upgrading from 1.5.0 to 1.5.1 fails at backup

[Hanz]

@tom-elliott Using Fedora 26 still failing with working branch and master…was working with 1.5.0 though. I’ll also add that I can create a backup via GUI just fine and this only occurs during installation.

**Extra information. I tried running the installer with the -X option to continue on fail. After finishing with backup db…failed being the only error, I try going to GUI and get

Service Unavailable
The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.

Error differs slightly

* Setting up Apache virtual host (no SSL).....................OK
 * Starting and checking status of web services................OK
 * Changing permissions on apache log files....................OK
 * Backing up database.........................................Failed!
[root@fog-server bin]#

Running MariaDB 10.1.32

[Wayne Workman]

@Tom-Elliott concerning Hanz issue, I am unsure if it’s related or not but all of my Fedora tests failed last night: http://theworkmans.us:20080/fog_distro_check/installer_dashboard.html

[Wayne Workman]

@hanz Check the status of the mysql / mariadb service. See what it says. Try to access it with the -h option. Specify both localhost and 127.0.0.1 Try both:
mysql -h 127.0.0.1 -D fog
mysql -h localhost -D fog

[Hanz]

@wayne-workman Both are working from server with additional -u root -p options (since I have password)

[Wayne Workman]

@hanz Can you figure out your PHP version please?

[Hanz]

@wayne-workman

PHP 7.1.16 (cli) (built: Mar 28 2018 07:11:55) ( NTS )
Copyright (c) 1997-2018 The PHP Group
Zend Engine v3.1.0, Copyright (c) 1998-2018 Zend Technologies
    with Zend OPcache v7.1.16, Copyright (c) 1999-2018, by Zend Technologies

[Wayne Workman]

@hanz Any errors in the apache error log when you try to run the upgrade?

[Hanz]

@wayne-workman

none other than my 3000 hosts complaining about not being able to connect to backend at 127.0.0.1

[Wayne Workman]

@Tom-Elliott that’s the same error I have in my test VMs:

[Tue Apr 10 02:04:17.644043 2018] [proxy_fcgi:error] [pid 17902] [client 10.0.0.22:51312] AH01079: failed to make connection to backend: 127.0.0.1

Considering that Fedora VMs on working, dev-branch, and master all failed the exact same way on the same day and that @Hanz is having the same issue, this must be a Fedora problem.

[Wayne Workman]

@hanz You could fix this the hard way by moving to something like Debian 9, or wait for us to try to help more. I’m at work right now and can’t focus on this. FWIW, Debian historically has had the least amount of problems with FOG.

[Sebastian Roth]

@Wayne-Workman Looking through the logs I found this:

Apr 10 02:17:55 Fedora27Workstation systemd[1]: Started MariaDB 10.2 database server.
ERROR 1064 (42000) at line 1: You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near 'BY ''' at line 1
ERROR 1064 (42000) at line 1: You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near 'BY ''' at line 1

[Sebastian Roth]

I was on the wrong track I suppose. Wayne is right, the PHP fpm seems to be the issue: FCGI: attempt to connect to 127.0.0.1:9000

@Wayne-Workman Can you check /var/ log/php-fpm/error.log

[Hanz]

@sebastian-roth Indeed this is from installer error log file

Apr 10 11:25:31 fog-server systemd[1]: Starting The PHP FastCGI Process Manager$
Apr 10 11:25:31 fog-server systemd[1]: Started The PHP FastCGI Process Manager.
--2018-04-10 11:25:33--  http://x.x.x.x//fog//maintenance/backup_db.php
Connecting to x.x.x.x:80... connected.
HTTP request sent, awaiting response... 503 Service Unavailable
2018-04-10 11:25:33 ERROR 503: Service Unavailable.

php-fpm/error.log

[10-Apr-2018 11:43:21] NOTICE: Terminating ...
[10-Apr-2018 11:43:21] NOTICE: exiting, bye-bye!
[10-Apr-2018 11:43:50] NOTICE: fpm is running, pid 20233
[10-Apr-2018 11:43:50] NOTICE: ready to handle connections
[10-Apr-2018 11:43:50] NOTICE: systemd monitor interval set to 10000ms

[Tom Elliott]

What’s in /var/log/php-fpm/www-error.log, or very similar?

[Hanz]

@tom-elliott

/var/log/php-fpm/error.log

[10-Apr-2018 11:43:21] NOTICE: Terminating ...
[10-Apr-2018 11:43:21] NOTICE: exiting, bye-bye!
[10-Apr-2018 11:43:50] NOTICE: fpm is running, pid 20233
[10-Apr-2018 11:43:50] NOTICE: ready to handle connections
[10-Apr-2018 11:43:50] NOTICE: systemd monitor interval set to 10000ms

[Tom Elliott]

@hanz no it will start with www

[Hanz]

@tom-elliott none in that directory…

[bcs@fog-server php-fpm]$ ls
error.log           error.log-20180325  error.log-20180408
error.log-20180318  error.log-20180401

[Wayne Workman]

@hanz Run this, it’ll find all files begining with www
find / -type f -name www*

[Tom Elliott]

This is now fixed. Please just run git pull and you will be able to install. I don’t know if this will work for arch based installs quite yet though.

[Wayne Workman]

@tom-elliott I manually started the installation tests just now.