heroku pgbackups:destroy fails w/ "Internal server error" - heroku

Here's what happened...
pgbackups:capture gave the following:
! must delete a backup before creating a new one
...so I tried deleting a past backup:
$ heroku pgbackups
ID Backup Time Size Database
---- ------------------- ----- ---------------
b003 2012/07/16 11:27.19 2.4MB SHARED_DATABASE
b004 2012/08/09 05:27.57 2.5MB SHARED_DATABASE
$ heroku pgbackups:destroy b003
! Internal server error.
! Run `heroku status` to check for known platform issues.
$ heroku status
=== Heroku Status
Development: No known issues at this time.
Production: No known issues at this time.
Any ideas?

Apparently, the error was a "rare occurrence" -- according to Heroku Support.
BTW, the following command succeeded:
$ heroku pgbackups:capture -a <app_name> --expire

Update your heroku gem - that solved the problem for me
gem install heroku

Related

Windows bash 'Cannot create property 'exitCode' on string 'autoupdate:' Installing Heroku CLI/Toolbelt

I am attempting to install the Heroku CLI/Toolbelt on an bash shell version 4.3.48(1) (x86_64-pc-linux-gnu) running on Windows 10. I'm entering wget -qO- https://cli-assets.heroku.com/install-ubuntu.sh | sh in the command line, and after the packages download I'm getting the below:
Preparing to unpack .../heroku_6.14.24-1_amd64.deb ...
Unpacking heroku (6.14.24-1) ...
Setting up heroku (6.14.24-1) ...
+ which heroku
+ LOCATION=/usr/bin/heroku
+ echo heroku cli installed to /usr/bin/heroku
heroku cli installed to /usr/bin/heroku
+ heroku version
▸ Cannot create property 'exitCode' on string 'autoupdate:'
And then it just takes me back to the command line. If I try heroku --v or any other heroku CLI command I get this same error
▸ Cannot create property 'exitCode' on string 'autoupdate:'
I can't find any record of this error code anywhere online. Any thoughts? I really do not know where to start here, usually a good search at least sets me on the path but I have crickets here. Let me know if any further details are required. FWIW the toolbelt was working as of a few days ago, I just had to uninstall and reinstall my bash shell due to an unfortunate incident with Linuxbrew so am rebuilding everything from scratch.
this is fixed but the CLI might have trouble updating. Delete ~/.local/share/heroku and run heroku update and it should be fixed.

Can't install heroku-accounts plugin in heroku cli

I'm trying to install heroku-accounts using this command:
$ heroku plugins:install heroku-accounts
but instead I got this:
heroku-cli: Updating plugins... !
/Users/hit/Library/Caches/heroku/update.lock is locked with a reader active: 17964
I try rm the update.lock but it still get same result
I solve the problem by updating Heroku (I use homebrew install heroku)
heroku update
brew upgrade heroku
None of the above answers seem to work for me. The issue happened when I updated my heroku-cli via brew to version heroku-cli/6.13.8 (darwin-x64) node-v8.2.1 .
Although Error was very explicit that one more process is running, ps | grep showed no such process.
The only solution was to Restart System Once.
There is another copy of the heroku CLI running with the PID 17964
You can find it by doing something like this:
ps | grep 17964
Once you stop that process try your install again.

Heroku command error .local/share/heroku/client/bin/heroku: line 21:

My heroku toolbelt had been working fine until last week that it's giving me this error on all heroku commands.
$ heroku run console -a abc
/Users/songserm/.local/share/heroku/client/bin/heroku: line 21: /Users/songserm/.local/share/heroku/client/bin/node: No such file or directory
/Users/songserm/.local/share/heroku/client/bin/heroku: line 21: /Users/songserm/.local/share/heroku/client/bin/node: No such file or directory
I didn't know that my heroku has been auto upgraded, but what has changed is I installed heroku-vi addon around the time. Not sure if it's the case.
Has anyone seen this issue before?
Noted that I installed heroku on my Mac using homebrew.
Cause:
This also happened to me, the client auto-updated but didn't do so properly:
heroku run console
heroku-cli: Updating to 6.11.14-c5f9179... 12.9 MB/12.9 MB
/Users/steves/.local/share/heroku/client/bin/heroku: line 21: /Users/steves/.local/share/heroku/client/bin/node: No such file or directory
/Users/steves/.local/share/heroku/client/bin/heroku: line 21: /Users/steves/.local/share/heroku/client/bin/node: No such file or directory
Solution:
The resolution was to re-install the client using the following (as found here https://github.com/heroku/cli/issues/485):
rm -rf ~/.local/share/heroku/client
I then checked using the status command and it re-installed correctly:
heroku status
heroku-cli: Updating to 6.11.14-c5f9179... 12.9 MB/12.9 MB
Migrating Heroku CLI v5 plugins... done
Apps: No known issues at this time.
Data: No known issues at this time.
Tools: No known issues at this time.
Heroku-error-Not a directory
Not a directory Error when executing Heroku commands
heroku
▸ stat /User/xxx/.local/share/heroku/client/bin/heroku: not a directory
▸ fork/exec /User/xxx/.local/share/heroku/client/bin/heroku: not a directory
Solution is to remove previous directory
rm -rf ~/.local/share/heroku/client
A quick workaround for me was to remove this file.
rm -rf ~/.local/share/heroku
And I was able to acces heroku console easily.
As everyone said the solution is:
deleting the heroku client with the following console command:
rm -rf ~/.local/share/heroku/plugins
But if you continue having problems you have to delete the heroku plugins folder:
rm -rf ~/.local/share/heroku/plugins
Then you can excecute the command
heroku status
To reinstall correctly
$ heroku status
heroku-cli: Updating to 6.99.0-ec9edad... 12.8 MB/12.8 MB
Apps: No known issues at this time.
Data: No known issues at this time.
Tools: No known issues at this time.

issues with configuration and deployment of my app

I tried restarting my server some days ago with sudo service httpd restart and all sorts of problems have started happening.
I still log into my VPS server with ssh myapp.com.
I enter my password.
But then this comes up, which usen't to happen :
Last login: Sun Jul 21 22:54:02 2013 from 123.145-678-89.dsl.completel.net
-bash: __rvm_add_to_path: command not found
In Terminal I tried some things which solved problems for me before like:
bundle exec rake db:seed
But I keep getting:
-bash: bundle: command not found
Any idea how I can solve this problem? From what I've read it's suggested I look at my bash_profile/bashrc files and edit some things in there. But where are these located? What should I change? Where can I find my deploy log and configuration?
When I do a cap production deploy:migrations to transfer my site online to my remote server it all used to go fine, but now I get the error:
failed: "rvm_path=$HOME/.rvm/ $HOME/.rvm/bin/rvm-shell 'ree-1.8.7-2012.02' -c 'cd /var/www/apps/myapp/releases/20130720123042 && bundle exec rake RAILS_ENV=production RAILS_GROUPS=assets assets:precompile'" on 12.34.567.8
mypc#ubuntu:~/myapp$
Thanks for any help.
Reinstalling RVM with \curl -L https://get.rvm.io | bash -s stable sorted it out...

Killing abandoned process on Heroku

I killed an IRB prompt in a not-so-graceful manner (started with heroku run irb), and now I've got a zombie process which I can't seem to kill:
Process State Command
------------ ------------------ ------------------------------
run.3 up for 2h irb -r ./init.rb
web.1 up for 0s thin -p $PORT -e $RACK_ENV -R $HER..
I've tried:
heroku restart
heroku ps:scale run=0
heroku ps:restart run.3
Anyone know how I can force quit it?
I noticed a new ps:stop command added to the Heroku command line client a few days ago:
https://github.com/heroku/heroku/commit/a6d9eb7f314bf2c5f162a508e8d764286fb577bb
I'm not sure if that change made it into version 2.9.0 but it would be worth a try.
Update
This is now in the Heroku Toolbelt. Just run:
heroku ps:stop <process id from heroku ps>
Example:
heroku ps:stop run.8729
Thanks James!! I needed to update my Heroku client first, and it worked like a charm.
$ sudo gem install heroku
$ heroku ps #to view processes
$ heroku stop <process name here>
edit: they've fixed this issue
Heroku doesn't have a way to kill abandoned run.x processes.
But it will be killed automatically after 24 hours.
If anyone else is struggling with killing using something like:
heroku run ps:stop run.789
Killing by id worked for me:
heroku ps:kill 61ff0687-eaf4-4299-9c65-f0b22af7ec67
I got the id using the platform api list dynos - https://devcenter.heroku.com/articles/platform-api-reference#dyno-list
Worked for a detached one-off that was abandoned.

Resources