Issues during installation on Debian 9.6 stretch

[Sebastian Roth]

@thegiantcat At the “Backing up database” stage it sends a HTTP web request to the FOG web server asking to deliver a DB backup. This can fail for various reasons. Could be the proxy or other things. You probably want to set no_proxy=127.0.0.1,x.x.x.x (while x.x.x.x is your FOG server IP).

But could be other things as well. First check web server state via systemctl status apache2. Running? Then you might check the logs, /var/log/php-fpm*.log and the error logs mentioned in my signature.

[thegiantcat]

@Sebastian-Roth
Apache2 shows running
The php log file mentioned only contains the following

[02-Dec-2018 06:25:02] NOTICE: error log file re-opened
[04-Dec-2018 09:42:25] NOTICE: Terminating ...
[04-Dec-2018 09:42:25] NOTICE: exiting, bye-bye!
[04-Dec-2018 10:08:03] NOTICE: fpm is running, pid 465
[04-Dec-2018 10:08:04] NOTICE: ready to handle connections
[04-Dec-2018 10:08:04] NOTICE: systemd monitor interval set to 10000ms
[04-Dec-2018 11:59:51] NOTICE: Terminating ...
[04-Dec-2018 11:59:51] NOTICE: exiting, bye-bye!
[04-Dec-2018 12:00:25] NOTICE: fpm is running, pid 470
[04-Dec-2018 12:00:25] NOTICE: ready to handle connections
[04-Dec-2018 12:00:25] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 13:42:01] NOTICE: Terminating ...
[05-Dec-2018 13:42:01] NOTICE: exiting, bye-bye!
[05-Dec-2018 14:09:03] NOTICE: fpm is running, pid 8503
[05-Dec-2018 14:09:03] NOTICE: ready to handle connections
[05-Dec-2018 14:09:03] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 14:11:37] NOTICE: Terminating ...
[05-Dec-2018 14:11:37] NOTICE: exiting, bye-bye!
[05-Dec-2018 14:12:21] NOTICE: fpm is running, pid 9732
[05-Dec-2018 14:12:21] NOTICE: ready to handle connections
[05-Dec-2018 14:12:21] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 14:15:19] NOTICE: Terminating ...
[05-Dec-2018 14:15:19] NOTICE: exiting, bye-bye!
[05-Dec-2018 14:16:11] NOTICE: fpm is running, pid 11029
[05-Dec-2018 14:16:11] NOTICE: ready to handle connections
[05-Dec-2018 14:16:11] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 14:18:55] NOTICE: Terminating ...
[05-Dec-2018 14:18:55] NOTICE: exiting, bye-bye!
[05-Dec-2018 14:19:53] NOTICE: fpm is running, pid 12261
[05-Dec-2018 14:19:53] NOTICE: ready to handle connections
[05-Dec-2018 14:19:53] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 14:28:53] NOTICE: Terminating ...
[05-Dec-2018 14:28:53] NOTICE: exiting, bye-bye!
[05-Dec-2018 14:30:01] NOTICE: fpm is running, pid 13502
[05-Dec-2018 14:30:01] NOTICE: ready to handle connections
[05-Dec-2018 14:30:01] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:01:27] NOTICE: Terminating ...
[05-Dec-2018 15:01:27] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:03:14] NOTICE: fpm is running, pid 14933
[05-Dec-2018 15:03:14] NOTICE: ready to handle connections
[05-Dec-2018 15:03:14] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:09:21] NOTICE: Terminating ...
[05-Dec-2018 15:09:21] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:10:18] NOTICE: fpm is running, pid 16312
[05-Dec-2018 15:10:18] NOTICE: ready to handle connections
[05-Dec-2018 15:10:18] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:29:00] NOTICE: Terminating ...
[05-Dec-2018 15:29:00] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:30:52] NOTICE: fpm is running, pid 18904
[05-Dec-2018 15:30:52] NOTICE: ready to handle connections
[05-Dec-2018 15:30:52] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:39:08] NOTICE: Terminating ...
[05-Dec-2018 15:39:08] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:40:03] NOTICE: fpm is running, pid 20203
[05-Dec-2018 15:40:03] NOTICE: ready to handle connections
[05-Dec-2018 15:40:03] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:54:07] NOTICE: Terminating ...
[05-Dec-2018 15:54:07] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:54:39] NOTICE: fpm is running, pid 470
[05-Dec-2018 15:54:39] NOTICE: ready to handle connections
[05-Dec-2018 15:54:39] NOTICE: systemd monitor interval set to 10000ms
[05-Dec-2018 15:56:55] NOTICE: Terminating ...
[05-Dec-2018 15:56:55] NOTICE: exiting, bye-bye!
[05-Dec-2018 15:57:56] NOTICE: fpm is running, pid 2079
[05-Dec-2018 15:57:56] NOTICE: ready to handle connections
[05-Dec-2018 15:57:56] NOTICE: systemd monitor interval set to 10000ms
[06-Dec-2018 11:15:22] NOTICE: Terminating ...
[06-Dec-2018 11:15:22] NOTICE: exiting, bye-bye!
[06-Dec-2018 11:16:06] NOTICE: fpm is running, pid 6848
[06-Dec-2018 11:16:06] NOTICE: ready to handle connections
[06-Dec-2018 11:16:06] NOTICE: systemd monitor interval set to 10000ms

The apache2 error log also doesn’t seem to be throwing to many errors

[Thu Dec 06 06:25:02.506492 2018] [mpm_prefork:notice] [pid 2080] AH00163: Apache/2.4.25 (Debian) OpenSSL/1.0.2l configured -- resuming normal operations
[Thu Dec 06 06:25:02.506529 2018] [core:notice] [pid 2080] AH00094: Command line: '/usr/sbin/apache2'
[Thu Dec 06 11:15:22.443225 2018] [mpm_prefork:notice] [pid 2080] AH00169: caught SIGTERM, shutting down
[Thu Dec 06 11:16:06.707548 2018] [mpm_prefork:notice] [pid 6849] AH00163: Apache/2.4.25 (Debian) OpenSSL/1.0.2l configured -- resuming normal operations
[Thu Dec 06 11:16:06.707626 2018] [core:notice] [pid 6849] AH00094: Command line: '/usr/sbin/apache2'

However, after doing some digging I thing i found a clue, if I disable the setting in /etc/wgetrc “no_proxy=“localhost”” it fails on “Downloading Binaries” if I have this enabled it fails on “Backing up Database” this backup isn’t using FTP is it?

Unfortunately due to the location, this machine has to go through a proxy to access the internet, is it possible to skip this step in the installation?

[Sebastian Roth]

@thegiantcat said in Issues during installation on Debian 9.6 stretch:

Unfortunately due to the location, this machine has to go through a proxy to access the internet, is it possible to skip this step in the installation?

Sure you can manipulate the scripts but you could also just run the installer like this: ./installfog.sh --exitFail (actually means “don’t exit on failure”)

Strange that “no_proxy” wouldn’t properly work. Have you tested to set to no_proxy="localhost,127.0.0.1,x.x.x.x" where x.x.x.x is the FOG server IP?

[Tom Elliott]

@thegiantcat can you get an exact screen shot of your /etc/wgetrc file?

If you literally have:

"no_proxy="localhost""

This may be a part of the problem.

I’d imagine you should have:

no_proxy=127.0.0.1,localhost,localhost.localdomain,X.X.X.X

Change X.X.X.X to whatever your fog IP actually is.

Also, I believe the installer uses curl for binaries and database backups, so I imagine editing the /etc/wgetrc is the wrong approach.

I believe you would edit the file for whom you’re installing at:
~/.curlrc

You could set the same information up in /etc/skel/curlrc too I suppose.

however you’d want lines like:

proxy = <proxyInfoHere>
noproxy = localhost,127.0.0.1,localhost.localdomain,127.0.1.1,X.X.X.X

You could also edit your ~/.bashrc To contain lines like:

 export http_proxy=<http://proxyInfoHere>
 export https_proxy=$http_proxy
 export ftp_proxy=$http_proxy
 export rsync_proxy=$http_proxy
 export no_proxy="localhost,127.0.0.1,localhost.localdomain,127.0.1.1,X.X.X.X"

If you do add to the .bashrc file, however, you’ll need to close the terminal/logout of the console, and log back in for the changes to take effect.

Hopefully this helps.

In the case you use the bashrc method, please change the X.X.X.X to that of your fog server’s actual IP address, and of course change your http_proxy line to be appropriate. It should include the schema (http{,s}://…) information as well.

[thegiantcat]

@Tom-Elliott
My wgetrc file contains

http_proxy = http://172.22.0.102:3128
use_proxy = on
no_proxy="localhost",127.0.0.0/8,172.22.0.102

I’m not showing a hidden file for curl under ~/ or a config file for curl under /etc/curlrc

I edited my .bashrc to contain the following

export http_proxy=<http://172.22.0.102>
export rsync_proxy=$http_proxy
export https_proxy=$http_proxy
export ftp_proxy=$http_proxy
export no_proxy="localhost,127.0.0.1,localhost.localadmin,127.0.0.1,172.22.0.102"

Is the syntax for that correct, I re-booted the machine after making the change. However its still looking like no dice. (at this point it actually failed at getting binaries

I made the following changes and it was able to get the binaries but it still failed at backing up the database

export http_proxy="http://172.22.0.102:3128"
export rsync_proxy=$http_proxy
export https_proxy=$http_proxy
export ftp_proxy=$http_proxy
export no_proxy="localhost,127.0.0.1,localhost.localadmin,127.0.0.1,172.22.0.102"

[Tom Elliott]

@thegiantcat What’s the installer’s error log look like?

It should show us more information as to why backing up is failing.

I am a little confused too. Is your FOG Server also your PROXY machine?

[thegiantcat]

@Tom-Elliott
No,

The fog server is seperate from the proxy.

[Tom Elliott]

@thegiantcat Then why do you have:

export no_proxy="localhost,127.0.0.1,localhost.localadmin,127.0.0.1,172.22.0.102"

The 172.22.0.102 (which shows as your proxy), should be your FOG Server’s IP address.

[thegiantcat]

@Tom-Elliott
Also here is the error part from the log about when it gets to “Backing up databases”

Dec 06 12:23:00 zstsvreefog systemd[1]: Starting The PHP 7.0 FastCGI Process Manager...
Dec 06 12:23:00 zstsvreefog systemd[1]: Started The PHP 7.0 FastCGI Process Manager.
--2018-12-06 12:23:02--  http://172.24.28.71//fog//maintenance/backup_db.php
Connecting to 172.22.0.102:3128... connected.
Proxy request sent, awaiting response... 302 Found
Could not parse String-Transport-Security header
Location: ?node=schema [following]
--2018-12-06 12:23:03--  http://172.24.28.71//fog//maintenance/backup_db.php?node=schema
Connecting to 172.22.0.102:3128... connected.
Proxy request sent, awaiting response... 500 Internal Server Error
2018-12-06 12:23:04 ERROR 500: Internal Server Error.

This is before making the requested change after the requested change i get

Dec 06 13:31:46 zstsvreefog systemd[1]: Starting The PHP 7.0 FastCGI Process Manager...
Dec 06 13:31:46 zstsvreefog systemd[1]: Started The PHP 7.0 FastCGI Process Manager.
--2018-12-06 13:31:48--  http://172.24.28.71//fog//maintenance/backup_db.php
Connecting to 172.24.28.71:80... connected.
HTTP request sent, awaiting response... 302 Found
Could not parse String-Transport-Security header
Location: ?node=schema [following]
--2018-12-06 13:31:48--  http://172.24.28.71//fog//maintenance/backup_db.php?node=schema
Connecting to 172.24.28.71:80... connected.
HTTP request sent, awaiting response... 500 Internal Server Error
2018-12-06 13:31:48 ERROR 500: Internal Server Error.

The cert on the proxy server is self signed, i assume that its failing because of that, is there a way to start the installation and ignore this? Or a way to add it as a trusted site?

[Sebastian Roth]

@thegiantcat said in Issues during installation on Debian 9.6 stretch:

500 Internal Server Error

Please check your /var/log/php-fpm/www-error.log!

[thegiantcat]

@Sebastian-Roth
I do not have the requested directory(php-fpm) under/var/log,
/var/log also doesn’t have an www-error.log
I also checked just to be through and there isn’t a log under
/etc/php/7.0/fpm

[Sebastian Roth]

@thegiantcat said in Issues during installation on Debian 9.6 stretch:

is there a way to start the installation and ignore this?

As mentioned earlier, you can simply run the installer like this: ./installfog.sh --exitFail (actually means “don’t exit on failure”)

[thegiantcat]

@Sebastian-Roth
Thank you so much, I missed it when you originally said that, running it with --exitFail didn’t work, however running it with -X did work. After this I was able to update/install the schema and actually log into the server via the web interface.

Additionally I tested booting to the server via PXE and was able too!