Thank you for the explaination. We are doing a graceful restart of Apache.

/Sebastian

Den fredagen den 16:e november 2012 kl. 23:58:49 UTC+1 skrev Graham 
Dumpleton:
>
> That can happen momentarily when a graceful restart of Apache is done as 
> Apache allows the child worker processes which are proxying requests to 
> mod_wgsi daemon processes to hang around until all requests on a HTTP 
> connection are handled.
>
> It comes about because the socket path has a generation number in it. 
> Every time you restart Apache the generation number goes up one. Thus an 
> older child worker created for a prior generation will be trying to connect 
> to a socket path which no longer exists, because the mod_wsgi daemon 
> processes for that older generation have been killed and replaced with 
> those corresponding to next generation.
>
> This is a self protection mechanism to ensure that old child worker 
> process that still haven't been killed and which are running against Apache 
> configuration for an older generation, are not able to connect to daemon 
> process running with a newer Apache configuration. Allowing them to would 
> produce undefined results, errors or potential security issues.
>
> Thus, so long as transitory, it is normal.
>
> Can you confirm whether do a graceful restart of Apache?
>
> It could also happen with a normal restart, but you would have to be quite 
> lucky for it to happen in that case.
>
> Graham
>
>
> On 17 November 2012 01:49, Sebastian Dahlgren 
> <[email protected]<javascript:>
> > wrote:
>
>> Hi!
>>
>> I have a mod_wsgi setup with Apache2 and Flask. If I have an active 
>> connection to the Apache2 server from a HTTP client and I reload the 
>> Apache2, the client will get a HTTP 503 back. The Apache error log says the 
>> following:
>>
>> [Fri Nov 16 14:27:30 2012] [error] [client 213.115.37.68] (2)No such file 
>> or directory: mod_wsgi (pid=19720): Unable to connect to WSGI daemon 
>> process 'upload_handler' on '/etc/apache2/run/wsgi.18907.3.1.sock' after 
>> multiple attempts.
>>
>> This is what my Apache config looks like:
>>
>> <VirtualHost *:443>
>>     ServerAdmin [email protected] <javascript:>
>>     ServerName my.example.com
>>
>>     WSGIDaemonProcess upload_handler user=www-data group=www-data 
>> threads=5 processes=4
>>     WSGIScriptAlias / /usr/local/uploader/uploader.wsgi
>>
>>     <Directory /usr/local/uploader>
>>         WSGIProcessGroup upload_handler
>>         WSGIApplicationGroup %{GLOBAL}
>>         WSGIScriptReloading Off
>>         Order deny,allow
>>         Allow from all
>>     </Directory>
>> </VirtualHost>
>>
>> I have Googled and read the mod_wsgi documentation to find an answer, but 
>> I can't find anything that matches. Does anyone here have a clue what the 
>> problem might be?
>>
>> Cheers
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "modwsgi" group.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msg/modwsgi/-/VuXuNqq6leEJ.
>> To post to this group, send email to [email protected]<javascript:>
>> .
>> To unsubscribe from this group, send email to 
>> [email protected] <javascript:>.
>> For more options, visit this group at 
>> http://groups.google.com/group/modwsgi?hl=en.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"modwsgi" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/modwsgi/-/kpa7_o3AXOsJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/modwsgi?hl=en.

Reply via email to