Re: [ZODB-Dev] Connection GC

2010-09-28 Thread Jim Fulton
On Tue, Sep 28, 2010 at 6:42 PM, Matthias  wrote:
> Am 29.09.2010, 00:34 Uhr, schrieb Jim Fulton :
>
>>> Is this a bug in ZODB?
>>
>> It appears so.  Are you specifying a pool timeout?
>
> I don't set a timeout anywhere specifically. Maybe there is one a default
> one buried in grok/ztk, but I couldn't find anything on quick inspection.
>
> However, the error is gone for me now, because I discovered a mistake in my
> custom publisher. It would sometimes leave an open transaction so that the
> connection could not be closed. Too many open connections then caused the
> traceback in question I guess. Now that there aren't too many historical
> connections, the traceback does not appear anymore.

OK. I'm glad it's cleared up.

I'll fix this for 3.10 final.

Jim

-- 
Jim Fulton
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


Re: [ZODB-Dev] Connection GC

2010-09-28 Thread Matthias
Am 29.09.2010, 00:34 Uhr, schrieb Jim Fulton :

>> Is this a bug in ZODB?
>
> It appears so.  Are you specifying a pool timeout?

I don't set a timeout anywhere specifically. Maybe there is one a default  
one buried in grok/ztk, but I couldn't find anything on quick inspection.

However, the error is gone for me now, because I discovered a mistake in  
my custom publisher. It would sometimes leave an open transaction so that  
the connection could not be closed. Too many open connections then caused  
the traceback in question I guess. Now that there aren't too many  
historical connections, the traceback does not appear anymore.

-Matthias
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


Re: [ZODB-Dev] Connection GC

2010-09-28 Thread Jim Fulton
On Tue, Sep 28, 2010 at 6:01 PM, Matthias  wrote:
> Hello,
>
> I keep getting this traceback with ZODB 3.9.4.
>
> Traceback (most recent call last):
>   File
> "c:\users\nitro\.buildout\eggs\zope.publisher-3.12.3-py2.6.egg\zope\publi
> sher\publish.py", line 130, in publish
>     obj = publication.getApplication(request)
>   File "c:\anima\sources\anima\src\anima\publication\publication.py", line
> 34, i
> n getApplication
>     conn = self.db.open()
>   File
> "c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
> ine 770, in open
>     self.historical_pool.availableGC()
>   File
> "c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
> ine 288, in availableGC
>     pool.availableGC()
>   File
> "c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
> ine 231, in availableGC
>     del self.available[t]
> TypeError: list indices must be integers, not float
>
> Is this a bug in ZODB?

It appears so.  Are you specifying a pool timeout?

Jim

-- 
Jim Fulton
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


[ZODB-Dev] Connection GC

2010-09-28 Thread Matthias
Hello,

I keep getting this traceback with ZODB 3.9.4.

Traceback (most recent call last):
   File  
"c:\users\nitro\.buildout\eggs\zope.publisher-3.12.3-py2.6.egg\zope\publi
sher\publish.py", line 130, in publish
 obj = publication.getApplication(request)
   File "c:\anima\sources\anima\src\anima\publication\publication.py", line  
34, i
n getApplication
 conn = self.db.open()
   File  
"c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
ine 770, in open
 self.historical_pool.availableGC()
   File  
"c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
ine 288, in availableGC
 pool.availableGC()
   File  
"c:\users\nitro\.buildout\eggs\zodb3-3.9.4-py2.6-win32.egg\ZODB\DB.py", l
ine 231, in availableGC
 del self.available[t]
TypeError: list indices must be integers, not float

Is this a bug in ZODB?

-Matthias
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


[ZODB-Dev] ZODB 3.9.7 released

2010-09-28 Thread Jim Fulton
http://pypi.python.org/pypi/ZODB3/3.9.7

This was mainly motivated by the Python 2.7 fix, since we'd previously
made Python 2.7 windows binaries available.

Jim

--
Jim Fulton
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


[ZODB-Dev] ZODB 3.10.0b7 released

2010-09-28 Thread Jim Fulton
http://pypi.python.org/pypi/ZODB3/3.10.0b7

This is the last release I plan to make before 3.10.0, hopefully, next week.

Jim

-- 
Jim Fulton
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


Re: [ZODB-Dev] Windows/ZODB Tests Passing

2010-09-28 Thread Jim Fulton
On Tue, Sep 28, 2010 at 3:12 AM, Adam GROSZER  wrote:
> Hello Alan,
>
> They pass since the summit, when Jim fixed the windows failures.
>
> http://winbot.zope.org/builders

Actually, they passed since I fixed a couple of ZEO cache management
bugs on August 30 and September 7.  What's interesting is that these
weren't windows related, but rather due to race conditions that
Windows tends to be good at finding.

Also interesting is that the test that stopped failing on Windows kept
failing (through less dramatically) on an odd-ball linux buildbot here
at ZC finally leading to finding and fixing a very subtle error
yesterday, involving a race condition on the server that led to
invalid cache data on clients.  The test only failed when this
buildbot ran it. If I logged into the buildbot machine, and ran the
test manually, it passed.

Jim

--
Jim Fulton
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


Re: [ZODB-Dev] ZODB history problem with Zope 2.12.7

2010-09-28 Thread Chris Withers
On 28/09/2010 12:55, Tres Seaver wrote:
> On 09/28/2010 03:36 AM, Chris Withers wrote:
>> On 28/09/2010 00:12, Marius Gedminas wrote:
>>> --- ./OFS/History.py.orig   2010-09-28 02:11:56.535745440 +0300
>>> +++ ./OFS/History.py2010-09-28 02:12:00.043764683 +0300
>>> @@ -151,6 +151,9 @@
>>>base = aq_base(self)
>>>base._p_activate()   # make sure we're not a ghost
>>>base.__setstate__(state) # change the state
>>> +for attr in dir(base):
>>> +if attr.startswith('_v_'):
>>> +delattr(base, attr)
>>>base._p_changed = True   # marke object as dirty
>>>self.manage_afterHistoryCopy()
>>
>> Thanks, I guess I'll monkey patch for now, here's the bug:
>>
>> https://bugs.launchpad.net/zope2/+bug/649605
>>
>> However, I'm curious, so the above will fix the object in the current
>> thread, but what about objects in other threads?
>>
>> (or do _v_ attributes get killed off at the start of each transaction?)
>
> Only when objects are ghostified (due to an invalidation from another
> thread or process) or evicted from the cache.  I'm not quite sure how
> the case you are triggering occurs, but if that code saves the new-old
> version of the template to the ZODB, then any instances in other
> threads' connection caches should be invalidated.

Feel free to repeat the steps I originally posted and are in the bug 
description ;-)

Looks like the History.py code isn't doing something it should, but I 
don't know when or what changed to make this so. I'll CC zodb-dev in 
case Jim knows of any changes in ZODB (I'm using 3.9.6) that might be 
relevant...

cheers,

Chris

-- 
Simplistix - Content Management, Batch Processing & Python Consulting
 - http://www.simplistix.co.uk
___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev


Re: [ZODB-Dev] Windows/ZODB Tests Passing

2010-09-28 Thread Adam GROSZER
Hello Alan,

They pass since the summit, when Jim fixed the windows failures.

http://winbot.zope.org/builders

Monday, September 27, 2010, 7:17:42 PM, you wrote:

AR> For anyone who cares.  ZODB trunk builds win32 and amd64 have been
AR> passing tests for almost a week without failures.  Thanks Jim for
AR> getting ZODB tests working reliably on Windows!

AR> cheers
AR> alan

AR> p.s. let us know if you need py2.7 builders and what to test.  you can
AR> find all the builders at hudson.enfoldsystems.com
AR> ___
AR> For more information about ZODB, see the ZODB Wiki:
AR> http://www.zope.org/Wikis/ZODB/

AR> ZODB-Dev mailing list  -  ZODB-Dev@zope.org
AR> https://mail.zope.org/mailman/listinfo/zodb-dev


-- 
Best regards,
 Adam GROSZERmailto:agros...@gmail.com
--
Quote of the day:
A bird in the bush usually has a friend in there with him.

___
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
https://mail.zope.org/mailman/listinfo/zodb-dev