ZServer on Windows 2000 w/Apache issue
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again. What can i do to optimize the server so this wont happen? It gets so bad that even just Apache served sites slow down also. This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also. The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage. I am more than willing to answer any questions you may need to help me figure out what is going on here. TIA Sherwood
Sherwood Robinson wrote:
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again.
What can i do to optimize the server so this wont happen? It gets so bad that even just Apache served sites slow down also.
uhumm...is the os using the swap? the ram is full?
This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also.
i ever had problems with ODBC when i was using it with iterbase until i switched to a native driver :-)
The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage.
What do u mean here? what's the process memory consumption when it hangs?
I am more than willing to answer any questions you may need to help me figure out what is going on here.
uhumm ....you are using zope 2.3.0...some questions: Do you have tried to flush Zope's cache istead of restart it? Do you have tried to activate the profiling system to see where the time is spent?
Neither ram or Swap is full... and only 50% of processor Resources are tied up by the Python process At 03:26 PM 4/6/2001 +0100, Alberto Berti wrote:
Sherwood Robinson wrote:
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again. What can i do to optimize the server so this wont happen? It gets so bad that even just Apache served sites slow down also.
uhumm...is the os using the swap? the ram is full?
This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also.
i ever had problems with ODBC when i was using it with iterbase until i switched to a native driver :-)
The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage.
What do u mean here? what's the process memory consumption when it hangs?
I am more than willing to answer any questions you may need to help me figure out what is going on here. uhumm ....you are using zope 2.3.0...some questions:
Do you have tried to flush Zope's cache istead of restart it?
Do you have tried to activate the profiling system to see where the time is spent?
Sherwood Robinson wrote -
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again.
...
This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also.
The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage.
Some ODBC drivers have memory leaks. Cold Fusion actually supplies a utility to stop and restart the CF server on a schedule (like at 2 AM) to work around this problem. Sounds like this could be it. Cheers, Tom P
We don't run Cold Fusion But I can stop and start the Zope Server at specified times I'm sure. But I shouldn't have to do that.... At 10:35 AM 4/6/2001 -0400, Thomas B. Passin wrote:
Sherwood Robinson wrote -
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again.
...
This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also.
The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage.
Some ODBC drivers have memory leaks. Cold Fusion actually supplies a utility to stop and restart the CF server on a schedule (like at 2 AM) to work around this problem. Sounds like this could be it.
Cheers,
Tom P
_______________________________________________ Zope maillist - Zope@zope.org http://lists.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope-dev )
I didn't mean to literally use CF or their utility. I just mentioned to indicate that CF took the problem seriously. Memory leaks aren't specific to Zope or Win2000, of course. But it's probably best to bite the bullet and try it out even though this "shouldn't" be necessary. Or, as someone else said, maybe a native database driver, if there is one for Zope, would clear up the problem. Cheers, Tom P Sherwood Robinson wrote -
We don't run Cold Fusion But I can stop and start the Zope Server at specified times I'm sure. But I shouldn't have to do that....
At 10:35 AM 4/6/2001 -0400, Thomas B. Passin wrote:
Sherwood Robinson wrote -
OK I have been trying to figure out why this keeps happening but the Zope server seems to run fine for a day and slow to a crawl as the day goes by. This happens to the point that I have to restart the Zope server. This works fine till the server gets hit with heavy traffic or as a day grows on and it slows down again.
...
This only happens with our main Zope install. This runs our primary site. It has the most stuff in it using ODBC to connect to access databases and some MySQL activity is happening on the server also.
The Python process for this server is only utilizing only about 50% of the processor recourses but this is steady. It also climbs up to 200K+ in memory usage.
Some ODBC drivers have memory leaks. Cold Fusion actually supplies a utility to stop and restart the CF server on a schedule (like at 2 AM) to work around this problem. Sounds like this could be it.
participants (3)
-
Alberto Berti -
Sherwood Robinson -
Thomas B. Passin