There is not enough information to say much. if possible try to stream directly rather than buffering data first in a variable, this should make things faster and lighter.
2013/10/7 Niphlod <niph...@gmail.com> > did you try upping the memory limits ? > limit-as, reload-on-as, reload-on-rss > > > PS: dumping 50k records in a csv is definitely not something for a > webserver (timeout, anyone ?) > > On Monday, October 7, 2013 10:46:34 PM UTC+2, viniciusban wrote: >> >> Hi guys. >> >> I'm experiencing a problem I think is not web2py fault, but uwsgi or >> nginx configuration. >> >> I'm running web2py inside a Vagrant box (Virtualbox) with nginx. >> >> I have a table with more than 50k lines and when I try to export as >> csv in admin, the generated file doesn't have all records. Each time >> it generates a >> different number of rows. >> >> >> Reading logs, I found something: >> >> in /var/log/nginx/error.log: >> 2013/10/07 20:25:15 [crit] 955#0: *291 open() >> "/var/lib/nginx/uwsgi/9/00/**0000000009" failed (13: Permission denied) >> while reading upstream, client: 10.11.12.1, server: wsl, request: "GET >> /wsl/appadmin/csv/db?query=db.**publicacao.id <http://db.publicacao.id>%3E0 >> HTTP/1.1", upstream: >> "uwsgi://unix:///tmp/web2py.**socket:", host: "10.11.12.13", referrer: >> "https://10.11.12.13/wsl/**appadmin/select/db?query=db.** >> publicacao.id%3E0<https://10.11.12.13/wsl/appadmin/select/db?query=db.publicacao.id%3E0>" >> >> >> >> in /var/log/uwsgi/uwsgi.log: >> SIGPIPE: writing to a closed pipe/socket/fd (probably the client >> disconnected) on request >> /wsl/appadmin/csv/db?query=db.**publicacao.id <http://db.publicacao.id>%3E0 >> (ip 10.11.12.1) !!! >> uwsgi_response_write_body_do()**: Broken pipe [core/writer.c line 296] >> IOError: write error >> [pid: 2168|app: 0|req: 53/153] 10.11.12.1 () {46 vars in 1015 bytes} >> [Mon Oct 7 20:25:06 2013] GET >> /wsl/appadmin/csv/db?query=db.**publicacao.id <http://db.publicacao.id>%3E0 >> => generated 0 bytes >> in 9064 msecs (HTTP/1.1 200) 7 headers in 371 bytes (4 switches on >> core 0) >> ...The work of process 2168 is done. Seeya! >> worker 4 killed successfully (pid: 2168) >> Respawned uWSGI worker 4 (new pid: 2326) >> mapping worker 4 to CPUs: 1 >> >> >> >> >> What is happening? It seems somehow the socket connection is being >> lost. Do you have an idea to fix that? >> >> BTW, I'm using webp2y v2.4.6 on Ubuntu 12.04 server and it was configured >> by >> setup-web2py-nginx-uwsgi-**ubuntu.sh distributed along with web2py. >> >> My /etc/uwsgi/web2py.ini file is: >> [uwsgi] >> socket = /tmp/web2py.socket >> pythonpath = /home/vagrant/web2py/ >> mount = /=wsgihandler:application >> processes = 4 >> master = true >> harakiri = 60 >> reload-mercy = 8 >> cpu-affinity = 1 >> stats = /tmp/stats.socket >> max-requests = 2000 >> limit-as = 512 >> reload-on-as = 256 >> reload-on-rss = 192 >> uid = vagrant >> gid = vagrant >> cron = 0 0 -1 -1 -1 python /home/vagrant/web2py/web2py.py -Q -S >> welcome -M -R scripts/sessions2trash.py -A -o >> no-orphans = true >> >> >> >> Any help? >> > -- > Resources: > - http://web2py.com > - http://web2py.com/book (Documentation) > - http://github.com/web2py/web2py (Source code) > - https://code.google.com/p/web2py/issues/list (Report Issues) > --- > You received this message because you are subscribed to the Google Groups > "web2py-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to web2py+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
