You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cocoon.apache.org by br...@bjohnson.net on 2004/05/21 00:00:06 UTC
Broken Pipes w/ Production Site
I'm running a production site using Tomcat 5.x and Cocoon 2.1.2. I'm
planning on upgrading to 2.1.4 later next week. But today I got a call from
the boss asking me what was wrong with the website. I checked it.. and I
was getting no response. No errors to the browser.. it just wasnt
responding.
I restarted and everything cleared up. But I had a look at the logs around
the time of the site failure and the only exception I could find around that
time was a SocketException for a Broken Pipe. It appears numerous times and
I'm not sure why.
I saw some threads about this in the archive but no real answers. Could
this cause our site not to respond? I MUST get this fixed asap (its a
production intranet site for a government contract).
Anyone else using Tomcat for production?
Thanks! Here's the full exception...
2004-05-20 13:10:07 StandardWrapperValve[Cocoon]: Servlet.service() for
servlet Cocoon threw exception
ClientAbortException: java.net.SocketException: Broken pipe
at
org.apache.catalina.connector.ResponseBase.flushBuffer(ResponseBase.java:680
)
at
org.apache.catalina.connector.HttpResponseBase.doFlushBuffer(HttpResponseBas
e.java:795)
at
org.apache.catalina.connector.HttpResponseBase.flushBuffer(HttpResponseBase.
java:785)
at
org.apache.catalina.connector.ResponseBase.write(ResponseBase.java:648)
at
org.apache.catalina.connector.ResponseStream.write(ResponseStream.java:313)
at
org.apache.catalina.connector.http.HttpResponseStream.write(HttpResponseStre
am.java:183)
at
org.apache.catalina.connector.ResponseStream.write(ResponseStream.java:287)
at
org.apache.cocoon.components.notification.Notifier.notifyHTML(Notifier.java:
144)
at
org.apache.cocoon.components.notification.Notifier.notify(Notifier.java:98)
at
org.apache.cocoon.servlet.CocoonServlet.manageException(CocoonServlet.java:1
257)
at
org.apache.cocoon.servlet.CocoonServlet.service(CocoonServlet.java:1150)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
at
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(Application
FilterChain.java:247)
at
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterCh
ain.java:193)
at
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.ja
va:256)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:643)
at
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:480)
at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:995)
at
org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.ja
va:191)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:643)
at
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:480)
at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:995)
at
org.apache.catalina.core.StandardContext.invoke(StandardContext.java:2416)
at
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:180
)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:643)
at
org.apache.catalina.valves.ErrorDispatcherValve.invoke(ErrorDispatcherValve.
java:171)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:641)
at
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:172
)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:641)
at
org.apache.catalina.valves.AccessLogValve.invoke(AccessLogValve.java:577)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:641)
at
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:480)
at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:995)
at
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java
:174)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:643)
at
org.apache.catalina.valves.AccessLogValve.invoke(AccessLogValve.java:577)
at
org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invok
eNext(StandardPipeline.java:641)
at
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:480)
at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:995)
at
org.apache.catalina.connector.http.HttpProcessor.process(HttpProcessor.java:
1040)
at
org.apache.catalina.connector.http.HttpProcessor.run(HttpProcessor.java:1151
)
at java.lang.Thread.run(Thread.java:534)
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
RE: Broken Pipes w/ Production Site
Posted by Upayavira <uv...@upaya.co.uk>.
On Fri, 21 May 2004 13:03:43 -0400, brent@bjohnson.net said:
> Good idea.. I'll check that software out. I used to use something
> called NetSaint back in the day.. dont know if its still around
> or still as hard to configure as it used to be.
>
Hah. I think Nagios used to be called NetSaint. There is a new
configuration system. Don't know if it is easier.
Upayavira
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
RE: Broken Pipes w/ Production Site
Posted by Askild Aaberg Olsen <as...@xangeli.com>.
No replacement for Nagios, but www.siteprobe.com offers a free 1-hour
interval service for checking the status of your server. I guess that there
are several other services like this on the net.
This can be a good supplement to Nagios, and you get it up and running in no
time.
Askild
-
brent@bjohnson.net wrote:
> Good idea.. I'll check that software out. I used to use
> something called NetSaint back in the day.. dont know if its
> still around or still as hard to configure as it used to be.
>
> > Not an answer to your immediate problem, but...
> >
> > Have you got any monitoring on your server?
> >
> > I've started using Nagios (www.nagios.org) an OS monitoring
> > tool, Unix based, and it is great. (Up until now, all my
> > monitoring was hand crafted). You can set it up to test your
> > server every five minutes (say), and send you an
> > email/text/pager message whenever it (a) doesn't respond (b)
> > doesn't give a response you'd expect (e.g. page must contain
> > a specific phrase).
> >
> > With something like this, at least you can be sure you'll
> > find out before your boss!
> >
> > HTH. Upayavira
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
RE: Broken Pipes w/ Production Site
Posted by br...@bjohnson.net.
Good idea.. I'll check that software out. I used to use something
called NetSaint back in the day.. dont know if its still around
or still as hard to configure as it used to be.
Thanks,
- Brent
> -----Original Message-----
> From: Upayavira [mailto:uv@upaya.co.uk]
> Sent: Friday, May 21, 2004 4:24 AM
> To: users@cocoon.apache.org
> Subject: Re: Broken Pipes w/ Production Site
>
> brent@bjohnson.net wrote:
>
> >Thanks - Ill try upgrading to 2.1.4 and see if that helps.
> >Yeah I wasnt sure whether those exceptions were the problem
> (since they
> >do appear in other parts of the log file) but that was all I found
> >around the time the problem occurred.
> >
> >I also increased the pool size in the cocooon.xconf file
> after having
> >problems with Cocoon/Tomcat returning blank pages (as in
> actual blank
> ><body> tags). I've been working on a side project called Webgate
> >(webgate.sourceforge.net) and I had the blank page problem and
> >increasing pool sizes seemed to fix that.. so I tried it on the
> >production site and that seems to have worked (so far).
> >
> >But this new issue with the whole webserver no longer responding has
> >started to worry me quite a bit. It's not good to have your
> boss (the
> >owner of the company) call me because the website isnt responding :(
> >
> >
> Not an answer to your immediate problem, but...
>
> Have you got any monitoring on your server?
>
> I've started using Nagios (www.nagios.org) an OS monitoring
> tool, Unix based, and it is great. (Up until now, all my
> monitoring was hand crafted). You can set it up to test your
> server every five minutes (say), and send you an
> email/text/pager message whenever it (a) doesn't respond (b)
> doesn't give a response you'd expect (e.g. page must contain
> a specific phrase).
>
> With something like this, at least you can be sure you'll
> find out before your boss!
>
> HTH. Upayavira
>
> >Luckily restarting seemed to fix it for now.
> >
> >This seems to only happen after the site has been running
> for a month
> >or so.. so a memory leak may definitely be the problem. I'll try
> >upgrading.. and also setting up a new cron job to reboot the server
> >every couple weeks just in case.
> >
> >Thanks,
> >
> >- Brent
> >
> >
> >
> >>-----Original Message-----
> >>From: Askild Aaberg Olsen [mailto:askild@xangeli.com]
> >>Sent: Thursday, May 20, 2004 6:28 PM
> >>To: users@cocoon.apache.org
> >>Subject: RE: Broken Pipes w/ Production Site
> >>
> >>It seems like the client aborted the connection, like when the user
> >>pressed "Stop" in the browser when the page loaded.
> >>I'm using Jetty, but I get a similar message here.
> >>
> >>Try that yourself, by pressing "Stop" when the page loads,
> and see if
> >>you get the same message.
> >>
> >>If this is the case, I don't know if this is the cause of your
> >>problem.
> >>
> >>There were some memory-leaks in Cocoon versions before 2.1.4.
> >>Maybe this is the cause?
> >>Then upgrading would certanly help.
> >>
> >>Have been running 2.1.4 for 40 days straight now, but it's
> not a wery
> >>high level of traffic on my sites, though.
> >>
> >>Playing with the pools in the sitemap and the store-settings in
> >>cocoon.xconf helped me a lot in terms of performance,
> especialy since
> >>my site seemed to respond slower and slower before i did these
> >>changes.
> >>
> >>Not much, but hopefully some of the more experienced users
> will have
> >>other ideas when they wake up in the morning (The
> Cocoon-lists seems
> >>to follow European times in terms of activity ;)
> >>
> >>Good luck!
> >>
> >>Askild
> >>-
> >>
> >>
> >>------------------------------------------------------------
> ---------
> >>To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
> >>For additional commands, e-mail: users-help@cocoon.apache.org
> >>
> >>
> >>
> >>
> >>
> >>
> >
> >
> >
> >---------------------------------------------------------------------
> >To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
> >For additional commands, e-mail: users-help@cocoon.apache.org
> >
> >
> >
> >
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
> For additional commands, e-mail: users-help@cocoon.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
Re: Broken Pipes w/ Production Site
Posted by Upayavira <uv...@upaya.co.uk>.
brent@bjohnson.net wrote:
>Thanks - Ill try upgrading to 2.1.4 and see if that helps.
>Yeah I wasnt sure whether those exceptions were the problem
>(since they do appear in other parts of the log file) but
>that was all I found around the time the problem occurred.
>
>I also increased the pool size in the cocooon.xconf
>file after having problems with Cocoon/Tomcat returning
>blank pages (as in actual blank <body> tags). I've been
>working on a side project called Webgate (webgate.sourceforge.net)
>and I had the blank page problem and increasing pool sizes
>seemed to fix that.. so I tried it on the production site
>and that seems to have worked (so far).
>
>But this new issue with the whole webserver no longer
>responding has started to worry me quite a bit. It's
>not good to have your boss (the owner of the company)
>call me because the website isnt responding :(
>
>
Not an answer to your immediate problem, but...
Have you got any monitoring on your server?
I've started using Nagios (www.nagios.org) an OS monitoring tool, Unix
based, and it is great. (Up until now, all my monitoring was hand
crafted). You can set it up to test your server every five minutes
(say), and send you an email/text/pager message whenever it (a) doesn't
respond (b) doesn't give a response you'd expect (e.g. page must contain
a specific phrase).
With something like this, at least you can be sure you'll find out
before your boss!
HTH. Upayavira
>Luckily restarting seemed to fix it for now.
>
>This seems to only happen after the site has been
>running for a month or so.. so a memory leak may
>definitely be the problem. I'll try upgrading.. and
>also setting up a new cron job to reboot the server
>every couple weeks just in case.
>
>Thanks,
>
>- Brent
>
>
>
>>-----Original Message-----
>>From: Askild Aaberg Olsen [mailto:askild@xangeli.com]
>>Sent: Thursday, May 20, 2004 6:28 PM
>>To: users@cocoon.apache.org
>>Subject: RE: Broken Pipes w/ Production Site
>>
>>It seems like the client aborted the connection, like when
>>the user pressed "Stop" in the browser when the page loaded.
>>I'm using Jetty, but I get a similar message here.
>>
>>Try that yourself, by pressing "Stop" when the page loads,
>>and see if you get the same message.
>>
>>If this is the case, I don't know if this is the cause of
>>your problem.
>>
>>There were some memory-leaks in Cocoon versions before 2.1.4.
>>Maybe this is the cause?
>>Then upgrading would certanly help.
>>
>>Have been running 2.1.4 for 40 days straight now, but it's
>>not a wery high level of traffic on my sites, though.
>>
>>Playing with the pools in the sitemap and the store-settings
>>in cocoon.xconf helped me a lot in terms of performance,
>>especialy since my site seemed to respond slower and slower
>>before i did these changes.
>>
>>Not much, but hopefully some of the more experienced users
>>will have other ideas when they wake up in the morning (The
>>Cocoon-lists seems to follow European times in terms of activity ;)
>>
>>Good luck!
>>
>>Askild
>>-
>>
>>
>>---------------------------------------------------------------------
>>To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
>>For additional commands, e-mail: users-help@cocoon.apache.org
>>
>>
>>
>>
>>
>>
>
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
>For additional commands, e-mail: users-help@cocoon.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
RE: Broken Pipes w/ Production Site
Posted by br...@bjohnson.net.
Thanks - Ill try upgrading to 2.1.4 and see if that helps.
Yeah I wasnt sure whether those exceptions were the problem
(since they do appear in other parts of the log file) but
that was all I found around the time the problem occurred.
I also increased the pool size in the cocooon.xconf
file after having problems with Cocoon/Tomcat returning
blank pages (as in actual blank <body> tags). I've been
working on a side project called Webgate (webgate.sourceforge.net)
and I had the blank page problem and increasing pool sizes
seemed to fix that.. so I tried it on the production site
and that seems to have worked (so far).
But this new issue with the whole webserver no longer
responding has started to worry me quite a bit. It's
not good to have your boss (the owner of the company)
call me because the website isnt responding :(
Luckily restarting seemed to fix it for now.
This seems to only happen after the site has been
running for a month or so.. so a memory leak may
definitely be the problem. I'll try upgrading.. and
also setting up a new cron job to reboot the server
every couple weeks just in case.
Thanks,
- Brent
> -----Original Message-----
> From: Askild Aaberg Olsen [mailto:askild@xangeli.com]
> Sent: Thursday, May 20, 2004 6:28 PM
> To: users@cocoon.apache.org
> Subject: RE: Broken Pipes w/ Production Site
>
> It seems like the client aborted the connection, like when
> the user pressed "Stop" in the browser when the page loaded.
> I'm using Jetty, but I get a similar message here.
>
> Try that yourself, by pressing "Stop" when the page loads,
> and see if you get the same message.
>
> If this is the case, I don't know if this is the cause of
> your problem.
>
> There were some memory-leaks in Cocoon versions before 2.1.4.
> Maybe this is the cause?
> Then upgrading would certanly help.
>
> Have been running 2.1.4 for 40 days straight now, but it's
> not a wery high level of traffic on my sites, though.
>
> Playing with the pools in the sitemap and the store-settings
> in cocoon.xconf helped me a lot in terms of performance,
> especialy since my site seemed to respond slower and slower
> before i did these changes.
>
> Not much, but hopefully some of the more experienced users
> will have other ideas when they wake up in the morning (The
> Cocoon-lists seems to follow European times in terms of activity ;)
>
> Good luck!
>
> Askild
> -
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
> For additional commands, e-mail: users-help@cocoon.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org
RE: Broken Pipes w/ Production Site
Posted by Askild Aaberg Olsen <as...@xangeli.com>.
brent@bjohnson.net :
> I'm running a production site using Tomcat 5.x and Cocoon
> 2.1.2. I'm planning on upgrading to 2.1.4 later next week.
> But today I got a call from the boss asking me what was wrong
> with the website. I checked it.. and I was getting no
> response. No errors to the browser.. it just wasnt responding.
>
> I restarted and everything cleared up. But I had a look at
> the logs around the time of the site failure and the only
> exception I could find around that time was a SocketException
> for a Broken Pipe. It appears numerous times and I'm not sure why.
>
> I saw some threads about this in the archive but no real
> answers. Could this cause our site not to respond? I MUST
> get this fixed asap (its a production intranet site for a
> government contract).
>
> Anyone else using Tomcat for production?
>
> Thanks! Here's the full exception...
>
> 2004-05-20 13:10:07 StandardWrapperValve[Cocoon]:
> Servlet.service() for servlet Cocoon threw exception
> ClientAbortException: java.net.SocketException: Broken pipe
It seems like the client aborted the connection, like when the user pressed
"Stop" in the browser when the page loaded. I'm using Jetty, but I get a
similar message here.
Try that yourself, by pressing "Stop" when the page loads, and see if you
get the same message.
If this is the case, I don't know if this is the cause of your problem.
There were some memory-leaks in Cocoon versions before 2.1.4. Maybe this is
the cause?
Then upgrading would certanly help.
Have been running 2.1.4 for 40 days straight now, but it's not a wery high
level of traffic on my sites, though.
Playing with the pools in the sitemap and the store-settings in cocoon.xconf
helped me a lot in terms of performance, especialy since my site seemed to
respond slower and slower before i did these changes.
Not much, but hopefully some of the more experienced users will have other
ideas when they wake up in the morning (The Cocoon-lists seems to follow
European times in terms of activity ;)
Good luck!
Askild
-
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org