fix orion environment not working#4857
Conversation
Fixes the current orion environment not working. I have also removed the squid-archive services in favour of our new endpoint `https://archive.joystream.org/graphql`
c9a643d to
dfd8667
Compare
mnaamani
left a comment
There was a problem hiding this comment.
Works well, but the default archive gateway on startup is the production chain which doesn't make sense. I think we should keep an archive + gateway in the setup. If we choose to run orion from this repo for the production chain then we set the CUSTOM_ARCHIVE_GATEWAY_URL env variable to https://archive.joystream.org/graphql
| TRUSTED_REVERSE_PROXIES=1 | ||
| SESSION_EXPIRY_AFTER_INACTIVITY_MINUTES=60 | ||
| SESSION_MAX_DURATION_HOURS=720 | ||
| EMAIL_CONFIRMATION_ROUTE=http://localhost:4074/api/v1/confirm-email?token={token} |
There was a problem hiding this comment.
obviously this will only work when opening email from the machine. But I gather we would most likely disable email confirmation and notifications in a playground setup so thats fine.
|
@ignazio-bovo The CI check that is failing: because it tries to use the adjacent docker-compose file Please apply the same fixes to that as well. |
mnaamani
left a comment
There was a problem hiding this comment.
A couple minor fixes needed of archive explorer config. Otherwise working nicely.
With so env variables now in the .env file and with some similarly named ones, i think in future we need to do a bit of refactoring to make it more manageable.
.env
Outdated
| TRUST_PROXY=uniquelocal | ||
|
|
||
| # Sendgrid API | ||
| SENDGRID_API_KEY= |
There was a problem hiding this comment.
There is an identical SENDGRID_API_KEY variable below for the faucet.
perhaps we can prefix the one for orion with ORION_SENDGRID_API_KEY then pass that env variable in docker-compose environment section (so you don't have to modify orion code to expect a different variable name)
| @@ -0,0 +1,4 @@ | |||
| listen_addresses = '*' | |||
There was a problem hiding this comment.
can we rename this file to orion-postgres.conf, there is another PR where I was planning to add a postgres.conf file for the query-node db. It makes sense to have two different conf files for each db instance.
docker-compose-no-bind-volumes.yml
Outdated
| - .env | ||
| environment: | ||
| DB_TYPE: postgres # set to `cockroach` for Cockroach DB | ||
| DB_HOST: archive_db |
There was a problem hiding this comment.
I think the hostname is incorrect, (should it be orion_archive_db), also the correct port for the archive db ARCHIVE_DB_PORT is not used here.
I'm seeing connection error log for the explorer container:
2023-09-11 10:54:19 {"level":3,"time":1694415259620,"ns":"sqd:substrate-explorer","msg":"failed to connect to postgres://postgres@archive_db:5432/squid-archive","err":{"errno":-3008,"code":"ENOTFOUND","syscall":"getaddrinfo","hostname":"archive_db","stack":"Error: getaddrinfo ENOTFOUND archive_db\n at GetAddrInfoReqWrap.onlookup [as oncomplete] (node:dns:109:26)"}}
2023-09-11 10:54:19 {"level":2,"time":1694415259730,"ns":"sqd:substrate-explorer","msg":"listening on port 3000"}
mnaamani
left a comment
There was a problem hiding this comment.
You need the DB_PORT also to be set for the explorer, and the renamed postgres.conf file needs to be adjusted in the docker-compose file for the orion-db service to match the new name:
volumes:
- orion-db-data:/var/lib/postgresql/data
- ./orion-postgres.conf:/etc/postgresql/postgresql.conf
|
|
||
| # Sendgrid API | ||
| ORION_SENDGRID_API_KEY= | ||
| ORION_SENDGRID_FROM_EMAIL=gateway@example.com |
There was a problem hiding this comment.
These need to be used in docker-compose to pass the expected env var name to orion like so:
environment:
- SENDGRID_API_KEY=${ORION_SENDGRID_API_KEY}
- SENDGRID_FROM_EMAIL=${ORION_SENDGRID_FROM_EMAIL}
Fixes the current orion environment not working when starting the playground
I have also removed the squid-archive services in favour of our new archive endpoint