" />

Forums

Resolved
1 votes
Super related to this discussion but I thought it be cleaner to create a new discussion.

Today a Tiki CMS installed on an AWS instance stopped to work.
No database connection.
After checking I couldn’t find the trace of the database used for this server (like it has been deleted). Super odd !

After 20mn of panic I remembered that I had an issue with mysql version on this server (see : Clearos 7.7.2 (final) new install and fully updated = Server version: 5.5.64-MariaDB (bad for most webapp))

I checked mysql version:
[root@ip-172-31-31-5 backups]# mysql -V
mysql Ver 15.1 Distrib 5.5.64-MariaDB, for Linux (x86_64) using readline 5.1



DArrnnn the old version was back...
After following and redoing part of the instruction found in the previous discussion I had about this, things almost got back to normal;
[root@ip-172-31-31-5 backups]# mysql -V
mysql Ver 15.1 Distrib 10.3.13-MariaDB, for Linux (x86_64) using EditLine wrapper


And the site is back on.

But the services doesn’t reflect this anymore (see attachement from this discussion and compare with this one https://www.clearos.com/clearfoundation/social/community?controller=attachment&task=displayFile&tmpl=component&id=10291

I can live with that, but more important, how can I be sure it doesn’t happen again ?
Thanks for any advise about this critical issue.
Tuesday, August 25 2020, 07:26 PM
Share this post:
Responses (1)
  • Accepted Answer

    Tuesday, August 25 2020, 08:30 PM - #Permalink
    Resolved
    0 votes
    There is a way of permanently disabling a systemd unit file. I don't know it so you may need to google it.
    The reply is currently minimized Show
Your Reply