Michael Howitz wrote at 2008-5-22 10:59 +0200:
> ...
>Am 22.05.2008 um 08:20 schrieb Dieter Maurer:
>> [...]
>> This looks like the misfeature "Zope does not start in debug mode
>> when a single product import fails". In non debug mode, product
>> import failures are logged but Zope nevertheless st
On Thu, May 22, 2008 at 8:20 AM, Dieter Maurer <[EMAIL PROTECTED]> wrote:
> Sasha Vincic wrote at 2008-5-21 12:47 +0200:
> >On Tue, May 20, 2008 at 8:30 PM, Rob Miller <[EMAIL PROTECTED]> wrote:
> >
> >> Philipp von Weitershausen wrote:
> >>
> >>> Believe something very very rotten in Five's regis
Am 22.05.2008 um 08:20 schrieb Dieter Maurer:
[...]
This looks like the misfeature "Zope does not start in debug mode
when a single product import fails". In non debug mode, product
import failures are logged but Zope nevertheless starts.
I don't think that this is a misfeature because in debu
Sasha Vincic wrote at 2008-5-21 12:47 +0200:
>On Tue, May 20, 2008 at 8:30 PM, Rob Miller <[EMAIL PROTECTED]> wrote:
>
>> Philipp von Weitershausen wrote:
>>
>>> Believe something very very rotten in Five's registerPackage was fixed by
>>> Rocky in r72986 [1]. As far as I can tell this was never me
On Wed, May 21, 2008 at 12:47:58PM +0200, Sasha Vincic wrote:
> So you run ./zopectl run do_nothing.py before you started each
> instance?
We do it just once, right after building a new instance; the instance
starts up fine forever after that.
> Is that better then my doInstall() check?
Well, no
On Tue, May 20, 2008 at 8:30 PM, Rob Miller <[EMAIL PROTECTED]> wrote:
> Philipp von Weitershausen wrote:
>
>> Believe something very very rotten in Five's registerPackage was fixed by
>> Rocky in r72986 [1]. As far as I can tell this was never merged to the 1.4
>> branch, but I could we wrong.
>>
Philipp von Weitershausen wrote:
Believe something very very rotten in Five's registerPackage was fixed
by Rocky in r72986 [1]. As far as I can tell this was never merged to
the 1.4 branch, but I could we wrong.
i can confirm that this seems to be helping. there's an easier way to
generate t
Believe something very very rotten in Five's registerPackage was fixed
by Rocky in r72986 [1]. As far as I can tell this was never merged to
the 1.4 branch, but I could we wrong.
[1] http://svn.zope.org/?rev=72986&view=rev
Sasha Vincic wrote:
Forgot to say that this is Zope 2.9.8, Five 1.4
Forgot to say that this is Zope 2.9.8, Five 1.4 branch from svn, Plone 2.5.5
/Sasha
On Fri, May 16, 2008 at 12:03 PM, Sasha Vincic <[EMAIL PROTECTED]>
wrote:
> Hi
> On a server we have a ZEO server with 6 clients. When we start/restart the
> server we often get on random instance an AttributeErr