You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@turbine.apache.org by Ilan Azbel <il...@autochartist.com> on 2005/11/01 17:21:33 UTC
Turbine server keeps going down
Hello,
I have a Linux turbine server using torque for db connections. Once in a
while it just stops serving pages and needs to be restarted - the process is
running but seems to be dead.
I am suspecting the db connection pool runs out of connections. Is there any
diagnostic tool available, or a technique I can use to see how my torque
connection pool is doing? Maybe I'm on the wrong track, maybe there could be
other reasons.
Any help would be appreciated.
Ilan
RE: Turbine server keeps going down
Posted by Ilan Azbel <il...@autochartist.com>.
Hi Paulo,
I have tried to change to SharedPoolDataSourceFactory. I will see how it
performs. I have not been able to find the "usenewconnection" option you
speak of; can you point in the right direction?
Thanks for the help!
Ilan
> -----Original Message-----
> From: Paulo Santos [mailto:paulo@isnet.com.br]
> Sent: 01 November 2005 08:38 PM
> To: Turbine Users List
> Subject: Re: Turbine server keeps going down
>
> Hi Ilan,
>
> I had exactly the same problem several months ago with PostgreSQL and
> Torque. I *think* I had to change the dsfactory inside my
> conf/torque.properties file to read:
>
> torque.dsfactory.eCommerce.factory =
> org.apache.torque.dsfactory.SharedPoolDataSourceFactory
>
> You should keep the property name though, the "eCommerce" above is the
> name for my project. As I remember, doing this solved the problem. It's
> been a long time, so I may be forgetting something important here.
>
> I also turned off the torque's idbroker and it's "usenewconnection"
> option and upgraded to use the latest postgres jdbc driver. I found
> there are problems when using Postgres 8 with a version 7 driver as I
> was doing before.
>
> I think you are on the right track, I used to have 100 idle connections
> to the DB and had to restart the app server every day until I found the
> fix. Too bad I can't remember exactly what I did back then.
>
> Paulo
>
> Peter Courcoux wrote:
>
> > Hi Ilan,
> >
> > It is a long time since I have hit a problem like this but I seem to
> > recall that when my postgresql db ran out of connections it was clear
> > from the process listing, (ps aux) that the database connections were
> > not being released.
> >
> > Peter
> >
> > Ilan Azbel wrote:
> >
> >> Hello,
> >>
> >>
> >>
> >> I have a Linux turbine server using torque for db connections. Once in
> a
> >> while it just stops serving pages and needs to be restarted - the
> >> process is
> >> running but seems to be dead.
> >>
> >>
> >>
> >> I am suspecting the db connection pool runs out of connections. Is
> >> there any
> >> diagnostic tool available, or a technique I can use to see how my
> torque
> >> connection pool is doing? Maybe I'm on the wrong track, maybe there
> >> could be
> >> other reasons.
> >>
> >>
> >>
> >> Any help would be appreciated.
> >>
> >>
> >>
> >> Ilan
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: turbine-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-user-help@jakarta.apache.org
Re: Turbine server keeps going down
Posted by Paulo Santos <pa...@isnet.com.br>.
Hi Ilan,
I had exactly the same problem several months ago with PostgreSQL and
Torque. I *think* I had to change the dsfactory inside my
conf/torque.properties file to read:
torque.dsfactory.eCommerce.factory =
org.apache.torque.dsfactory.SharedPoolDataSourceFactory
You should keep the property name though, the "eCommerce" above is the
name for my project. As I remember, doing this solved the problem. It's
been a long time, so I may be forgetting something important here.
I also turned off the torque's idbroker and it's "usenewconnection"
option and upgraded to use the latest postgres jdbc driver. I found
there are problems when using Postgres 8 with a version 7 driver as I
was doing before.
I think you are on the right track, I used to have 100 idle connections
to the DB and had to restart the app server every day until I found the
fix. Too bad I can't remember exactly what I did back then.
Paulo
Peter Courcoux wrote:
> Hi Ilan,
>
> It is a long time since I have hit a problem like this but I seem to
> recall that when my postgresql db ran out of connections it was clear
> from the process listing, (ps aux) that the database connections were
> not being released.
>
> Peter
>
> Ilan Azbel wrote:
>
>> Hello,
>>
>>
>>
>> I have a Linux turbine server using torque for db connections. Once in a
>> while it just stops serving pages and needs to be restarted - the
>> process is
>> running but seems to be dead.
>>
>>
>>
>> I am suspecting the db connection pool runs out of connections. Is
>> there any
>> diagnostic tool available, or a technique I can use to see how my torque
>> connection pool is doing? Maybe I'm on the wrong track, maybe there
>> could be
>> other reasons.
>>
>>
>>
>> Any help would be appreciated.
>>
>>
>>
>> Ilan
>>
>>
>>
>>
>>
>>
>>
>>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-user-help@jakarta.apache.org
Re: Turbine server keeps going down
Posted by Peter Courcoux <pe...@courcoux.biz>.
Hi Ilan,
It is a long time since I have hit a problem like this but I seem to
recall that when my postgresql db ran out of connections it was clear
from the process listing, (ps aux) that the database connections were
not being released.
Peter
Ilan Azbel wrote:
> Hello,
>
>
>
> I have a Linux turbine server using torque for db connections. Once in a
> while it just stops serving pages and needs to be restarted - the process is
> running but seems to be dead.
>
>
>
> I am suspecting the db connection pool runs out of connections. Is there any
> diagnostic tool available, or a technique I can use to see how my torque
> connection pool is doing? Maybe I'm on the wrong track, maybe there could be
> other reasons.
>
>
>
> Any help would be appreciated.
>
>
>
> Ilan
>
>
>
>
>
>
>
>
--
------------------------------
Peter Courcoux
Telephone : +44 (0)1923 661488
Mobile : 07880 605626
email : peter@courcoux.biz
---------------------------------------------------------------------
To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-user-help@jakarta.apache.org
RE: Turbine server keeps going down
Posted by Ilan Azbel <il...@autochartist.com>.
Thanks Jeff, I would really appreciate the script!
> -----Original Message-----
> From: Jeffery Painter [mailto:painter@kiasoft.com]
> Sent: 01 November 2005 06:38 PM
> To: Turbine Users List
> Subject: Re: Turbine server keeps going down
>
>
> Does you log file show anything weird or does the process just up and die?
> Check to make sure you aren't running out of memory... does it happen when
> other processes become active (cron jobs maybe)... are you running any
> kind of monitor to log when it actually dies? I run a perl script in cron
> to test to make sure my servlet engine is still alive and sends an emaim
> alert when it goes down.
>
> just some thoughts... if you want my monitor script, just let me know.
>
> Jeff Painter
>
> On Tue, 1 Nov 2005, Ilan Azbel wrote:
>
> > Hello,
> >
> >
> >
> > I have a Linux turbine server using torque for db connections. Once in a
> > while it just stops serving pages and needs to be restarted - the
> process is
> > running but seems to be dead.
> >
> >
> >
> > I am suspecting the db connection pool runs out of connections. Is there
> any
> > diagnostic tool available, or a technique I can use to see how my torque
> > connection pool is doing? Maybe I'm on the wrong track, maybe there
> could be
> > other reasons.
> >
> >
> >
> > Any help would be appreciated.
> >
> >
> >
> > Ilan
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: turbine-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-user-help@jakarta.apache.org
Re: Turbine server keeps going down
Posted by Jeffery Painter <pa...@kiasoft.com>.
Does you log file show anything weird or does the process just up and die?
Check to make sure you aren't running out of memory... does it happen when
other processes become active (cron jobs maybe)... are you running any
kind of monitor to log when it actually dies? I run a perl script in cron
to test to make sure my servlet engine is still alive and sends an emaim
alert when it goes down.
just some thoughts... if you want my monitor script, just let me know.
Jeff Painter
On Tue, 1 Nov 2005, Ilan Azbel wrote:
> Hello,
>
>
>
> I have a Linux turbine server using torque for db connections. Once in a
> while it just stops serving pages and needs to be restarted - the process is
> running but seems to be dead.
>
>
>
> I am suspecting the db connection pool runs out of connections. Is there any
> diagnostic tool available, or a technique I can use to see how my torque
> connection pool is doing? Maybe I'm on the wrong track, maybe there could be
> other reasons.
>
>
>
> Any help would be appreciated.
>
>
>
> Ilan
---------------------------------------------------------------------
To unsubscribe, e-mail: turbine-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: turbine-user-help@jakarta.apache.org