Process completed with exit code 1.
kerim opened this issue · comments
Describe the bug
Was working fine until 5 to 6 days ago. (Would only fail about 1 in 10 times.) Since then, it has failed every single time. Error is:
Process completed with exit code 1.
To Reproduce
Steps to reproduce the behavior:
I didn't do anything different, was working fine beforel
Expected behavior
Backup
Traceback
Please use http://gist.github.com/ or similar, and report the last line here.
Don't understand this request.
Run roam-to-git --debug notes/ and report what you get.
It should open a Chrome front-end, an do the scrapping. The repository content will not be modified. If applicable, add screenshots to help explain your problem.
Don't know/understand how to do this.
Please complete the following information:
- OS: [e.g. MacOs, Linux] I'm on a Mac, but seems irrelevant?
- Do you use Github Action? Yes.
- Do you use multiple Roam databases? No.
- Does roam-to-git use to work for you? When precisely did it stopped to work? See above.
- Does some backup runs are still working? No.
Additional context
Add any other context about the problem here.
I am also receiving the same error message "Process completed with exit code 1". I just set up my Roam backup using the instructions at https://kuvee.medium.com/a-step-by-step-guide-to-backing-up-your-roam-research-database-356107fe82ee
OS: Windows 10
Do you use Github Action? Yes.
Do you use multiple Roam databases? No.
Does roam-to-git use to work for you? I just set it up today and it failed on the first run
When precisely did it stopped to work? December 20, 2020
Does some backup runs are still working? No
I tried creating starting fresh with a new script and GitHub repository, but the same problem: "Process completed with exit code 1."
Here is the log file in text format from the most recent failed backup with the error code "Process completed with exit code 1". I have added an empty line between the sections of the backup log for sake of clarity.
https://www.dropbox.com/s/nmjhw93346euqek/Roam%20backup%20using%20GitHub%20error%20log.txt?dl=0
Thanks for your help with this error!
EDIT: Backup worked on its own (about 9 hours ago) without any changes! Will update if it breaks again. Leaving the old comment below for posterity.
Experiencing the same issue!
OS: Windows 10
Do you use Github Action? Yes.
Do you use multiple Roam databases? No.
Does roam-to-git use to work for you? Yes, have been using it since Sep.
When precisely did it stopped to work? December 20, 2020
Does some backup runs are still working? No, fails consistently.
Still not working for me. Fails every time.
Same here, fails every time with the same error code.
Thanks in advance for help solving this issue
@kerim All errors end with Process completed with exit code 1. Please attach your log file. (You can check at the end of my guide on how to download your log files.)
@bobmhughes If you're getting this error every hour, you may have entered the wrong email or are using a Google account.
@everruler12 Thanks for the suggestion to "Double check that your ROAMRESEARCH_DATABASE value is just the name of your graph, and not a URL". I don't know if this was entered wrong or not as when you go to update it, the previous value does not show. I also checked the "All workflows" page and it turns out that the backup is not failing every time. I incorrectly assumed that it was failing every time as I have received so many "backup failed" emails. I didn't realize that I would not receive an email if the backup was successful. When there is an error, I see this in the log "AssertionError: All roads lead to Roam, but that one is too long. Try again when Roam servers are faster. Error: Process completed with exit code 1." I also noticed in the log file that the backup was trying to access "https://roamresearch.com/#/app/***" which doesn't make sense to me.
@bobmhughes Your graph name is a Secret, that's why it uses *** in the log. I've been getting the same kind of error every so often. I don't know what causes it. It may be Roam timing out. As long as most hourly runs are successful, these errors can be ignored.
@everruler12 Thanks. I was able to find the error on my own and it is working again. What happened is that I had imported some notes (via Readwise) that had a double colon "::" (probably an OCR error in the original PDF) at the end of a huge block of text. Roam treated this as an attribute, hence a page name, and the backup script was giving an error for having a page name that was too long. I wonder if the backup script could be made more robust so that errors like this don't interrupt the backup?
@everruler12 It looks like I am getting an 83% success rate with my backups. The most common errors are Assertion and Network errors
Most of my hourly backups are still successful, with about 3 errors per day.
Is the time out happening between Wait for interface to load and Closing browser markdown? Also see #76
Error: Process completed with exit code 1.