#[SOLVED] Old Cloud Functions Not Working After 1.4.2 Migration
97 messages · Page 1 of 1 (latest)
no error or logs in console .. will paste traceback from docker container logs
errors like these i wasnt getting in 1.3.4 ;-;
According to this :#🏠│general message it's not the expected behaviour.
cc @worthy veldt
Hope you don't mind me tagging you LOL
probably you are using an older version of sdk?
and the runtime is python?
whats on line 40 of your main.py?
can you share the impl? and the sdk version used?
btw can it be because i directly upgraded from 1.3.4 to 1.4.2
there are major changes so its quite possible. there's a possibility of older sdk used as well. hence i requested for the details.
sure one sec actually i was restoring my backup rn and seems like it didnt work so i will get back to u after i fix it oof
what values are you passing into list_documents?
If env variable, then migration maynedidnt go well because of problematic value in your encrypt key in .env. That results in false as value of env variables
oh whats the env var name for encrypt key?
openssl key?
_APP_OPENSSL_KEY_V1 yes
But wait
THere's also a _APP_EXECUTOR_SECRET
Yep that's the default value
okay so i think i might have to make a new issue ..
I tried running backup after re installing 1.3.4 image and then ./restore.sh in appwrite dir
i did replace .env fle with the backed up one
but i am getting this for functions
i hope its okay to continue in the functions issue
from docker logs appwrite
so auth and functions are showing internal 500 error
ok i think i fixed it somehow with fresh installation
and again restoring 1.3.4
so i am currently back to 1.3.4
now i am thinking of upgrading to 1.3.8 then 1.4.0 migrate then 1.4.2 migrate?
anything else i should keep in mind?
_APP_OPENSSL_KEY_V1 and _APP_EXECUTOR_SECRET env variables i should check after 1.4.2 migration?
This is the way I think.
Did you mention before that you moved to 1.4 from 1.3.4?
i directly moved to 1.4.2
another issue i get while going to 1.3.8
weird thing is it doesnt show it if i go for 1.4.2
maybe smth wrong with my installation cmd ;-;
Okay are you on 1.3.4 right now?
can you confirm by running docker ps | grep appwrite
yes
yep confirmed
sorry for so much confusion
that’s okay
okay are you in production right now?
like are there active users of your app?
yea 3 production projects
yep
thats why i tried reverting back to 1.3.4 just after the function issue in 1.4.2
okay here's what I recommend you to do. Perform a backup of your current instance. Make sure to stop traefik while data is being backed up.
Take this backup, and restore it on another appwrite instance v1.3.4
Then try to upgrade to 1.3.8, then migrate
Then try 1.4.2 and migrate
See if you face any issues.
thanks!
tho the main problem is this
will try searching if anyone else faced this when going to 1.3.8
Fatal error: Uncaught Utopia\Database\Exception\Query: Invalid query: Invalid cursor: UID must contain at most 36 chars
going from 1.3.8 to 1.4.1 migration
left 1.4.0 because of this error .. shouldnt be an issue cuz some other ppl in this server did the same 🤔
1.3.8 > 1.4.2
also, after updating to 1.4.2, do this...https://github.com/appwrite/appwrite/pull/6199/files
oh
Fatal error: Uncaught Utopia\Database\Exception\Query: Invalid query: Invalid cursor: UID must contain at most 36 chars. Valid chars are a-z, A-Z, 0-9, and underscore. Can't start with a leading underscore in /usr/src/code/vendor/utopia-php/database/src/Database/Database.php:4061 Stack trace: #0 /usr/src/code/src/Appwrite/Migration/Migration.php(197): Utopia\Database\Database->find(Object(Utopia\Database\Document), Array) #1 /usr/src/code/src/Appwrite/Migration/Migration.php(154): Appwrite\Migration\Migration->documentsIterator('cache') #2 {main}
btw is this resolved in 1.4.2
but which environment
just below appwrite: one?
look into the details of the PR
i mean if i compare this with current docker-compose.yml then i am still not sure
am i missing the point
oh i am dumb
i can just open the full file whats wrong with me đź’€
appwrite-schedule got it
and regarding compose.phtml
is it possible to update that?
it's helpful to learn how to use these tools
this file is used to when generating your docker-compose.yml file during install/upgrade so you don't need to update that exact file.
just your docker-compose.yml file
i see gotcha so i updated the yml file
and did docker compose down , docker compose up -d
i still have this listDocuments issue lemme get it brb
I checked the guest perms and read perm is still there 🤔
btw i have logged in from a verified account
fixed
it was because of a relation collection (gave the perrms)
probably because i directly moved from 1.3.4 > 1.3.8 > 1.4.2