These are chat archives for getredash/redash

28th
Mar 2016
ChiragKParmar
@ChiragKParmar
Mar 28 2016 01:05
@arikfr no luck, still getting stuck at the same stage. Any other way of upgrading redash?
Arik Fraimovich
@arikfr
Mar 28 2016 05:59
@ChiragKParmar it looks like it stuck on downloading from GitHub. Is it possible your Redash instance can't access the internet?
ChiragKParmar
@ChiragKParmar
Mar 28 2016 17:01
@arikfr It does have internet connection.
Mariusz Litwin
@mariuszlitwin
Mar 28 2016 17:05
@ChiragKParmar I had some issues in corporate network when due to some company traffic limitations I wasn't able to access sftp, git or ssh. Have you checked on this? I had this super-duper-simple check where I tried to pip some git repository by 'pip install git+git://...' and in corporate network I received timeout.
ChiragKParmar
@ChiragKParmar
Mar 28 2016 17:06
hm..I better check that! Thanks @mariuszlitwin How did you solve it then?
Mariusz Litwin
@mariuszlitwin
Mar 28 2016 17:08
@ChiragKParmar As I have used local server I simply plugged in my mobile and used mobile network. You can try some SSH tunnel or VPN if any of this is working.
ChiragKParmar
@ChiragKParmar
Mar 28 2016 17:09
Cool! Let me try! Thanks @mariuszlitwin
Mariusz Litwin
@mariuszlitwin
Mar 28 2016 17:11
@ChiragKParmar Hope that helps :)
Mariusz Litwin
@mariuszlitwin
Mar 28 2016 17:27
Have someone tried integrating redash with Google DBs other than Big Query? In specific I'm interested in Google Datastore as their other services are provided with some kind of proxy from machine to their cloud.
Arik Fraimovich
@arikfr
Mar 28 2016 17:58
@mariuszlitwin I don't know of anyone who did this, but it should be possible and probably simple.
Mariusz Litwin
@mariuszlitwin
Mar 28 2016 17:59
@arikfr I will look if I can handle this by myself. Hope that I will be able to contribute.
Arik Fraimovich
@arikfr
Mar 28 2016 18:00
It will be great and very appreciated!
Sean Auriti
@seanauriti
Mar 28 2016 18:11
Hi there! I've recently updated and need to run the migrations, however the fabfile is not working, maybe because I only allowed VPN access, is there a command to run the migrations from the command line?
Arik Fraimovich
@arikfr
Mar 28 2016 18:12
@seanauriti unfortunately there is no automatic command. Basically what the fabfile does is:
Sean Auriti
@seanauriti
Mar 28 2016 18:13
i tried this in the migrations folder - sudo sudo -u redash PYTHONPATH=. python *
Arik Fraimovich
@arikfr
Mar 28 2016 18:13
  1. Find new migration by diffing current folder and new version: https://gist.github.com/arikfr/440d1403b4aeb76ebaf8#file-fabfile-py-L57
  2. Run each migration found in step #1: https://gist.github.com/arikfr/440d1403b4aeb76ebaf8#file-fabfile-py-L94
@seanauriti that's not a good idea, as the migration are not idempotent...
Sean Auriti
@seanauriti
Mar 28 2016 18:13
ok
I'll try to get the fabfile working
Arik Fraimovich
@arikfr
Mar 28 2016 18:13
@seanauriti if you already this, it might have messed up your database. if it's a fresh install, you can just drop the db and recreate it.
Sean Auriti
@seanauriti
Mar 28 2016 18:14
the command did not work because of the ENV
Arik Fraimovich
@arikfr
Mar 28 2016 18:14
oh, good :)
Sean Auriti
@seanauriti
Mar 28 2016 18:14
:)
Arik Fraimovich
@arikfr
Mar 28 2016 18:14
then just run them manually
Sean Auriti
@seanauriti
Mar 28 2016 18:14
how can I do that?
with the fabfile you mean?
Arik Fraimovich
@arikfr
Mar 28 2016 18:15
I mean, you can run the commands the fab file runs (see the lines I've linked to)
or if you're brave enough, you can try this version of the fabfile that is supposed to run locally on the redash server : https://gist.github.com/idanh/457795c77b653e764731
let me know if it worked
Sean Auriti
@seanauriti
Mar 28 2016 18:18
I am trying the local fabfile, I ran it like this fab apply_migrations - it says TypeError: apply_migrations() takes exactly 1 argument (0 given)
probably missing a command line argument?
Arik Fraimovich
@arikfr
Mar 28 2016 18:18
yes, it's need the new version as param. but you can just run the deploy task again
it will override previous download
Sean Auriti
@seanauriti
Mar 28 2016 18:18
ok cool

Warning: local() encountered an error (return code 2) while executing 'diff <(cd /opt/redash/redash.0.10.0.b1664/migrations && find . | sort) <(cd /opt/redash/current/migrations && find . | sort)'

No new migrations in this version.

I may have to relink the old version
Arik Fraimovich
@arikfr
Mar 28 2016 18:23
This message was deleted
yep
Sean Auriti
@seanauriti
Mar 28 2016 18:24
same error even after relinking
Arik Fraimovich
@arikfr
Mar 28 2016 18:24
strange
Sean Auriti
@seanauriti
Mar 28 2016 18:24
is there a version token that needs to switch back?
Arik Fraimovich
@arikfr
Mar 28 2016 18:24
nope, just /opt/redash/current need to link to the old version
Sean Auriti
@seanauriti
Mar 28 2016 18:24
hmm
are there no new migrations from redash.0.9.1.b1377 to redash.0.10.0.b1664 ?
there are
< ./0021_create_api_keys_table.py
< ./0022_add_new_permissions.py
< ./0023_make_sure_correct_group_type.py
Arik Fraimovich
@arikfr
Mar 28 2016 18:26
yep
Sean Auriti
@seanauriti
Mar 28 2016 18:27
This message was deleted
This message was deleted
This message was deleted
This message was deleted
Arik Fraimovich
@arikfr
Mar 28 2016 18:33
? :)
Sean Auriti
@seanauriti
Mar 28 2016 18:33
Warning: local() encountered an error (return code 1) while executing 'diff /opt/redash/current/requirements.txt /opt/redash/redash.0.10.0.b1664/requirements.txt'
Warning: local() encountered an error (return code 2) while executing 'diff <(cd /opt/redash/redash.0.10.0.b1664/migrations && find . | sort) <(cd /opt/redash/current/migrations && find . | sort)'
get both of those
i'm running
fab deploy_latest_release
Arik Fraimovich
@arikfr
Mar 28 2016 18:36
strange. maybe just apply those migrations manually and be done with it:
from the new version folder, run PYTHONPATH=. bin/run python migrations/0021_.... for each one of them in their order.
then run pip install for the new requirements file , symlink and restart
Sean Auriti
@seanauriti
Mar 28 2016 18:38
This message was deleted
peewee.OperationalError: FATAL: role "ubuntu" does not exist
getting that, tried sudo and get same thing for "root"
Arik Fraimovich
@arikfr
Mar 28 2016 18:40
right, prefix the above with sudo -u redash
Sean Auriti
@seanauriti
Mar 28 2016 18:40
ok thank you
back in business! thanks!
is there a way to export / import all queries?
Arik Fraimovich
@arikfr
Mar 28 2016 18:44
for backup purposes?
Sean Auriti
@seanauriti
Mar 28 2016 18:44
yes
and quick import of queries
I can add that feature request...
Arik Fraimovich
@arikfr
Mar 28 2016 18:48
I've recently wrote a simple script to export queries into a folder, but it's a one way process: https://gist.github.com/arikfr/598590356c4da18be976
For backup purposes, I recommend backing up the postgres database.
If you're on AWS, my other recommendation is to use RDS.
Sean Auriti
@seanauriti
Mar 28 2016 18:48
awesome, thanks!
Arik Fraimovich
@arikfr
Mar 28 2016 18:50
happy to help :) where are you using Redash?
Sean Auriti
@seanauriti
Mar 28 2016 18:50
for a test project
any suggestions for a data pipeline tool to get data from Salesforce into Redshift?
Arik Fraimovich
@arikfr
Mar 28 2016 18:52
If you want a SaaS solution, then check out alooma.com. I didn't do it myself so not sure how involved it is to do it yourself.
Sean Auriti
@seanauriti
Mar 28 2016 18:53
thank you
Arik Fraimovich
@arikfr
Mar 28 2016 18:54
If you end up building something yourself it will be very cool to open source it & share your Redash dashboards (assuming some of them will be generic and not specific to your needs).
Sean Auriti
@seanauriti
Mar 28 2016 18:55
cool, yes of course, it would be cool if there was a query library that could be used across instances
Alex Sukstansky
@alexsukstansky
Mar 28 2016 19:27
Hi everybody. Here is a situation. Using redash with redshift. Redshift shows that query with id N is complete, but the redash doesn't update a view and stays in "waiting" state. Any suggestions where to dig?
Arik Fraimovich
@arikfr
Mar 28 2016 19:35
@alexsukstansky it might be connectivity settings between your EC2 instance and Redshift. I stumbled at this a few times. I would start with verifying your MTU setting is 1500 (see http://docs.aws.amazon.com/redshift/latest/mgmt/connecting-drop-issues.html for details) and if this doesn't help verify the keep alive settings (http://docs.aws.amazon.com/redshift/latest/mgmt/connecting-firewall-guidance.html).
But if both the cluster and Redash instance on the same VPC, it's more likely the MTU than the keep alive.
Alex Sukstansky
@alexsukstansky
Mar 28 2016 19:36
@arikfr Thanks a lot!
ChiragKParmar
@ChiragKParmar
Mar 28 2016 20:50
@mariuszlitwin @arikfr Since I was not using private-key based authentication (running redash on linux box using VMWare fusion), fab script was prompting for password at each stage but because of python with hide('running', 'stdout', 'stderr'): was supressing the output from console!
Arik Fraimovich
@arikfr
Mar 28 2016 20:58
oh, crap. I really need to refactor this script to run from the redash instance. will solve so much issues...
anyway, did you get it working?
ChiragKParmar
@ChiragKParmar
Mar 28 2016 20:59
Yup