These are chat archives for getredash/redash

22nd
Feb 2016
Ben Gracewood
@nzben
Feb 22 2016 02:34
So I've asked this before but am still struggling with it. What would be the likely source of a persistent "Outdated Query Count" and how would I go about diagnosing it? I'm guessing it's some bad queries that can't execute so the results aren't updating?
Arik Fraimovich
@arikfr
Feb 22 2016 05:11
@nzben this is my guess too. Last time someone asked this question I posted a query that can help you find these queries. Let me know if you can't find it in the message history, I'll try to find it in my redash instance.
Ben Gracewood
@nzben
Feb 22 2016 18:18
The only thing I can see that might be related is a list of query result by date?
Arik Fraimovich
@arikfr
Feb 22 2016 19:02
@nzben this one:
select q.id, name, schedule, retrieved_at
from queries q
join query_results qr on qr.id = q.latest_query_data_id
where retrieved_at < current_date - 1 and schedule is not null
should find scheduled queries that were not refreshed in the past day. you might want to calibrate the threshold
Scott Reynen
@sreynen
Feb 22 2016 19:32
hi! after editing HTML files in rd_ui/app, what are the steps to see those changes in the UI?
i expected the steps to be gulp build and supervisorctl restart redash_server, but still not seeing my changes
Arik Fraimovich
@arikfr
Feb 22 2016 19:39
@sreynen what version of redash do you have? Try clearing browser cache
Scott Reynen
@sreynen
Feb 22 2016 19:40
i just did a git clone this morning. says 0.9.1+b1377
still no changes after clearing browser cache. does nginx cache those files?
Arik Fraimovich
@arikfr
Feb 22 2016 19:53
No it doesn't. What is the path of the file you're changing?
Scott Reynen
@sreynen
Feb 22 2016 20:01
rd_ui/app/views/new_widget_form.html
i also tried changing rd_ui/app/views/dashboard.html and rd_ui/app/robots.txt and don't see my changes to those either
Scott Reynen
@sreynen
Feb 22 2016 20:09
hmm, i think maybe it's not using my copy of redash at all. is the /opt/redash/current symlink not what controls the active version?
Ben Gracewood
@nzben
Feb 22 2016 20:12
Thanks @arikfr I'll give that a crack later today.
Scott Reynen
@sreynen
Feb 22 2016 20:15
yeah, it's definitely loading from /opt/redash/redash.0.9.1.b1377 even though i have current symlinked to a different copy
so how do i tell supervisor to use /opt/redash/current?
Scott Reynen
@sreynen
Feb 22 2016 20:49
restarting my VM seems to have reloaded /opt/redash/current, but now nginx is throwing 502 Bad Gateway, though supervisorctl status says redash_server is running
is it just a bad idea to try to run this in a VM?
Arik Fraimovich
@arikfr
Feb 22 2016 21:38
It does use current. If for some reason restart doesn't pick up the right one, try restarting supervisor itself.
As for 502, try checking the logs to identify the problem.
I don't see an issue with using a vm as long as it had enough resources.
Scott Reynen
@sreynen
Feb 22 2016 21:39
i tried restarting supervisor itself, but got some error i didn't understand, so i restarted the VM, which fixed that part
the nginx error log says 2016/02/22 20:49:00 [error] 1097#0: *1 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.33.1, server: , request: "GET /favicon.ico HTTP/1.1", upstream: "http://127.0.0.1:5000/favicon.ico", host: "192.168.33.10", referrer: "http://192.168.33.10/dashboard/test" which i don't know what to do with
http://192.168.33.10/ is my VM, and it looks like i could connect directly to redash at http://192.168.33.10:5000/ but that gets ERR_CONNECTION_REFUSED
oh, i see now that supervisor is restarting redash_server every 2 seconds. so i guess i need to figure out why
Scott Reynen
@sreynen
Feb 22 2016 21:45
api_error.log appears to have a python error, which is something i know how to debug
Scott Reynen
@sreynen
Feb 22 2016 21:52
it looks like maybe the DB schema has changed between the first version i ran and the git clone i'm now running. is there some way to update it?
i'm getting ProgrammingError: relation "organizations" does not exist LINE 1: ...", "t1"."name", "t1"."slug", "t1"."settings" FROM "organizat...
Scott Reynen
@sreynen
Feb 22 2016 22:54
i solved that by dropping the schema in postgress and re-running manage.py database create_tables