error on production

[derekeder]

http://chicagoelections.datamade.us/war-chest is throwing an error.

Dug in to the nginx error logs and found this:

2014/04/16 19:40:12 [error] 25091#0: *3 upstream prematurely closed connection while reading response header from upstream, client: 50.200.5.115, server: chicagoelections.datamade.us, request: "GET /war-chest/ HTTP/1.1", upstream: "http://127.0.0.1:7777/war-chest/", host: "chicagoelections.datamade.us"

Though that doesn't mean anything to me. Is there a better place to see python level error messages?

[evz]

It probably means the flask app is taking too long to respond. This is likely because that database is way huger now (since I scraped all the committees).

Two things you can do to fix this: increase the gunicorn timeout and increase the nginx timeout. The gunicorn process is being managed by upstart so the conf file for that is /etc/init/war-chest.conf . In that file you'll see a line that starts the gunicorn process. To increase the timeout, you can pass a -t flag and give it a timeout in seconds. The restart the app with sudo restart war-chest. For nginx, you'll need to set

proxy_connect_timeout 600; proxy_send_timeout 600; proxy_read_timeout 600; send_timeout 600;

if it's not already set. Actually, 600 seconds is probably to generous but you get the idea.

Hope that helps,

Eric

On Wednesday, April 16, 2014, Derek Eder notifications@github.com wrote:

http://chicagoelections.datamade.us/war-chest is throwing an error.

Dug in to the nginx error logs and found this:

2014/04/16 19:40:12 [error] 25091#0: *3 upstream prematurely closed connection while reading response header from upstream, client: 50.200.5.115, server: chicagoelections.datamade.us, request: "GET /war-chest/ HTTP/1.1", upstream: "http://127.0.0.1:7777/war-chest/", host: "chicagoelections.datamade.us"

Though that doesn't mean anything to me. Is there a better place to see python level error messages?

— Reply to this email directly or view it on GitHubhttps://github.com/datamade/war-chest/issues/33 .

[derekeder]

@evz thanks - not a critical thing to fix right now. we can wait till you get back.

I think we should look in to fixing the query that takes so long instead of increasing the timout for the site.

On Wed, Apr 16, 2014 at 5:12 PM, Eric van Zanten notifications@github.comwrote:

It probably means the flask app is taking too long to respond. This is likely because that database is way huger now (since I scraped all the committees).

Two things you can do to fix this: increase the gunicorn timeout and increase the nginx timeout. The gunicorn process is being managed by upstart so the conf file for that is /etc/init/war-chest.conf . In that file you'll see a line that starts the gunicorn process. To increase the timeout, you can pass a -t flag and give it a timeout in seconds. The restart the app with sudo restart war-chest. For nginx, you'll need to set

proxy_connect_timeout 600; proxy_send_timeout 600; proxy_read_timeout 600; send_timeout 600;

if it's not already set. Actually, 600 seconds is probably to generous but you get the idea.

Hope that helps,

Eric

On Wednesday, April 16, 2014, Derek Eder notifications@github.com wrote:

http://chicagoelections.datamade.us/war-chest is throwing an error.

Dug in to the nginx error logs and found this:

2014/04/16 19:40:12 [error] 25091#0: *3 upstream prematurely closed connection while reading response header from upstream, client: 50.200.5.115, server: chicagoelections.datamade.us, request: "GET /war-chest/ HTTP/1.1", upstream: "http://127.0.0.1:7777/war-chest/", host: "chicagoelections.datamade.us"

Though that doesn't mean anything to me. Is there a better place to see python level error messages?

— Reply to this email directly or view it on GitHub< https://github.com/datamade/war-chest/issues/33> .

— Reply to this email directly or view it on GitHubhttps://github.com/datamade/war-chest/issues/33#issuecomment-40659039 .

Derek Eder @derekeder https://twitter.com/#!/derekeder derekeder.com derek.eder@gmail.com