[Zope] Re: MemoryError that brings my zope to its knees

[Tres Seaver]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Calvin Hendryx-Parker wrote:
> In the last month I have started to get these errors using Zope 2.7.7 in
> a ZEO configuration with one client and one storage server:
> 
> ------
> 2005-09-12T23:46:07 ERROR(200) ZODB Couldn't load state for 0x055ba7
> Traceback (most recent call last):
>   File "/usr/local/Zope/lib/python/ZODB/Connection.py", line 600, in
> setstate
>     self._set_ghost_state(obj, p)
>   File "/usr/local/Zope/lib/python/ZODB/Connection.py", line 639, in
> _set_ghost_state
>     state = unpickler.load()
> MemoryError
> ------
> 2005-09-12T23:46:07 ERROR(200) SiteError Error while logging
> Traceback (most recent call last):
>   File
> "/usr/local/Zope/lib/python/Products/SiteErrorLog/SiteErrorLog.py", line
> 142, in raising
>     tb_text = ''.join(
> MemoryError
> 
> 
> Sometimes I see this too:
> 
> ------
> 2005-09-06T15:23:30 ERROR(200) ZServer uncaptured python exception,
> closing channel <select-trigger (pipe) at bd9104c>
> (exceptions.MemoryError: [/usr/local/lib/python2.3/asyncore.py|read|69]
> [/usr/local/lib/python2.3/asyncore.py|handle_read_event|390]
> [/usr/local/Zope/lib/python/ZServer/medusa/thread/select_trigger.py|handle_read|77]
> [/usr/local/lib/python2.3/asyncore.py|recv|342]
> [/usr/local/lib/python2.3/asyncore.py|recv|523])

That is not a problem -- some client disconnected before Zope finished
processing the request.

> And this one also brings the show down:
> 
> ------
> 2005-09-06T05:35:51 ERROR(200) zrpc:30832 can't decode message:
> "(M2'K\x00U\x06.reply(TW\x00\x01\x00((U\tOFS.Imageq\x01U..."
> ------
> 2005-09-06T05:35:51 ERROR(200) zrpc-conn(C):localhost:6100 Error caught
> in asyncore
> Traceback (most recent call last):
>   File "/usr/local/lib/python2.3/asyncore.py", line 69, in read
>     obj.handle_read_event()
>   File "/usr/local/lib/python2.3/asyncore.py", line 390, in
> handle_read_event
>     self.handle_read()
>   File "/usr/local/Zope/lib/python/ZEO/zrpc/smac.py", line 217, in
> handle_read
>     self.message_input(msg)
>   File "/usr/local/Zope/lib/python/ZEO/zrpc/connection.py", line 339, in
> message_input
>     msgid, flags, name, args = self.marshal.decode(message)
>   File "/usr/local/Zope/lib/python/ZEO/zrpc/marshal.py", line 39, in decode
>     return unpickler.load() # msgid, flags, name, args
> MemoryError
> 
> It completely locks up the zope until I restart.  I added more RAM to
> the box and also bumped up the zeo cache-size to 100MB and 10000 objects
> in the zope.conf.  I also changed over from binding to an external
> ethernet interface and bound the zeo stuff to the localhost interface.
> 
> Any ideas on other things I could try?

 - Try 'memtest' or an equivalent on the box to verify that you don't
   have a flaky RAM chip.

 - If you are RAM constrained, *reduce* the number of objects in the
   connection cache;  more objects in that cache requires more RAM.
   Leaving the on-disk cache size high shouldn't be a problem.

 - Add swap space, if possible (swapping will be painful, but better
   than a lockup).


Tres.
- --
===================================================================
Tres Seaver          +1 202-558-7113          tseaver at palladion.com
Palladion Software   "Excellence by Design"    http://palladion.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFDJqbB+gerLs4ltQ4RAj4HAJ4vCNeuoqIH8kDXbVzp6YTbjDn2LgCgqnzo
bNaH6hDz9QBA5gCg+xcCyCs=
=osoi
-----END PGP SIGNATURE-----