June 2002 - Zope - Zope lists

FTP Crash
by Jim Kutter 12 Jun '02

12 Jun '02

Hi Folks - Any idea what might cause this? 2002-06-11T15:50:30 ERROR(200) ZServer uncaptured python exception, closing channel <ZServer.FTPServer.zope_ftp_channel connected 168.100.185.93:61104 at 0x88c91d4> (exceptions.AttributeError:'None' object has no attribute '_fileno' [/usr/local/zope/ZServer/medusa/asyncore.py|poll|94] [/usr/local/zope/ZServer/medusa/asyncore.py|handle_read_event|391] [/usr/local/zope/ZServer/medusa/asynchat.py|handle_read|82] [/usr/local/zope/ZServer/medusa/asyncore.py|recv|347] [/usr/local/zope/ZServer/medusa/asynchat.py|handle_close|149] [/usr/local/zope/ZServer/medusa/ftp_server.py|close|181] [/usr/local/zope/ZServer/medusa/asyncore.py|close|360] [/usr/local/zope/ZServer/medusa/asyncore.py|del_channel|245] [/usr/local/zope/ZServer/medusa/asyncore.py|__getattr__|366]) -jim

2 1

Zope and memory
by Jerome hebert 12 Jun '02

12 Jun '02

Zope 2.4 on a Debian 2.2 linux server. After 3 days uptime, Zope is taking lot and lot of memory on my server although this is a very small site with very few hits. socrate:~# memstat | sort -rn | more 1958684k 157416k: PID 14967 (/opt/Zope-2.4.0-linux2-x86/bin/python) 157416k: PID 14966 (/opt/Zope-2.4.0-linux2-x86/bin/python) 157416k: PID 14965 (/opt/Zope-2.4.0-linux2-x86/bin/python) 157416k: PID 14964 (/opt/Zope-2.4.0-linux2-x86/bin/python) 157416k: PID 14963 (/opt/Zope-2.4.0-linux2-x86/bin/python) 157416k: PID 14692 (/opt/Zope-2.4.0-linux2-x86/bin/python) What can cause Zope to use that much memory ? How Zope manage its memory ? Regards, Jerome. ______________________________________________________________________________ ifrance.com, l'email gratuit le plus complet de l'Internet ! vos emails depuis un navigateur, en POP3, sur Minitel, sur le WAP... http://www.ifrance.com/_reloc/email.emailif

3 4

Sessions without cookies
by Heimo Laukkanen 12 Jun '02

12 Jun '02

Hola, now I have tested the UrlSession Adapter and confirmed that it works. ( Download from http://www.zope.org/Members/k_vertigo/Products/URLSessionAdapter ) While creating an instance somehow AccessRule was not created to the folder where I created the URL Session Adapter - so I had to create the Access Rule manually. If you do the same then just remember to give the URL Session Adapters ID to the Access rule... Now Zope creates - when needed - session id:s to the URL. So http://www.mydomain.com/foo/ will transform into http://www.mydomain.com/xytadsekhgs*session-id-here*hga/foo/ Not bad. Eventhough the php way of giving the session id as a parameter was sexier ,-) -huima

1 0

Re: [Zope] Using a template for all pages, or calling dtml as a function, or perhaps switch to Zpt ??
by a.p.bourne＠salford.ac.uk 12 Jun '02

12 Jun '02

Max, Is your index_html object a DTML Method or a DTML Document? If it's a DTML Document then acquisition won't work in the way you expected. Change it into a DTML Method and the structure you describe below will work. You will not need to create an index_html object in every sub-folder (but each subfolder can optionally have its own unique primary_content and secondary_content objects). Regards, Andy Bourne > Date: Wed, 12 Jun 2002 09:52:03 +0200 > From: Max M <maxm(a)mxm.dk> > To: zope(a)zope.org > Subject: [Zope] Using a template for all pages, or calling dtml as a > function, or > perhaps switch to Zpt ?? > > I have a site where i want to use the same template for all the pages. > Lets say that I have the following structure: > > index_html: > <b> > <dtml-var primary_content> > </b> > <i> > <dtml-var secondary_content> > </i> > > I then want to use this template for all my pages. Most of my products > will them implement "primary_content()" and "secondary_content()" > methods. If a product hasn't implemented one of the methods, it just > finds it from somewhere up the aquisition (aq) path > > So far so good. > > My problem is that it seems that I have to define "index_html" in > every product for this to work. > > If I just put an "index_html" in the root of my aq path zope will > traverse it's way up to it allright, but naturally won't find the > objects "primary_content()" and "secondary_content()" methods. So the > view is wrong. > > If I create a simple dtml attribute that looks like: > > index_html: > <dtml-var "aq_parent.index_html(None, _)"> to call the aquired > index_html within the current namespace, it doesn't work either. > > I want to use this type of layout in a site using a lot of products, > so putting all the index_html code inside each product is A Bad Thing > for many reasons. > > Has anybody got an idea as to how I can approach this? Perhaps by > using Zpt ? > > Isn't it possible to call the index_html like a function from within > the current namespace ?? If so, isn't index_html(None, _) the right > way to do it? > > > regards Max M > > -- > > "Sorry I would Really Like To Help More On This Project, > But Am To Busy Doing Paid Work On A Tight Deadline" > Max M Andy Bourne Information Officer Information Services Division University of Salford Tel: 0161-295-6229 Fax: 0161-295-6189

1 0

[HELP] UltraseekDA product problem
by flemaitre＠fede.generali.fr 12 Jun '02

12 Jun '02

Hello, I have a problem using the UltraseekDA product. Explication : - In one hand, i have an UltraSeek server (now named Inktomi Search 4.4.0) - In the other hand, i have my Zope server (Zope 2.4.4 on NT Server) In order to access my Ultraseek server in Zope, a have installed the UltraSeek product : i followed the installation instructions : both Zope side and Ultraseek side. When a access to my Ultraseek server (by : http://myUltraseek/da) it's works But when i create a new "Ultraseek DA" from the "Add..." form, i specify the fields id, title, UltraseekURL, EcryptKey, .... i have the error : "An error occurred connecting to the server. " I found nothing on the archive. Can someone help me ? Thanks in advance. PS : Execuse my english. Frédéric.

1 0

Sessions: without cookies? url rewrite?
by Heimo Laukkanen 12 Jun '02

12 Jun '02

>Subject: Re: [Zope] Sessions: without cookies? url rewrite? >From: Chris McDonough <chrism(a)zope.com> >To: Frank Tegtmeyer <fte-sub-zope(a)fte.to> >Cc: zope(a)zope.org >Date: 12 Jun 2002 09:07:52 -0400 > >Yes, it is. Kapil Thangavelu made a URLSession Adapter product that >automates this, but I don't have the link at the moment. Took a couple of minutes of searching: http://www.zope.org/Members/k_vertigo/Products/URLSessionAdapter Haven't yet taken a look into it. -huima

1 0

Checkboxes in formulator
by Michael Montagne 12 Jun '02

12 Jun '02

I'm trying to create a multi-select checkbox that offers each item in a folder. The idea is to select items to print. I thought I'd use formulator and put a dtml-in statement in the "items" box. That did not work. Is there a good way to do this? -- Michael Montagne montagne(a)boora.com http://www.boora.com

1 0

Standard logging?
by Jørgen Hermanrud Fjeld 12 Jun '02

12 Jun '02

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi. I wondered why Zope didn't: - -Came with a default rotating log schema that adheres to apache standard, with LOGDIR=$INSTANCE_HOME/logs LOGFILE=$LOGDIR/access_log - -And used rotating logs with weekly/daily rotations by default, by using rotating_file_logger This would, in my eyes, make it easier to generate statistics, without having to edit z2.py Are there any arguments for not doing this? And why isn't the logfile in CLF? As far as I can see, it incorporates all logs, that is i.e. FTP that isn't understood by i.e. webalizer. How are one planning to this in the future? - -- Sincerely Jørgen Hermanrud Fjeld jhf(a)hex.no -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iEYEARECAAYFAj0HTpsACgkQ9jvTqPy5VsoQEgCeKD5UhdE5w7bGv2OeJv+HUNuK RooAniVpyQghoYlatZvEoXGSj7meXLDz =/0Fd -----END PGP SIGNATURE-----

3 3

Python script: problem with session
by p.t. 12 Jun '02

12 Jun '02

I did already post this with a different thread: probably the formulation was to complex. Or may be there is some very basic knowledge I'm missing. I'll try again. If I assign to variable in a Python script (say context.REQUEST['foo']) a value stored in SESSION (say context.SESSION['foo']), and then I change the retrieved value (the one in REQUEST, not the original one), also the variable in SESSION is changed. In this way I have lost my original information stored in SESSION! I've tryied to avoid this, but I can't manage to find how. Any suggestion? TIA, p.t.

3 6

URL relative to document, not folder. How to fix?
by James Hazard 12 Jun '02

12 Jun '02

Hi, I'm new. I loaded a bunch of html documents into a couple of folders under CMF and discover that the usual notation for a relative link <a href="doc2.html">doc2</a> looks for doc1.html/doc2.html ! Because of inheritance (?), it finds the document as long as the calling and called documents are in the same folder, but if I want a document in another folder <a href="../introduction.html">Up for Intro</a>, it doesn't work. It seems that the path is relative to the document and not the folder, so to speak. Have I got something misconfigured? Thanks, Jim

3 4