6.0.0p1: Production system is down. Only shows white screen - error logs show key creation not working?

Situation
After docker could not restart due to failed attempt to secure apache server (with 3rd party ssl certificates and other security recommendations - thinking they were working, but then failed when I rebooted the web server): Restored production system database with mysqlbackup dump onto an OpenEMR system created from the latest 6.0.0 cloud formation stack- now only shows white screen (no login screen). It is possible to access the database instances and EC2 instances. However, there is no user interface for the users to log into. Logs show key creation errors. This makes me wonder if there are certain tables (the ones for the keys?) that should be excluded in a backup and restore.

My recovery strategy going forward is twofold:

  1. Restore from 5.0.2(5) backups to a system and upgrade it to 6.0.0p1 - There would not be webserver access to medical records for a couple of weeks, but I know this backup restore strategy works because I have tested it numerous times. I should have tested the 6.0.0 backup and restore - my bad.
  2. Attempt a mysqldump excluding the keys table. Maybe a backup from the 6.0.0 system with the broken web server but intact RDS database node without the keys and restore will prevent the problem with key creation errors and decryption errors.
  3. Access the still working RDS instance MySQL database with database scripts for current appointments, most recent progress notes, vitals, etc.

OpenEMR Version
I’m using OpenEMR version : 6.0.0(1) on AWS

Browser:
I’m using: chrome or safari - latest versions for mac

Operating System
I’m using: AWS

Logs
[pid 16] AH01909: server certificate does NOT include an ID which matches the server name
pid file /run/apache2/httpd.pid overwritten – Unclean shutdown of previous Apache run?
[Mon Jun 14 10:21:05.483404 2021] [php7:notice] [pid 19] [client ] OpenEMR Error : Key creation is not working - Exiting.
[php7:notice] [pid 16] [client] OpenEMR Error : Decryption failed authentication.

You prob need to allow oemr to regen the file keys by going to and deleting
sites/default/documents/logs_and_misc/methods/sixa and sixb. then login

1 Like

The sites/default/documents/logs_and_misc/methods/ directory just has a README.md file in it…

/var/www/localhost/htdocs/openemr/sites/default/documents/logs_and_misc/methods # ls
README.md

On that same restore attempt system (I have several attempts to restore the system going in parallel):
[Mon Jun 14 12:44:45.859139 2021] [php7:error] [pid 38] [client ] PHP Fatal error: require_once(): Failed opening required ‘sites/default/sqlconf.php’ (include_path=’.:/usr/share/php7’) in /var/www/localhost/htdocs/openemr/index.php on line 22 … I’m going to check permissions on that file.

I would look at folder permissions. The system attempts to create a sixa and sixb key file in that folder location. If the files are not there, then php does not have access to write to that location.

1 Like