I have 5 databases running on different ports i.e., seperate instances and all 275 services running on single apache web server will connect to these databases
On Tuesday 22 October 2024 at 09:30:18 UTC+5:30 Graham Dumpleton wrote: > Let me rephrase that then, one database server, but with separate database > instances hosted by that server? > > Or do you really have 275 completely separate database servers? > > > On 22 Oct 2024, at 2:15 PM, Graham Dumpleton <[email protected]> > wrote: > > And it is a single database shared by all front ends right? > > On 22 Oct 2024, at 2:13 PM, RajKumar Ambadipelli <[email protected]> > wrote: > > And spike in cpu while forwarding request to datababse or sending response > to client is in between 30 to 35% of a single core. > > On Tuesday 22 October 2024 at 08:41:53 UTC+5:30 RajKumar Ambadipelli wrote: > >> Each mod_wsgi is using an average of 7 to 10mb of ram. And coming to cpu >> usage i got to see spike in cpu only while forwarding request or while >> sending response to client. >> >> And apache mpm: mpm_event_module >> >> On Tuesday 22 October 2024 at 01:22:19 UTC+5:30 Graham Dumpleton wrote: >> >>> And more questions. >>> >>> What Apache MPM are you using? Prefork, Worker or Event. >>> >>> How much memory are the Django web application processes using? IOW, the >>> mod_wsgi daemon processes. >>> >>> On 21 Oct 2024, at 5:35 PM, RajKumar Ambadipelli <[email protected]> >>> wrote: >>> >>> Only REST API type services. >>> >>> On Monday 21 October 2024 at 12:04:33 UTC+5:30 Graham Dumpleton wrote: >>> >>>> Sorry, one more question for now while think of what to suggest. >>>> >>>> Are these REST API type services, HTML page based URL handlers, or >>>> mixed of both? >>>> >>>> On 21 Oct 2024, at 5:11 PM, RajKumar Ambadipelli <[email protected]> >>>> wrote: >>>> >>>> Yes all those (275) are virtualhosts like below >>>> >>>> #MyApp1 Webservice Config >>>> Listen 9013 >>>> >>>> <VirtualHost *:9013> >>>> ErrorLog /var/log/webservice_error.log >>>> >>>> WSGIPassAuthorization On >>>> WSGIDaemonProcess 9013 >>>> python-path=/home/admin/myapp1:/home/admin/shared display-name=%{GROUP} >>>> WSGIProcessGroup 9013 >>>> >>>> WSGIApplicationGroup %{GLOBAL} >>>> WSGIScriptAlias / /home/admin/myapp1/conf/wsgi.py >>>> >>>> <Directory /home/admin/myapp1/conf> >>>> <Files wsgi.py> >>>> Require all granted >>>> </Files> >>>> </Directory> >>>> </VirtualHost> >>>> >>>> and below is config of my wsgi.py >>>> >>>> import os >>>> >>>> from django.core.wsgi import get_wsgi_application >>>> >>>> os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'conf.settings') >>>> >>>> application = get_wsgi_application() >>>> >>>> I have for 275 let's say myapp1... ,to myapp275 and each is running on >>>> a seperate wsgi daemon with same python interpreter. >>>> >>>> And all of them are different services. >>>> And all of them are having different code base. >>>> On Monday 21 October 2024 at 11:32:38 UTC+5:30 Graham Dumpleton wrote: >>>> >>>>> When you say you have around 275 services, do you mean you 275 >>>>> VirtualHost's and thus also 275 mod_wsgi daemon processes? >>>>> >>>>> How do the services differ? Are they completely different code bases, >>>>> or are they the same service for but for different customers/users >>>>> distinguished by host name? >>>>> >>>>> On 21 Oct 2024, at 4:56 PM, RajKumar Ambadipelli <[email protected]> >>>>> wrote: >>>>> >>>>> Hello Graham, >>>>> >>>>> I have python django microservice running on apache web server using >>>>> mod_wsgi in daemon mode, It takes the request and forwards to database >>>>> and >>>>> takes response from database and forwards it to front-end service. >>>>> >>>>> But like the above I have around 275 services and there is no >>>>> cpu-bound tasks on them, they simple take request and forwards to >>>>> database >>>>> and gets response from database and forwards to front-end service. >>>>> >>>>> So, All of these wsgi daemons virtualhost configurations have same >>>>> line i.e., >>>>> WSGIApplicationGroup %{GLOBAL} >>>>> all of them are using same python interpreter and there are no sub >>>>> interpreters if I am correct. >>>>> >>>>> The below is example for my one of my virtualhost config >>>>> >>>>> #MyApp1 Webservice Config >>>>> Listen 9013 >>>>> >>>>> <VirtualHost *:9013> >>>>> ErrorLog /var/log/webservice_error.log >>>>> >>>>> WSGIPassAuthorization On >>>>> WSGIDaemonProcess 9013 >>>>> python-path=/home/admin/myapp1:/home/admin/shared display-name=%{GROUP} >>>>> WSGIProcessGroup 9013 >>>>> >>>>> WSGIApplicationGroup %{GLOBAL} >>>>> WSGIScriptAlias / /home/admin/myapp1/conf/wsgi.py >>>>> >>>>> <Directory /home/admin/myapp1/conf> >>>>> <Files wsgi.py> >>>>> Require all granted >>>>> </Files> >>>>> </Directory> >>>>> </VirtualHost> >>>>> >>>>> and below is config of my wsgi.py >>>>> >>>>> import os >>>>> >>>>> from django.core.wsgi import get_wsgi_application >>>>> >>>>> os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'conf.settings') >>>>> >>>>> application = get_wsgi_application() >>>>> >>>>> Does this create any problem, Or there is any other way I can do it? >>>>> I have tried using dockers and kubernetes but for each container it is >>>>> consuming more memory and cpu compare to deploying all of them in a >>>>> single >>>>> apache web server. >>>>> I am able to run all my services on apache web server but not on >>>>> docker containers in kubernetes. >>>>> >>>>> Thanking You, >>>>> RajKumar >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> 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 view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/modwsgi/e508f452-1f63-46a1-ba7a-5f2b96a6739cn%40googlegroups.com >>>>> >>>>> <https://groups.google.com/d/msgid/modwsgi/e508f452-1f63-46a1-ba7a-5f2b96a6739cn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>> . >>>>> >>>>> >>>>> >>>> -- >>>> 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 view this discussion on the web visit >>>> https://groups.google.com/d/msgid/modwsgi/e4362b25-6078-480a-9abd-d5111aa4fd8cn%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/modwsgi/e4362b25-6078-480a-9abd-d5111aa4fd8cn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>>> >>>> >>> -- >>> 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 view this discussion on the web visit >>> https://groups.google.com/d/msgid/modwsgi/4bb11732-f75c-4e0a-bfe2-73c485fc09edn%40googlegroups.com >>> >>> <https://groups.google.com/d/msgid/modwsgi/4bb11732-f75c-4e0a-bfe2-73c485fc09edn%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >>> >>> > -- > 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 view this discussion on the web visit > https://groups.google.com/d/msgid/modwsgi/3c0865ba-c36f-43f2-bef9-5562964918ban%40googlegroups.com > > <https://groups.google.com/d/msgid/modwsgi/3c0865ba-c36f-43f2-bef9-5562964918ban%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > > > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/modwsgi/a1468122-f0fd-4f85-a8bd-81a7aa1b9ffbn%40googlegroups.com.
