- Zope - Zope lists

RE: ZopeProfiler issue (found root cause)
by Pascal Peregrina 27 Jun '05

27 Jun '05

I just finished reading xmlrpclib.py and ZopeProfiler code. Here is the root cause of the issue : In my code : getattr(ServerProxy(self.url),self.rpc_method_expr) returns an xmlrpclib._Method object Then ZopeProfiler calls ZopeProfiler.ZopeProfiler.getHLFuncId(self,fn,frame), which contains: gP= getattr(s,'getPhysicalPath',None) So this calls xmlrpclib._Method.__getattr__, which is: def __getattr__(self, name): return _Method(self.__send, "%s.%s" % (self.__name, name)) So this returns another xmlrpclib._Method object for a "getPhysicalPath" RPC method Because gP is not None, ZopeProfiler then does : p= gP() And this makes an RPC method call on the service, and of course raises an Exception ! So for my tests I hacked xmlrpclib._Method.__getattr__ to return None for 'getPhysicalPath' : def __getattr__(self, name): if name=='getPhysicalPath': return None return _Method(self.__send, "%s.%s" % (self.__name, name)) But a real fix will be needed on ZopeProfiler ! Thanks. Pascal -----Message d'origine----- De : Pascal Peregrina Envoyé : lundi 27 juin 2005 10:42 À : 'zope(a)zope.org' Objet : ZopeProfiler issue Hi, In a previous mail, I was asking if anyone had issues with ZopeProfiler and Zope 2.8. I have made many more tests and my issue is completely unrelated with Zope 2.8 (I got the same issue with Zope 2.7.6) The issue has to deal with an XML-RPC call to a java service (using the standard xmlrpclib module). The code is (method of a class of my own): def query(self): from xmlrpclib import ServerProxy, loads, ProtocolError try: if self.protocol=='HTTP GET': self.result=loads(urlopen(self.url).read())[0][0] elif self.protocol=='XMLRPC': self.result=getattr(ServerProxy(self.url),self.rpc_method_expr)(*self.params ) except ProtocolError: self.result=None self.zeroconf_exception=1 self.log_exception() except SocketError: self.result=None self.zeroconf_exception=1 self.log_exception() except IOError: self.result=None self.zeroconf_exception=1 self.log_exception() except: self.result=None self.log_exception() This works fine with disabled profiler. However, when I enable the profiler, I get this error : Fault: <Fault 0: 'java.lang.Exception: RPC handler object "cat.getProducts" not found and no default handler registered\n\tat org.apache.xmlrpc.SelfDocumentingHandlerMapping.getHandler(Ljava.lang.String ;)Ljava.lang.Object;(Unknown Source)\n\tat org.apache.xmlrpc.XmlRpcWorker.execute(Ljava.io.InputStream;Lorg.apache.xmlr pc.XmlRpcContext;)[B(XmlRpcWorker.java:183)\n\tat org.apache.xmlrpc.DocXmlRpcServer.execute(Ljava.io.InputStream;Lorg.apache.x mlrpc.XmlRpcContext;)[B(Unknown Source)\n\tat org.apache.xmlrpc.DocXmlRpcServer.execute(Ljava.io.InputStream;Ljava.lang.St ring;Ljava.lang.String;)[B(Unknown Source)\n\tat org.apache.xmlrpc.DocXmlRpcWebServer$Connection.run()V(Unknown Source)\n\tat org.apache.xmlrpc.DocXmlRpcWebServer$Runner.run()V(Unknown Source)\n\tat java.lang.Thread.run()V(Optimized Method)\n\tat java.lang.Thread.startThreadFromVM(Ljava.lang.Thread;)V(Optimized Method)\n'> I can reproduce that error if I omit (*self.params) on that line : self.result=getattr(ServerProxy(self.url),self.rpc_method_expr)(*self.params ) But when I add some logging, self.params looks fine. Changing that line into : self.result=apply(getattr(ServerProxy(self.url),self.rpc_method_expr),self.p arams) Gives the same error. I also tried hardcoding the arguments, and I still get the same issue ! I don't know what to do, I really need to profile my Zope instance. So any suggestion would be highly appreciated... Thanks. Pascal ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com **********************************************************************

2 1

ZMySQLDB problems
by Dennis Allison 27 Jun '05

27 Jun '05

I am seeing occasional errors like those below. Any have any idea what my problem here is? Zope 2.7.6 MySQL-server-4.0.20-0 MySQLDB aka MySQL-python-1.2.tc3 ZMySQL 2.0.9 Python 2.4.1 Typically the errors occur at initial login when the database is consulted to populate various session variables, etc. The my.cnf file is, in part: # change timeouts to close rather than wait # interactive == 1 hour, programs == 5 mins set-variable=interactive_timeout=3600 set-variable=wait_timeout=300 # uncomment to not log connection problems #set-variable=log-warnings=off ------------------------------------------------------------------------ 2005-06-26T23:11:06 ERROR(200) ZMySQLDA exception during _begin Traceback (most recent call last): File "/opt/zope/zproducts/standard/ZMySQLDA/db.py", line 349, in _begin self.db.ping() OperationalError: (2013, 'Lost connection to MySQL server during query') ------ 2005-06-26T23:11:06 PANIC(300) ZODB A storage error occurred in the last phase of a two-phase commit. This shouldn't happen. The application will not be allowed to commit until the site/storage is reset by a restart. Traceback (most recent call last): File "/usr/local/src/zope/Zope2.7/lib/python/ZODB/Transaction.py", line 377, in _finish_many jar.tpc_finish(self) File "/usr/local/src/zope/Zope2.7/lib/python/Shared/DC/ZRDB/TM.py", line 59, in tpc_finish try: self._finish() File "/opt/zope/zproducts/standard/ZMySQLDA/db.py", line 376, in _finish self._tlock.release() error: release unlocked lock ------ 2005-06-26T23:11:06 ERROR(200) ZODB A storage error occured during object abort. This shouldn't happen. Traceback (most recent call last): File "/usr/local/src/zope/Zope2.7/lib/python/ZODB/Transaction.py", line 412, in _commit_error j.tpc_abort(self) # This should never fail File "/usr/local/src/zope/Zope2.7/lib/python/Shared/DC/ZRDB/TM.py", line 63, in abort try: self._abort() File "/opt/zope/zproducts/standard/ZMySQLDA/db.py", line 389, in _abort self._tlock.release() error: release unlocked lock ------ 2005-06-26T23:11:06 ERROR(200) SiteError http://dev.agilemind.com/logged_in Traceback (most recent call last): File "/usr/local/src/zope/Zope2.7/lib/python/ZPublisher/Publish.py", line 107, in publish transactions_manager.commit() File "/usr/local/src/zope/Zope2.7/lib/python/Zope/App/startup.py", line 222, in commit get_transaction().commit() File "/usr/local/src/zope/Zope2.7/lib/python/ZODB/Transaction.py", line 260, in commit self._finish_many(jars) File "/usr/local/src/zope/Zope2.7/lib/python/ZODB/Transaction.py", line 377, in _finish_many jar.tpc_finish(self) File "/usr/local/src/zope/Zope2.7/lib/python/Shared/DC/ZRDB/TM.py", line 59, in tpc_finish try: self._finish() File "/opt/zope/zproducts/standard/ZMySQLDA/db.py", line 376, in _finish self._tlock.release() error: release unlocked lock ------ -- Dennis Allison * Computer Systems Laboratory * Gates 227 * Stanford University * Stanford CA 94305 * (650) 723-9213 * (650) 723-0033 fax * allison(a)shasta.stanford.edu * allison(a)sumeru.stanford.edu

2 1

Timeout?
by Jan-Ole Esleben 27 Jun '05

27 Jun '05

Hi! Is there a way to have a ZOPE method (called, for example, via a TALES expression) time out after a specific (changeable!) amount of time? TIA, Ole

3 2

2.5.1 to 2.8.0
by Tim Suter 27 Jun '05

27 Jun '05

I have a Zope install 2.5.1 that I am wanting to migrate to another box that has 2.8.0. After the 2.8.0 install I import the .zexp's and copy the data.fs file from the old var to the new. This never goes successfully as I get the following when pointing to localhost:8080/manage and after I authenticate: Zope Error Zope has encountered an error while publishing this resource. Error Type: AttributeError Error Value: getItem Can anyone let me know what I might be doing wrong? I have stopped and restarted zope after making the changes. Thanks, Tim

3 6

RE: [Zope] Debugging a python routine
by Pascal Peregrina 27 Jun '05

27 Jun '05

Well, I guess that you want to debug here.lib.parse_file() as the rest is quiet obvious... If "parse_file" is a PythonScript, you can test it from the ZMI, and you should get a stack trace if an exception is raised. Pascal -----Message d'origine----- De : zope-bounces(a)zope.org [mailto:zope-bounces@zope.org]De la part de Jonathan Envoyé : lundi 27 juin 2005 17:57 À : John Poltorak; zope(a)zope.org Objet : Re: [Zope] Debugging a python routine ----- Original Message ----- From: "John Poltorak" <jp(a)warpix.org> > Can anyone suggest how I would go about debugging a Python routine like > this through Zope? > > > tal:define="opts > python:here.lib.parse_file(file=here.members,sepr=',',clone=1)"> > <tal:block repeat="opt opts"> > <li><a tal:content="python:opt[1]" > tal:attributes="href python:opt[0]"></a></li> > </tal:block> > I would start by testing the python routine on its own (ie. without any ZPT stuff) to make sure it is producing what you expect, then add in the ZPT stuff bit-by-bit. Jonathan _______________________________________________ Zope maillist - Zope(a)zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev ) ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com **********************************************************************

1 0

Re: [Zope] Debugging a python routine
by Barry Drake 27 Jun '05

27 Jun '05

More info that may be of help: * Ken Manheimer's wiki on Zope debugging at zope.org: http://zope.org/Members/klm/ZopeDebugging/FrontPage * Ken's original paper on the matter: http://zope.org/Members/klm/ZopeDebugging/ConversingWithZope * A more advanced coverage of the matter at zopewiki.org: http://zopewiki.org/DebuggingZopeWithPythonDebugger2 I got these from here: http://docs.neuroinf.de/programming-plone/debug Cheers! Barry Drake --- John Poltorak wrote: > > Can anyone suggest how I would go about debugging a > Python routine like > this through Zope? > > > tal:define="opts > python:here.lib.parse_file(file=here.members,sepr=',',clone=1)"> > <tal:block repeat="opt opts"> > <li><a tal:content="python:opt[1]" > tal:attributes="href > python:opt[0]"></a></li> > </tal:block> > > > > -- > John > > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://mail.zope.org/mailman/listinfo/zope > ** No cross posts or HTML encoding! ** > (Related lists - > http://mail.zope.org/mailman/listinfo/zope-announce > http://mail.zope.org/mailman/listinfo/zope-dev ) >

1 0

Removing old FileStorage backups from repozo
by Felix Ulrich-Oltean 27 Jun '05

27 Jun '05

Hi I'm using the repozo.py script to back up some large FileStorages, as outlined in: http://www.zope.org/Wikis/ZODB/FileStorageBackup I need to clear out old backups regularly, as there are several storages which are packed every night, and so there are lots of ~2G backups files around for every day. Can I just go in and delete files older than X days from the backup directory, or will this upset repozo, which apparently keeps info about its backup files in the .dat file? Thanks, Felix.

4 5

Re: [Zope] Debugging a python routine
by Barry Drake 27 Jun '05

27 Jun '05

I use a Script Python that writes to a ZPT. It first reads the ZPT then appends new info; first with a time stamp, container info, and the message. This works from other Script Python objects or ZPTs. It's quick and dirty; but, I've been able to use it for most of my debugging. There are other methods mentioned in any book on Zope that are more powerful. I'm sure you'll receive responses with those suggestions as well. Barry Drake USDA --- John Poltorak wrote: > > Can anyone suggest how I would go about debugging a > Python routine like > this through Zope? > > > tal:define="opts > python:here.lib.parse_file(file=here.members,sepr=',',clone=1)"> > <tal:block repeat="opt opts"> > <li><a tal:content="python:opt[1]" > tal:attributes="href > python:opt[0]"></a></li> > </tal:block> > > > > -- > John > > > > _______________________________________________ > Zope maillist - Zope(a)zope.org > http://mail.zope.org/mailman/listinfo/zope > ** No cross posts or HTML encoding! ** > (Related lists - > http://mail.zope.org/mailman/listinfo/zope-announce > http://mail.zope.org/mailman/listinfo/zope-dev ) >

1 0

ANN: MailManager 2.0 rc2 & news from us
by Andrew Veitch 27 Jun '05

27 Jun '05

To keep traffic down I don't announce every MailManager release but 2.0 is a milestone for us. This is still very much a Zope product but with 2.0 we're now storing the mail in PostgreSQL (and shortly other RDBMs too) rather than the ZODB. We are still fans of the ZODB but for our particular use case (multi terabyte storage requirement and very heavy amounts of writing) we found that PostgreSQL performed better. We have now migrated our commercial customers to the 2.0 architecture so we are pretty happy with it. Our other news is that we received a small venture capital investment in the company last week. It's just a bit over $0.5 million and we're going to use it to work through the feature suggestions we've had. I would like to thank everyone who's helped us develop MM from going to the bother of submitting a bug report or suggesting a feature or doing a translation. There's more info about our company news here if you are interested: http://www.theregister.co.uk/2005/06/23/logicalware_vc/ http://scotlandonsunday.scotsman.com/business.cfm?id=702472005 The final bit of news is that we've changed the name of the company from Logical Progression to Logicalware. It's a bit shorter and easier to type! Best regards Andrew -- Logicalware Ltd Stuart House, Eskmills, Station Road, Musselburgh EH21 7PQ, UK Tel +44 (0)131 273 5130 Web: http://www.logicalware.com/

1 0

RE: [Zope] Sharing global data between threads / locking a method
by Jim Abramson 27 Jun '05

27 Jun '05

> A possible solution: create a property field on the folder > where the external methods are stored. Have your external > method update this property field when the external method > starts and again when it exits. This way you can test > whether or not the external method is currently in operation. > You could store this property field on a temp_folder for > faster performance. Curious, - to save variables in this way, must a Folder property be used, or would a class variable set on something that extends Folder work the same? - why does temp_folder perform faster for this technique? Thx Jim

2 1