Hello, We having been using uwsgi and nginx with much success. However recently we have been seeing these errors. Nginx is connecting to uwsgi via local domain socket.
2013/08/12 16:15:33 [error] 1502#0: *825741 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *.clearcareonline.com, request: "GET /clients/31538/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " hcasandiego.clearcareonline.com", referrer: " https://hcasandiego.clearcareonline.com/clients/31537/"; 2013/08/12 16:15:33 [error] 1502#0: *825743 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *.clearcareonline.com, request: "GET /clients/67959/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " privatehc.clearcareonline.com", referrer: " https://privatehc.clearcareonline.com/clients/67959/"; 2013/08/12 16:15:33 [error] 1502#0: *825745 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *.clearcareonline.com, request: "GET /settings/admins/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " renew.clearcareonline.com", referrer: " https://renew.clearcareonline.com/settings/"; 2013/08/12 16:15:33 [error] 1502#0: *825747 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.126, server: *.clearcareonline.com, request: "GET /carelogs/edit-carelog/4712190/?_=1376324137022 HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " wpc.clearcareonline.com", referrer: " https://wpc.clearcareonline.com/accounting/billing/"; 2013/08/12 16:15:33 [error] 1502#0: *825751 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *.clearcareonline.com, request: "POST /event-notification/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " admin.clearcareonline.com", referrer: " https://admin.clearcareonline.com/event-notification/"; 2013/08/12 16:15:33 [error] 1502#0: *825753 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.126, server: *.clearcareonline.com, request: "POST /event-notification/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " admin.clearcareonline.com", referrer: " https://admin.clearcareonline.com/event-notification/"; 2013/08/12 16:15:34 [error] 1502#0: *825756 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *. clearcareonline.com, request: "GET /scheduling/add-shift/38743/?start=201309011200&end=201309011400&_=1376324130909 HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " hcasaltlake.clearcareonline.com", referrer: " https://hcasaltlake.clearcareonline.com/dashboard/live/weekly/?weeks=2"; 2013/08/12 16:15:34 [error] 1502#0: *825758 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *. clearcareonline.com, request: "GET /carelogs/edit-carelog/4803202/?_=1376324134812 HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " aberdeen.clearcareonline.com", referrer: " https://aberdeen.clearcareonline.com/accounting/payroll/"; 2013/08/12 16:15:34 [error] 1503#0: *825762 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.127, server: *. clearcareonline.com, request: "POST /apply/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " cacaregivers.clearcareonline.com", referrer: " https://cacaregivers.clearcareonline.com/apply/"; 2013/08/12 16:15:34 [error] 1503#0: *825764 connect() to unix://home/ec2-user/run/uwsgi.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: 10.0.0.126, server: *. clearcareonline.com, request: "GET /clients/79621/ HTTP/1.1", upstream: "uwsgi://unix://home/ec2-user/run/uwsgi.sock:", host: " visitingangelsriverside.clearcareonline.com", referrer: " https://visitingangelsriverside.clearcareonline.com/clients/"; Here are the uwsgi logs from that time period. Aug 12 16:15:31 localhost webex3: {address space usage: 939761664 bytes/896MB} {rss usage: 406831104 bytes/387MB} [pid: 35114|app: 0|req: 10487/30014] 10.0.0.126 () {48 vars in 1890 bytes} [Mon Aug 12 16:15:28 2013] GET /dashboard/get-tasks/?assigned_to=41411&date_filter=today_overdue&page=1&_=1376324127862 => generated 129 bytes in 3608 msecs (HTTP/1.1 200) 6 headers in 206 bytes (1 switches on core 0) Aug 12 16:15:34 localhost webex3: {address space usage: 926412800 bytes/883MB} {rss usage: 358572032 bytes/341MB} [pid: 35110|app: 0|req: 3991/30017] 10.0.0.127 () {46 vars in 3966 bytes} [Mon Aug 12 16:15:30 2013] GET /clients/12282/ => generated 55473 bytes in 4415 msecs (HTTP/1.1 200) 7 headers in 330 bytes (1 switches on core 0) Aug 12 16:15:34 localhost webex3: {address space usage: 819707904 bytes/781MB} {rss usage: 256200704 bytes/244MB} [pid: 35102|app: 0|req: 84/30016] 10.0.0.127 () {46 vars in 721 bytes} [Mon Aug 12 16:15:31 2013] POST /event-notification/ => generated 0 bytes in 3530 msecs (HTTP/1.1 200) 6 headers in 204 bytes (1 switches on core 0) Aug 12 16:15:34 localhost webex3: {address space usage: 843046912 bytes/803MB} {rss usage: 289374208 bytes/275MB} [pid: 35106|app: 0|req: 566/30015] 10.0.0.126 () {48 vars in 1855 bytes} [Mon Aug 12 16:15:30 2013] GET /dashboard/live/weekly/?weeks=0 => generated 760278 bytes in 3739 msecs (HTTP/1.1 200) 7 headers in 331 bytes (4 switches on core 1) This is the only relevant kernel setting I've been able to find. [ec2-user@webex3 ~]$ cat /proc/sys/net/unix/max_dgram_qlen 1024 I raised it from the default of 10 but it sure seems like its hitting that limit. Aug 12 16:42:14 localhost webex3: *** Starting uWSGI 1.9.12 (64bit) on [Mon Aug 12 16:42:10 2013] *** Aug 12 16:42:14 localhost webex3: compiled with version: 4.6.3 20120306 (Red Hat 4.6.3-2) on 28 June 2013 23:00:18 Aug 12 16:42:14 localhost webex3: os: Linux-3.4.37-40.44.amzn1.x86_64 #1 SMP Thu Mar 21 01:17:08 UTC 2013 Aug 12 16:42:14 localhost webex3: nodename: webex3 Aug 12 16:42:14 localhost webex3: machine: x86_64 Aug 12 16:42:14 localhost webex3: clock source: unix Aug 12 16:42:14 localhost webex3: detected number of CPU cores: 4 Does anyone have any ideas on what could be causing the errors? Thanks, Derrick
_______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
