On 06/03/2015, at 9:20 PM, [email protected] wrote: > Sorry Graham my mistake the mod_wsgi i am using is 3.2 which i had to compile > myself, i downloaded the source code and did the compile. I am on a Solaris > 10 and for web server i am using Oracle OHS (Oracle HTTP Server). > > The problem is that i am having issues with compiling the latest version of > mod_wsgi in my solaris. > > Can you help me with that ??
Sure, what is the specific problem/error with the very latest mod_wsgi version? Graham > On Friday, March 6, 2015 at 12:09:47 PM UTC+2, Graham Dumpleton wrote: > Do you really mean 3.1.2, or perhaps 3.1 where some OS distribution is > tagging on a '-2' qualifier? > > Although there was a version 3.1, there never was a 3.1.2 as I only started > using 3 part version numbers with 4.1.0. > > If it is 3.1, which if it is Oracle Linux I wouldn't be surprised, then it > was released 4 1/2 years ago and is way out of date. > > Is there any chance at all you can upgrade to a recent version of mod_wsgi by > compiling from mod_wsgi source code yourself? > > In using such an old version of mod_wsgi I am not sure there is much I can do > to help you. If it is an issue which was since fixed, you wouldn't have much > choice but to upgrade anyway. > > Graham > > On 06/03/2015, at 2:02 AM, [email protected] wrote: > >> Server version: Oracle-HTTP-Server/2.2.22 >> >> mod_wsgi: 3.1.2 >> >> Can this be the cause of the DB errors i am getting i.e. the process is >> crashing and with it together the DB connection ???? >> >> Regards, >> Alex >> >> >> >> On Thursday, March 5, 2015 at 12:46:23 PM UTC+2, Graham Dumpleton wrote: >> What version of mod_wsgi are you using? >> >> What patch level revision of Apache are you using? >> >> The crash of first daemon process instance but not subsequent ones sounds a >> bit like a bug I fixed not long back. >> >> I can't right now remember what the issue was and can't find any note about >> it in release notes for 4.4.X versions. >> >> Graham >> >> On 04/03/2015, at 10:47 PM, [email protected] wrote: >> >>> Shouldn't this cause a core dump ?? Because i can not find any in order to >>> analyze it.... >>> >>> On Wednesday, March 4, 2015 at 1:36:45 PM UTC+2, [email protected] >>> wrote: >>> Yes i tried it and the same thing happens: >>> >>> [Wed Mar 04 13:31:53 2015] [notice] child pid 10101 exit signal >>> Segmentation fault (11) >>> [Wed Mar 04 13:31:54 2015] [debug] proxy_util.c(1818): proxy: grabbed >>> scoreboard slot 0 in child 10148 for worker proxy:reverse >>> [Wed Mar 04 13:31:54 2015] [debug] proxy_util.c(1837): proxy: worker >>> proxy:reverse already initialized >>> [Wed Mar 04 13:31:54 2015] [debug] proxy_util.c(1914): proxy: initialized >>> worker 0 in child 10148 for (*) min=0 max=25 smax=25 >>> [Wed Mar 04 13:31:54 2015] [info] mod_wsgi (pid=10148): Initializing Python. >>> [Wed Mar 04 13:31:54 2015] [info] mod_wsgi (pid=10148): Attach interpreter >>> ''. >>> >>> It crashes the first interpreter and its attaching a new one... >>> >>> On Wednesday, March 4, 2015 at 12:42:32 PM UTC+2, Graham Dumpleton wrote: >>> Are you setting: >>> >>> WSGIApplicationGroup %{GLOBAL} >>> >>> directive. >>> >>> http://code.google.com/p/modwsgi/wiki/ApplicationIssues#Python_Simplified_GIL_State_API >>> >>> That is often a good remedy for process crashes when using Python third >>> party extensions modules which aren't implemented properly to work in sub >>> interpreters. >>> >>> Graham >>> >>> On 04/03/2015, at 9:25 PM, [email protected] wrote: >>> >>>> Also from the logs its seems that there are some Segmentation faults >>>> >>>> mod_wsgi (pid=9760): Create interpreter >>>> ........ >>>> >>>> child pid 9760 exit signal Segmentation fault (11) >>>> >>>> Is this normal ??? >>>> >>>> On Monday, March 2, 2015 at 12:39:02 PM UTC+2, [email protected] >>>> wrote: >>>> Hi Graham, >>>> >>>> In my project i already have DEBUG = True which i think it overwrites >>>> global settings. >>>> >>>> Is DEBUG = True enough to force Django to log details of unhanded >>>> exceptions?? Or should i do anything else? >>>> >>>> I already have print outs of os.environ and i have checked that >>>> ORACLE_HOME and TNS_ADMIN is set to the correct paths , i think that those >>>> are the only 2 variables needed for ORACLE... If i am wrong please correct >>>> me. >>>> >>>> I do not know what else should i do in order to get to log all the >>>> exceptions in the logs.... >>>> >>>> But i think this is like the other case you help me solve with >>>> "hashlib"... The true problem/exception was never logged, until you told >>>> me to try to import _hashlib instead of hashlib. THEN AND OLNY then the >>>> problem was clear and i was able to fix it. I think that the same >>>> circumstances apply here... Somehow i am not getting the real issue... >>>> >>>> Regards, >>>> Alex >>>> >>>> On Friday, February 27, 2015 at 11:47:28 PM UTC+2, Graham Dumpleton wrote: >>>> If this is only the latter consequence, then have you got Django setup to >>>> send you details of unhanded exceptions in emails or otherwise log them. >>>> If you don't then Django will simply translate an error into a generic 500 >>>> response page and you will not see the error anywhere. >>>> >>>> If you are on a development system, you can alternatively enable >>>> temporarily in the Django settings file: >>>> >>>> DEBUG = True >>>> >>>> and if it was creating a 500 page then it will then include the exception >>>> details that caused it. >>>> >>>> BTW other possible problems with Oracle specifically are missing >>>> ORACLE_??? environment variables set in environment of Apache. >>>> >>>> If you add debug statements in WSGI script file to print out contents of >>>> os.environ from 'os' module, dod you see all the ORACLE_?? environment >>>> variables you expect to see that might be set in the shell of your user >>>> environment. >>>> >>>> Graham >>>> >>>> On 28/02/2015, at 8:40 AM, [email protected] wrote: >>>> >>>>> Hi Graham, >>>>> >>>>> Can you elaborate more on this... Based on the Django project i have >>>>> created, in the settings for the Oracle DB i have provided the user (db >>>>> schema) the application should use to connect to the db. >>>>> >>>>> I have not changed anything since i migrated the project to the new >>>>> infrastructure and the project was working fine before. I am on the same >>>>> machine i just installed everything again basically.. >>>>> >>>>> I know that somewhere there is an exception occurring, but the problem is >>>>> that it is never caught and printed in the logs.... >>>>> >>>>> As you can see in the logs the only errors are when it tries to close the >>>>> connection and most probably never finds one.. either because it was >>>>> never created or it is closed before everything else is executed. >>>>> >>>>> I want to find a way to debug this ... Is there a way to trace each call >>>>> and every step so i can place some sort of print to see where the code is >>>>> not working ... Because i do not know any other way to fix this. >>>>> >>>>> I another problem i had, you were the one that suggested something that >>>>> relieved the actual problem and i was able to fix it, because before that >>>>> the only errors in the logs i was seeing were not relevant to the >>>>> problem. >>>>> >>>>> Thanks >>>>> >>>>> On Thursday, February 26, 2015 at 11:42:26 AM UTC+2, Graham Dumpleton >>>>> wrote: >>>>> If Oracle is providing access based on the identity of the user >>>>> connecting though local UNIX socket connection to the database, it could >>>>> be failing because your application will be running as the Apache user >>>>> and maybe you haven't configured Oracle to allow that user to connect. >>>>> >>>>> Graham >>>>> >>>>> On 25/02/2015, at 10:09 PM, [email protected] wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> I migrated my project to a new infrastructure but i kept the same >>>>>> versions Apache 2.2 + mod_wsgi + python 2.6.1 + Django 1.2.1 + ....... >>>>>> >>>>>> I am trying to run my project or even access the admin console of Django >>>>>> /admin and i can not. >>>>>> >>>>>> I tried to connect directly from the Oracle Client (sqlplus) and i can, >>>>>> i created a simple python script that uses cx_oracle and i am able to >>>>>> connect to the database without any problems, i have tested "python >>>>>> manage.py dbshell" and it connects. I also used "python manage.py >>>>>> runserver" and by navigating to http://127.0.0.1:8000/admin i was able >>>>>> to access the console without any errors, but when i try using apache + >>>>>> mod_wsgi i get the following error: >>>>>> >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] mod_wsgi >>>>>> (pid=27565): Exception occurred processing WSGI script >>>>>> '/opt/wgt_proxy/wgtproxyProj/ >>>>>> wgtproxy/apache/django.wsgi'. >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] Traceback (most >>>>>> recent call last): >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] File >>>>>> "/opt/wgt_proxy/wgtproxyProj/wgtproxy/apache/django.wsgi", line 50, in >>>>>> __call__ >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] return >>>>>> self.__application(environ, _start_response) >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] File >>>>>> "/opt/webtier/python_64/lib/python2.6/site-packages/django/core/handlers/wsgi.py", >>>>>> line 248, in __call__ >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] >>>>>> signals.request_finished.send(sender=self.__class__) >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] File >>>>>> "/opt/webtier/python_64/lib/python2.6/site-packages/django/dispatch/dispatcher.py", >>>>>> line 162, in send >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] response = >>>>>> receiver(signal=self, sender=sender, **named) >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] File >>>>>> "/opt/webtier/python_64/lib/python2.6/site-packages/django/db/__init__.py", >>>>>> line 82, in close_connection >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] conn.close() >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] File >>>>>> "/opt/webtier/python_64/lib/python2.6/site-packages/django/db/backends/__init__.py", >>>>>> line 70, in close >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] >>>>>> self.connection.close() >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] >>>>>> OperationalError: ORA-03114: not connected to ORACLE >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] >>>>>> [Fri Feb 20 17:40:51 2015] [error] [client xx.xx.xx.xx] Request Failed >>>>>> for : /wgtproxy/admin/, Resp Code : [500] >>>>>> >>>>>> Can anyone help me solve my issue? >>>>>> >>>>>> Thanks >>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "modwsgi" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>>> an email to [email protected]. >>>>>> To post to this group, send email to [email protected]. >>>>>> Visit this group at http://groups.google.com/group/modwsgi. >>>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google Groups >>>>> "modwsgi" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send an >>>>> email to [email protected]. >>>>> To post to this group, send email to [email protected]. >>>>> Visit this group at http://groups.google.com/group/modwsgi. >>>>> For more options, visit https://groups.google.com/d/optout. >>>> >>>> >>>> -- >>>> You received this message because you are subscribed to the Google Groups >>>> "modwsgi" group. >>>> To unsubscribe from this group and stop receiving emails from it, send an >>>> email to [email protected]. >>>> To post to this group, send email to [email protected]. >>>> Visit this group at http://groups.google.com/group/modwsgi. >>>> For more options, visit https://groups.google.com/d/optout. >>> >>> >>> -- >>> You received this message because you are subscribed to the Google Groups >>> "modwsgi" group. >>> To unsubscribe from this group and stop receiving emails from it, send an >>> email to [email protected]. >>> To post to this group, send email to [email protected]. >>> Visit this group at http://groups.google.com/group/modwsgi. >>> For more options, visit https://groups.google.com/d/optout. >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "modwsgi" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at http://groups.google.com/group/modwsgi. >> For more options, visit https://groups.google.com/d/optout. > > > -- > You received this message because you are subscribed to the Google Groups > "modwsgi" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/modwsgi. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "modwsgi" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/modwsgi. For more options, visit https://groups.google.com/d/optout.
