Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Octopus install fails #75

Closed
arielqgold opened this issue Sep 22, 2010 · 12 comments
Closed

Octopus install fails #75

arielqgold opened this issue Sep 22, 2010 · 12 comments

Comments

@arielqgold
Copy link

On vanilla ubuntu lucid slicehost install Barracuda installs fine, Octopus fails with these errors:
Octopus user: admin.

==> Wed Sep 22 17:26:31 UTC 2010 INIT A: _STATUS is INIT
==> Wed Sep 22 17:26:31 UTC 2010 INIT A: Hot Sauce check..
==> Wed Sep 22 17:26:35 UTC 2010 INIT A: Shared platforms code v.001 (latest available) will be used for this install
==> Wed Sep 22 17:26:35 UTC 2010 INIT A: Create directories and/or set good permissions, please wait...
==> Wed Sep 22 17:26:39 UTC 2010 INIT A: Add user, please wait...
==> Wed Sep 22 17:26:43 UTC 2010 INIT A: Prepare scripts, please wait...
==> Wed Sep 22 17:26:47 UTC 2010 INIT A: pre-install start
ERROR 1064 (42000) at line 2: You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '1+R=' WITH GRANT OPTION;
GRANT ALL PRIVILEGES ON . TO 'admin'@'xxx-xxx-xxx-xx.' at line 1
==> Wed Sep 22 17:26:56 UTC 2010 INIT A: pre-install completed
==> Wed Sep 22 17:27:00 UTC 2010 INIT A: Switch user and run AegirSetupB, please wait...
==> Wed Sep 22 17:27:04 UTC 2010 INIT B: Aegir automated install script part B
==> Wed Sep 22 17:27:05 UTC 2010 INIT B: Checking status..
==> Wed Sep 22 17:27:09 UTC 2010 INIT B: Create dirs and/or set good permissions
chmod: cannot access /data/disk/admin/distro': No such file or directory ==> Wed Sep 22 17:27:13 UTC 2010 INIT A: Aegir Hostmaster instance install completed ==> Wed Sep 22 17:27:13 UTC 2010 INIT A: post-install start, please wait... /opt/tmp/nginx-for-drupal/aegir/barracuda/AegirSetupA.sh.txt: line 311: /data/disk/admin/config/admin.nginx.conf: No such file or directory /opt/tmp/nginx-for-drupal/aegir/barracuda/AegirSetupA.sh.txt: line 315: cd: /data/disk/admin/config/includes: No such file or directory chgrp: cannot access/data/disk/admin/aegir/distro/001/sites/host.example.com/files': No such file or directory
chgrp: cannot access `/data/disk/admin/aegir/distro/001/sites/host.example.com/settings.php': No such file or directory
/opt/tmp/nginx-for-drupal/aegir/barracuda/AegirSetupA.sh.txt: line 337: cd: /data/disk/admin/aegir/distro/001: No such file or directory

@omega8cc
Copy link
Owner

Octopus was broken when you tried it because we had to test new fixes for redirects and ssl. Now it should work as expected. We will probably use separate repo for testing to avoid such problems, sorry!

@zzack
Copy link

zzack commented Sep 24, 2010

I just tried installing octopus and I got the same (more or less) errors, and ultimately a failed installation (the site says "under construction"). Should this be working atm?

@linuxgeneral
Copy link

As of today I tested 7 installs with octopus on Debian Lenny 64bit linodes without error. However, I am having problems on hosts listening on a second IP.

@zzack
Copy link

zzack commented Sep 24, 2010

Just tried again, and it worked. Maybe I had a typo in my mysql password? iono. I am using ubuntu 10.04 for the record

@omega8cc
Copy link
Owner

It is possible some errors can be caused by upgrading only Octopus without prior upgrade of core hostmaster using Barracuda, which holds main Nginx configuration files (Octopus simply includes its configuration in the Barracuda main config). We need to introduce a simple check and deny Octopus upgrade if Barracuda was not upgraded first.

@arielqgold
Copy link
Author

I just tried again, on clean install, and got the same errors. The only thing to add is that the Octopus installer wouldn't go forward saying that the domain didn't point to the IP, but it did. The domain previously pointed to a different ip. So I added a new DNS record and changed domain in installer config file and the installer executed, but with same/similar errors and finally a "Under Construction" page.

@omega8cc
Copy link
Owner

You probably need to wait a bit (a few hours or even one day) for DNS propagation and then use the domain for Octopus.

@arielqgold
Copy link
Author

Got it to work!

So I don't think the issue was DNS.

I changed user from "admin" to "octopus" in Octopus install script, reran the Octopus installer and it worked. To cleanup from the failed Octopus install should I just remove the user admin and remove /data/disk/admin ?

@omega8cc
Copy link
Owner

Great! Yes, to clean up the failed install remove the user and its home directory. Check also if there is a broken symlink to be removed in /var/aegir/config/server_master/nginx/platform.d. Also remove its entry from /etc/sudoers. It is another report about failed install with "admin" user. Curious what is causing this on the system without "admin" user yet.

@linuxgeneral
Copy link

Considering the admin usr is reserved for each instance you may want to consider blocking this username for octopus installs.

@linuxgeneral
Copy link

How about and un-install option or script to clean-up orphaned or unwanted installs?

@omega8cc
Copy link
Owner

Yes, it's a good idea to not allow "admin" as an Aegir username on any Hostmaster.

As for cleaning the system, I think we shouldn't automate it to avoid real disaster. We never know if something is really orphaned and should be removed/archived when it is some unknown % of partial install.

Instead, I think we need two useful things here: good (easy to follow) documentation on the files structure used by Barracuda and Octopus, and second - a script to detect and list all things looking like a zombies - dirs, users, databases, entries in sudo etc. The user could then review the report and use suggested commands/actions to clean the system manually, with recommended mv to /var/backups before permanent deletion. We just have to assume that bad things can/will happen and we should never delete anything when we are not 300% sure it should/can be safely deleted.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants