- Zope - Zope lists

Newbie - exUserFolder
by Ashley Lloyd 10 Apr '03

10 Apr '03

I have Zope 2.5.1, exUserFolder 0.10.10, Win 2k, Interbase Authentication database. In my root folder, I have a standard acl_users. I have a remote database for authentication, in which the passwords are not encrypted (I don't know if this is my problem - but if it is, I don't know where the code I need to change for Login is, I have found the code for update/add user). I wish to use this database to log users into a site I am developing, and have been trying to test exUserFolder. I have a "Test" subfolder (from root), and a "userPages" subfolder of Test. The exUserFolder instance is in /Test >From what I have read I believe it is best if the exUserFolder instance is in a folder closer to root than the pages you wish to protect with it, so the pages are in /Test/userPages The exUserFolder instance links to the correct database, and seems to retrieve the list of users (as listed on the users page of the management screens), the table and fields for users/passwords/roles have been set up correctly. Currently the exUserFolder uses standard basic authentication. I have changed the security settings for the /Test/userPages folder only for View, to be accessible to anyone apart from anonymous. If I attempt to view this index_html, I get the login box, but username/password combinations from the database do not work, I can only login with usernames from the acl_users in the root folder. Also, if I change the Authentication Type to cookie-based, immediately on updating (after the confirmation screen) in the ZMI, I get a Login page. No matter what username/password I try (from the database or the root acl_users) I cannot login. I also cannot view the /Test folder itself (trying to access it from the ZMI tree) or any subfolders of /Test, but I can view all other folders, without having to re-enter the password I used to get into the ZMI in the first place. Can anyone direct me as to where I am going wrong? TIA, Kind regards Ashley _________________________________________________________________ Overloaded with spam? With MSN 8, you can filter it out http://join.msn.com/?page=features/junkmail&pgmarket=en-gb&XAPID=32&DI=1059

1 0

tracebacks
by Paavo Parkkinen 10 Apr '03

10 Apr '03

I'm an idiot. Just figured it out two minutes after sending the previous message. Sorry. -- paavo. wu wei er wu bu wei

1 0

Re: i have problems installing Zope 2.6.1
by Josef Meile 10 Apr '03

10 Apr '03

> please tell me what to do (i don't know to upgrade my > python version because i use it with a lot of other > applications) In fact, you don't have to replace your old python version, you can install the new one in another folder and use it from zope (assuming that you're using linux or unix). This nice how-to explains the way it can be done: http://www.zope.org/Members/beacon/install_instructions This how-to doesn't say it, but you also have to uncomment the modules that you need from the Modules/Setup file in the python source code installation (like zlib for example). Note: if your old python is installed under /usr/local, then you can specify another path in the configure script (read the comments inside this file). Regards, Josef

1 0

Can't get tracebacks to show
by Paavo Parkkinen 10 Apr '03

10 Apr '03

I have Zope 2.6.0 installed and can't get tracebacks to show anywhere. I've been searching the net for clues, but nothing seems to help. I've tried the Z_DEBUG_MODE, BOBO_DEBUG_MODE, -D. Here is my 'start'-script, if someone could point out what I'm doing wrong. ----- start ----- #!/bin/sh umask 077 reldir=`dirname $0` cwd=`cd $reldir; pwd` # Zope's event logger is controlled by the "EVENT_LOG_FILE" environment # variable. If you don't have a "EVENT_LOG_FILE" environment variable # (or its older alias "STUPID_LOG_FILE") set, Zope will log to the standard # output. For more information on EVENT_LOG_FILE, see doc/ENVIRONMENT.txt. BOBO_DEBUG_MODE=yes Z_DEBUG_MODE=yes EVENT_LOG_FILE="var/event.log" ZLOGFILE=$EVENT_LOG_FILE if [ -z "$ZLOGFILE" ]; then ZLOGFILE=$STUPID_LOG_FILE fi if [ -z "$ZLOGFILE" ]; then EVENT_LOG_FILE="" export EVENT_LOG_FILE fi exec /usr/bin/python2.1 $cwd/z2.py -D "$@" -w 8080 -f 8021 -u zope & ----------------- -- paavo. wu wei er wu bu wei

1 0

Re: big problem when upgrading from2.5.1 to 2.6.1
by Josef Meile 10 Apr '03

10 Apr '03

> I think you should add "traceback info may contain clues" in your > howto so that googleing for this very specific message would be > successfull. Ok, I'll search my message in the newsgroup and include it. > Also your solution says "I will assume that you have also a fresh > installation of Zope 2.6.0 without anything on it and of course your old Zope 2.5.1". > > As specified in my previous message, my problem is (was I hope) that > I don't have my old Zope 2.5.1 ! This means that your solution, in > case I had found it, was unusable at that time, i.e. before I tried to > downgrade. Sorry for that, I thought you had a backup. > PS : I already know who I have to blame for not having a backup, > thank you very much. Who me? I think it's a responsibility of each administrator to do a backup before doing a software upgrade. Anyway, now I just remmember that somebody disscussed just some days ago a way of correcting mistakes if you have already upgrade your zope. He said that you had to export as XML, look for a string (If I'm not bad it was something like <object>, you've to look the archives) erased it and reimport the object again. Regards, Josef.

2 1

Zope temporary files bloating /tmp/
by Heimo Laukkanen 10 Apr '03

10 Apr '03

Hi all, a sysadmin-friend of mine noticed that on their system Zope 2.6 is creating a large number of tempfiles that bloat the tmp partition on the system and do not get deleted/released for a long time. In the worst case there are over 200 megabytes of these tempfiles bloating the temp and nothing else than a restart seems to help. I would need some help for debugging this kind of behaviour and what could be done. So far I know that tempfiles are used for example in situations of upload or downloads, where Zope writes the file to a tempfile before/after it is written to ZODB. Any help and insight would be appreciated. -huima

2 3

Re: big problem when upgrading from2.5.1 to 2.6.1
by Josef Meile 10 Apr '03

10 Apr '03

--- In zope(a)yahoogroups.com, Jerome Alet <alet@l...> wrote: > On Wed, Apr 09, 2003 at 08:47:06PM +0200, Dieter Maurer wrote: > > > > There are known problems with (some) ZClasses when upgrading to > > Zope 2.6.1. There are solution for them (--> archive). > > No solution for this problem there... Wrong! It was discussed several months ago. Anyway, you can look my how-to where I wrote the solution that helped me to cope with some issues: http://www.zope.org/Members/jmeile/upgradingZope > Sorry, it is something like (not at work yet to post the exact > message) : > > impossible to unpickle object oid "garbage" > Error Value : None, traceback may contain clues. > > "garbage" is a bucnh of non-printable characters You should search the name of the ZClass that originated the message in the garbage. Then see if it inherits from ObjectManager. If so, see the second point of my how-to. > I've succeeded at accessing it with a 2.5.1 installation, > and I've made a backup, now I'll try to reimport into > 2.6.1 tomorrow if possible. Actually, that was the solution proposed several months ago. (see my how-to) Regards, Josef.

2 1

Re: can't start new thread
by seila＠gmx.de 10 Apr '03

10 Apr '03

>> we have just installed zope to a new server (RedHat 7.3, Zope 2.6.1, Python >> 2.1.3, Apache 1.3.27, multiprocessor: Dual Xeon HotSwap, ICP Vortex Hardware >> Raid 10). >> Since we have a problem reaching zope behind apache, we started prozess in >> debug modus. >> Zope seems to start normally (port 8081, 8021), after the first call of the >> domain in the webbrowser (we get 500 server error), we find these 2 lines in >> the debug log: >> >> ERROR(200) ZServer Server Error: thread.error, can't start new thread >> file: /zope/Zope-2.6.1/ZServer/PubCore/ZRendezvous.py line: 32 >I've had a similar problem with RedHat 7.3 / 8 (upgraded from 7.2). Zope >ran with only one (!!!) thread, regardless if I used my Redhat's Python, >Zope's binary python or my own Python-Build. I didn't worked it out, so >I switched to Debian and I'm lucky again...:) Hi Maik, thanks a lot for your disencouraging answer ;-) we already found your posting concerning this problem on the dzug maillinglist. Unfortunately we cannot switch to debian at the moment. We think the problem might be related to the "--with-thread" Option when compiling python. Zope 2.6.1 seems to need "libpthread.so". At the moment we have the impression python 2.1.3 on redhat 7.3 does not compile the way we wanted to do (./configure --prefix=/usr/local --with-lcrypt --with-thread=/usr/lib/). We are still trying .... ;-) Seila -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!

1 0

can't start new thread
by seila＠gmx.de 10 Apr '03

10 Apr '03

Hi all, we have just installed zope to a new server (RedHat 7.3, Zope 2.6.1, Python 2.1.3, Apache 1.3.27, multiprocessor: Dual Xeon HotSwap, ICP Vortex Hardware Raid 10). Since we have a problem reaching zope behind apache, we started prozess in debug modus. Zope seems to start normally (port 8081, 8021), after the first call of the domain in the webbrowser (we get 500 server error), we find these 2 lines in the debug log: ERROR(200) ZServer Server Error: thread.error, can't start new thread file: /zope/Zope-2.6.1/ZServer/PubCore/ZRendezvous.py line: 32 Has anybody an idea what the problem could be? Maybe the problem mentioned in: http://www.zope.org/Members/glpb/solaris/report_ps (Python Global Interpreter Lock)? Any ideas? Thanx a lot, Seila -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!

2 1

P.S.: Add ZCTextIndex Error on Zope 2.6.1 stable release
by Chris 10 Apr '03

10 Apr '03

I did the same with the ZCatalog product directory and now it works fine. Could there be any problems caused by that? perhaps you have a bug fix ready or a hint for me ;)

2 1