Subject: Re: Resin Robustness - msg#00135
List: text.xml.resin.user
Rob Lockstone wrote:
Running Resin 2.1.10 on Win2K machines with Sun's JDK 1.4.1_03 and IIS
in front.
Servers are very busy for about 8 hours per day, dual CPU's averaging
about 75%. Handling millions of (mostly small) requests per day.
Because of relatively frequent code updates, resin usually gets
restarted every few days. But on those rare occasions when we manage to
go for a week or more without restarting, things can start to get
unstable and Resin just stops responding to servlet requests. Restarting
fixes everything up and it's fine. I don't see any (obvious) memory or
threading issues. And I'm willing to live with how it is. I'm just
curious if anyone else experiences this.
We've run Resin for months without Resin causing problems. Sometimes we
do get memory or thread issues, but they always end up being app
programs. We're starting to adopt httpunit on all projects more heavily
to help stress test and replicate these seemingly tough-to-diagnose
problems.
--
Serge Knystautas
President
Lokitech >> software . strategy . design >>
http://www.lokitech.com/
p. 1.301.656.5501
e. sergek@xxxxxxxxxxxx
Was this page helpful?
Thread at a glance:
Previous Message by Date:
click to view message preview
Resin Robustness
Running Resin 2.1.10 on Win2K machines with Sun's JDK 1.4.1_03 and IIS
in front.
Servers are very busy for about 8 hours per day, dual CPU's averaging
about 75%. Handling millions of (mostly small) requests per day.
Because of relatively frequent code updates, resin usually gets
restarted every few days. But on those rare occasions when we manage to
go for a week or more without restarting, things can start to get
unstable and Resin just stops responding to servlet requests.
Restarting fixes everything up and it's fine. I don't see any (obvious)
memory or threading issues. And I'm willing to live with how it is. I'm
just curious if anyone else experiences this.
Rob
Next Message by Date:
click to view message preview
Problems with a Java-Application running under Resin 2.1.6
Hi,
right now i am at a loss.
our departement has been developed a java-application (which is connecting to
another database-server [oracle 9i] outside the DMZ) under the following
premises:
- Suse Linux 6.4
- Unify ServletExec (jsp-engine)
- jdk 1.3.0
our production server is working in our DMZ and got the following
infrastructure:
- Sun Solaris 5.8
- Apache 1.3.27 (SSL)
- Resin 2.1.6
- jdk 1.3.1_03b
so, we brought the application from the development-server to our
production-server. after starting Resin it is possible to work for round about
one minute with this application, after this timelimit we do not get an
response. after restarting Resin the problem is gone (but unfortunately only
for one minute =8-( ). we got no problem with the database connection or with
any firewalls (we already checked this).
Funny, because if we connect to the application either with Resin as standalone
or with Apache(SSL)+Tomcat the application is working fine.
Anyone know what is causing this to happen?
Markus
Previous Message by Thread:
click to view message preview
Resin Robustness
Running Resin 2.1.10 on Win2K machines with Sun's JDK 1.4.1_03 and IIS
in front.
Servers are very busy for about 8 hours per day, dual CPU's averaging
about 75%. Handling millions of (mostly small) requests per day.
Because of relatively frequent code updates, resin usually gets
restarted every few days. But on those rare occasions when we manage to
go for a week or more without restarting, things can start to get
unstable and Resin just stops responding to servlet requests.
Restarting fixes everything up and it's fine. I don't see any (obvious)
memory or threading issues. And I'm willing to live with how it is. I'm
just curious if anyone else experiences this.
Rob
Next Message by Thread:
click to view message preview
Re: Resin Robustness
Serge Knystautas wrote:
Rob Lockstone wrote:
Running Resin 2.1.10 on Win2K machines with Sun's JDK 1.4.1_03 and
IIS in front.
Servers are very busy for about 8 hours per day, dual CPU's averaging
about 75%. Handling millions of (mostly small) requests per day.
Because of relatively frequent code updates, resin usually gets
restarted every few days. But on those rare occasions when we manage
to go for a week or more without restarting, things can start to get
unstable and Resin just stops responding to servlet requests.
Restarting fixes everything up and it's fine. I don't see any
(obvious) memory or threading issues. And I'm willing to live with
how it is. I'm just curious if anyone else experiences this.
We've run Resin for months without Resin causing problems. Sometimes
we do get memory or thread issues, but they always end up being app
programs. We're starting to adopt httpunit on all projects more
heavily to help stress test and replicate these seemingly
tough-to-diagnose problems.
That's generally been most people's experience: most of the
threading/deadlock issues have been application code. You need to be
able to get and understand thread dumps.
However, Resin does have bugs, so sometimes the deadlock is Resin's
fault. You really need a thread dump to have any idea of where the
cause is. So if you get a thread dump that looks like it's a
Resin-caused deadlock, send it to one of the support email addresses.
-- Scott