I have repeated the rapid click test after I updated the
connectionTimeout=600000, still getting the -53 and also -54 errors. This
time I have set log levels to debug. I'll send you a separate e-mail where
you can download them when you are back on Friday. Again thank you for your
help so far.

J.Neuhoff




Rainer Jung-3 wrote:
>
> JNeuhoff schrieb:
>>> connection timeout on mod_jk side is in seconds, on tomcat side is in
>>> milliseconds. So Mladens suggestion had a missing trailing 0 to make the
>>> params on the two sides fit. I think he meant connectionTimeout=600000
>>> to make it fit the 600 on the mod_jk side.
>>
>> Thanks, you are right, there was a missing zero. I'll repeat the test
>> tomorrow with the correct value.
>
> I don't expect that to fix the -53 problem, but the behaviour of the
> timeouts will be more consistent.
>
>>
>> Your explanation with the threads agrees with my observations.
>>
>>> Finally: from your last statement I'm not sure that I understand, what
>>> the real problems are now:
>>
>> It is not a memory leak problem, even though with the growing number of
>> threads it can grow to over 150MB, but it then stays at that level.
>>
>> The problem are the err=-53 conditions. While both Apache2 and Tomcat
>> continue to run and receive and respond to incoming requests, Apache2 is
>> unable to receive Tomcat's reponses (err=-53). Also, though Apache2
>> continues to serve other static pages just fine according to its
>> access.log,
>> the client web browser client never receives the responses. Instead, it
>> complaints about failed connections. The only static pages the web
>> browser
>> was able to receive was the standard 503 error page (Service Unavailable)
>> when attempting to access the Tomcat web service. I ended up re-starting
>> our
>> live server twice during the last 24 hours.
>
> Aha. Maybe there are two different things, the 53 problem and the fact,
> that your clients are not able to retrieve even static content, although
> apache thinks it has delivered the content. For the second problem it
> could be interesting to check, if this observation holds true, even if
> you try to receive a static page from a client not included in your
> click test.
>
> Do I remember correctly, that Apache and Tomcat are on the same machine?
> Is there a firewall on this machine?
>
>>
>>> If you can easily reproduce, a mod_jk log with
>>> JkLogLevel debug would be helpful.
>>
>> I'll do that tomorrow.
>
> The file might get big. I could download from a web site or similar.
> I'll be away for two days, so don't expect an answer before friday.
>
>>
>>> mod_jk error 53: the only remaining problem. Does it imply a problem on
>>> the user side?
>>
>> I had our sysadmin check the hardware and network equipment today, and he
>> also examined the Windows 2003 event logs. He couldn't find anything
>> abnormal.
>>
>> J.Neuhoff
>>
>
> ---------------------------------------------------------------------
> To start a new topic, e-mail: users@(protected)
> To unsubscribe, e-mail: users-unsubscribe@(protected)
> For additional commands, e-mail: users-help@(protected)
>
>
>

--
Sent from the Tomcat - User mailing list archive at Nabble.com.


---------------------------------------------------------------------
To start a new topic, e-mail: users@(protected)
To unsubscribe, e-mail: users-unsubscribe@(protected)
For additional commands, e-mail: users-help@(protected)