Hello,

I had also encountered this issue. What solved it for me was two things. 
One is that I had to add my meta domain to my host file and point it to 
127.0.0.1 . I believe that is what solved the meta service always 
crashing.

The second issue I had was that the oauth2 page would error out. For this 
I had to add a section in my apache config to also proxy requests to 
/query to 127.0.0.1:5100/query.

After that everything seems to work alright as far as I can see.

Marc

So I will reply to my own thread with a fix incase anyone else runs into 
this.

Since my email was not reach the list (it was having issues at the time) 
I jumped on IRC. With Drew's help I was able to figure out that the 
trouble was with the webhook services not all running.

Also somehow had to fix the service files as the services would not start 
up. In the celery commands they had --loglevel=info when it should have 
been --loglevel=INFO. After that change to both the meta and paste 
service files on my system they started up and everything works now.

Hope this can help someone someday.

Hello,

I host my own git,paste instance on Arch Linux. After seeing in this months announcement that gopher and Gemini URL are now accepted on meta profile I wanted to include mine on my profile. While I was able to enter it on the profile settings page. When I view my profile at https://git.marcg.pizza/~marcg It still shows the old https address. What am I missing here? I've tried restarting all services.
I have the latest version of all packages installed.

Thank you

Marc

> maybe that's what I'll try to do.


Sorry for replying to myself but i got it all to apply.


First thing i did was make 8fbeb080c434_add_repository_readme.py revise dacab1dcba45 so i was able to apply 8fbeb080c43 successfully. Then i applied the rest by running the migrate upgrade to the "latest" d42e577c5dcd. Then i put the revise of 8fbeb080c434 back to what it was.


So all the migrations have been applied which is great. However now i am unable to access any repos in the web interface. All returning permission denied. Something else is up. I'll figure that out. Hopefully it is not caused by something done here today.

> apart from temporarily moving the file away.

It did not like that as each one references the one before so it errors out if one is missing. Probably not idea to skip one anyways. It's gotta get applied. If only I could change the order somehow...maybe that's what I'll try to do.

Thank you Denis.


>so I'd suggest the attached patch, if you want to try it out.

While that did permit me to migrate to dacab1dcba45 the following migration(01412986a44d) fails again with the same error.

I suspect again because it grabs all columns

for repo in tqdm(session.query(Repository).all())

Is the only column needed for this migration repository.visibility? Is/can it be that simple?

Hello,


I am facing the same issue upgrading from 0.55.3 to 0.60.4. Error is the same as above just at a different migration step.


INFO [alembic.runtime.migration] Context impl PostgresqlImpl.

INFO [alembic.runtime.migration] Will assume transactional DDL.

INFO [alembic.runtime.migration] Context impl PostgresqlImpl.

INFO [alembic.runtime.migration] Will assume transactional DDL.

I run my own install of SourceHut as well and just want to
second the fact that it should be configurable.

Thank you Thorben. That worked. Finally i can move things over.

So this is an area where you will need to pardon my ignorance. I am not
familiar with how to interact with redis.


Is there cli tools for interacting with it. I assume so.


Thank you