no more SIG11 --without-pymallog --without-cycle-gc (was Re: [Zope-dev] Bravery & Zope 2.5 Testers)

Harald Koschinski harald.koschinski@friatec.de
Wed, 16 Jan 2002 09:24:21 +0100


--------------060906040602040105010008
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

  same for me, no more crashes - but zope needs very much memory (logo :-(()

I will try know python 2.1.2.

Leonardo Rochael Almeida wrote:

>Hi folks,
>
>
>I also tried the CVS ExtensionClass.h but it didn't help. We kept
>getting the SIG11's.
>
>However, after reading all comments on this problem report on
>sourceforge:
>http://sourceforge.net/tracker/?func=detail&atid=105470&aid=471942&group_id=5470
>pointed out in this list by Joseph Wayne Norton, and the diff -u between
>the CVS ExtensionClass.h and the one that came with zope 2.4.3, I
>decided to try recompiling python --without-cycle-gc (in addidtion to
>--without-pymalloc, which we used before causing an increase in time
>between restarts from 15 min to 40 min).
>
>Now our zope is running for "17 hours 29 min 40 sec" (pasted from the
>Control_Panel) without a single SIG11, YAY!!!
>
>I strongly recommend anyone running into these "Aiieee! 54321 exited
>with error code: 11
>" to recompile their pythons --without-cycle-gc" to
>see if the SIGSEGVs go away. Maybe even for Stefano Noferi's script?
>
>Throw in a --without-pymalloc for good measure :-)
>
>Now our zope is probably slowly leaking some cycles, but what the heck,
>I can live with restarting zope once per week. With ZEO it comes back in
>2 seconds anyway :-)
>
>	Cheers, Leo
>
>PS: No, I have no intention to check if --with-pymalloc
>--without-cycle-gc also works :-)
>
>On Wed, 2001-12-12 at 19:57, Matthew T. Kromer wrote:
>
>>OK, I am formulating a hypothesis that if you replace ExtensionClass.h 
>>in Zope 2.4 and Zope 2.5 with the following one:
>>
>>http://cvs.zope.org/~checkout~/Zope/lib/Components/ExtensionClass/src/ExtensionClass.h?rev=1.16.140.2&content-type=text/plain
>>
>>(Which goes in lib/Components/ExtensionClass/src/ExtensionClass.h btw)
>>
>>and rebuild zope .... python2.1 wo_pcgi.py
>>
>>and rebuld any other modules you may have which depend on extension class
>>
>>that some of the segfaults could go away.
>>[...]
>>
>

regards


Harald Koschinski
_____________________________________________________________________

FRIATEC Aktiengesellschaft
Informationssysteme und Organisation            
Tel. 0621 - 486 - 1454
Fax. 0621 - 486 - 1840
Steinzeugstrasse 50
mailto:harald.koschinski@friatec.de
D-68229 Mannheim



--------------060906040602040105010008
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <title></title>
</head>
<body>
<title></title>
       same for me, no more crashes - but zope needs very much memory (logo
:-(()<br>
<br>
 I will try know python 2.1.2.<br>
<br>
 Leonardo Rochael Almeida wrote:<br>
<blockquote type="cite" cite="mid:1008351449.5381.3.camel@spitfire">
  <pre wrap="">Hi folks,<br><br><br>I also tried the CVS ExtensionClass.h but it didn't help. We kept<br>getting the SIG11's.<br><br>However, after reading all comments on this problem report on<br>sourceforge:<br><a class="moz-txt-link-freetext" href="http://sourceforge.net/tracker/?func=detail&amp;atid=105470&amp;aid=471942&amp;group_id=5470">http://sourceforge.net/tracker/?func=detail&amp;atid=105470&amp;aid=471942&amp;group_id=5470</a><br>pointed out in this list by Joseph Wayne Norton, and the diff -u between<br>the CVS ExtensionClass.h and the one that came with zope 2.4.3, I<br>decided to try recompiling python --without-cycle-gc (in addidtion to<br>--without-pymalloc, which we used before causing an increase in time<br>between restarts from 15 min to 40 min).<br><br>Now our zope is running for "17 hours 29 min 40 sec" (pasted from the<br>Control_Panel) without a single SIG11, YAY!!!<br><br>I strongly recommend anyone running into these "Aiieee! 54321 exited<br>with er
ror code: 11
" to recompile their pythons --without-cycle-gc" to<br>see if the SIGSEGVs go away. Maybe even for Stefano Noferi's script?<br><br>Throw in a --without-pymalloc for good measure :-)<br><br>Now our zope is probably slowly leaking some cycles, but what the heck,<br>I can live with restarting zope once per week. With ZEO it comes back in<br>2 seconds anyway :-)<br><br>	Cheers, Leo<br><br>PS: No, I have no intention to check if --with-pymalloc<br>--without-cycle-gc also works :-)<br><br>On Wed, 2001-12-12 at 19:57, Matthew T. Kromer wrote:<br></pre>
  <blockquote type="cite">
    <pre wrap="">OK, I am formulating a hypothesis that if you replace ExtensionClass.h <br>in Zope 2.4 and Zope 2.5 with the following one:<br><br><a class="moz-txt-link-freetext" href="http://cvs.zope.org/%7Echeckout%7E/Zope/lib/Components/ExtensionClass/src/ExtensionClass.h?rev=1.16.140.2&amp;content-type=text/plain">http://cvs.zope.org/~checkout~/Zope/lib/Components/ExtensionClass/src/ExtensionClass.h?rev=1.16.140.2&amp;content-type=text/plain</a><br><br>(Which goes in lib/Components/ExtensionClass/src/ExtensionClass.h btw)<br><br>and rebuild zope .... python2.1 wo_pcgi.py<br><br>and rebuld any other modules you may have which depend on extension class<br><br>that some of the segfaults could go away.<br>[...]<br></pre>
    </blockquote>
    <pre wrap=""><!----><br></pre>
    </blockquote>
    <br>
 regards<br>
    <pre class="moz-signature" cols="$mailwrapcol"><br>Harald Koschinski<br>_____________________________________________________________________<br><br>FRIATEC Aktiengesellschaft<br>Informationssysteme und Organisation            <br>Tel. 0621 - 486 - 1454<br>Fax. 0621 - 486 - 1840<br>Steinzeugstrasse 50<br><a class="moz-txt-link-freetext" href="mailto:harald.koschinski@friatec.de">mailto:harald.koschinski@friatec.de</a><br>D-68229 Mannheim<br></pre>
    <br>
    </body>
    </html>

--------------060906040602040105010008--