN@   IDLE   JCPU  
PCPU WHAT

smallufo pts/1destiny  Wed166:09   2.91s
 1.92s top
smallufo pts/2destiny  Wed221.00s  2.67s
 0.01s w
smallufo pts/3destiny  Wed233:36   7:07  
6:56  /usr/java/jdk/bin/java -jar

/usr/local/share/resin-4.0.25/lib/resin.jar -root-directory /hom

The load peaks to 500 or even higher !
After 5 or more minutes , the load goes down...


Can you get a thread dump?

Or better yet, if you have resin-pro, get a "resinctl
pdf-report --profile".

-- Scott



It does not happen in 4.0.19 !


___
resin-interest mailing list
resin-interest@caucho.com  
http://maillist.caucho.com/mailman/listinfo/resin-interest



___
resin-interest mailing list
resin-interest@caucho.com 
http://maillist.caucho.com/mailman/listinfo/resin-interest



___
resin-interest mailing list
resin-interest@caucho.com  
http://maillist.caucho.com/mailman/listinfo/resin-interest



___
resin-interest mailing list
resin-interest@caucho.com 
http://maillist.caucho.com/mailman/listinfo/resin-interest



__

[Resin-interest] Encoding for javax.servlet.http.Part.getInputStream()

2012-02-06 Thread Mattias Jiderhamn

Hi. For multipart String parameters, Resin returns a UTF-8 encoded 
stream from javax.servlet.http.Part.getInputStream() without regards to 
request.getCharacterEncoding().

I cannot find whether the UTF-8 encoding is stipulated by the spec, or 
if this is a Resin bug.
Anyone knows...? Reference?

Thanks in advance.

-- 

   



___
resin-interest mailing list
resin-interest@caucho.com
http://maillist.caucho.com/mailman/listinfo/resin-interest

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

Re: [Resin-interest] [4.0.25] tremendously increases my server load when starts-up

[Resin-interest] Encoding for javax.servlet.http.Part.getInputStream()

4 matches

Site Navigation

Mail list logo

Footer information