Wheezy upgrade problem

[vascoleiria]

Hi

I just upgraded from wheezy to jessie and my zabbix installation just won't start
Since I have it is working on a VM, I restored it. Is there a walkthrough I can use to do this without have the zabbix crash?
Should I instead upgrade to buster?
I have Zabbix 3.4
Thx for the help

[Markku]

Hi,

First, what is the error you get in Zabbix?

Second, I wouldn’t upgrade to jessie at this point since that is very old too. That said, I haven’t upgraded my servers to buster yet. Upgrades from jessie to stretch have worked fine for me. In any case you need to check the install/upgrade notes when going up from wheezy.

Which database are you using? I’ve used MySQL but moved to MariaDB when Debian changed it default in stretch. No specific problems then either.

Markku

[Markku]

Also note that there are no Zabbix packages for 3.4 for buster in repo.zabbix.com (as 3.4 is out of support now), so that may affect your upgrade planning as well.

Markku

[vascoleiria]

Hi Markku thx for quick reply. The error was site not found. So something went very wrong starting the services. So what is your advise in the course of action? First upgrade Zabbix then debian?

[Markku]

Is that error message in your browser or where? If yes, you would need to check your Apache (assumed) setup. Are there any problems in zabbix_server.log?

Regarding the upgrade order, there are no Zabbix 4.0 packages for wheezy. What I would do is first reading the Debian upgrade documentation and upgrade to stretch, get everything working there, and then upgrade Zabbix to 4.0 (as it is LTS, long-term support). Later I would then upgrade Debian to buster.

These all assume use of packages from Zabbix repo and quite clean Debian installations, so your situation can be different of course.

Markku

[vascoleiria]

Yes it is in my browser. Ok I'll start upgrading to stretch. I'll keep you posted how it goes.

[vascoleiria]

In case someone else finds this post. here's what I add to do to far:
I used this tutorial to upgrade to stretch http://yab.x10.mx/archives/47 but add to do some modifications along way to work.

1. change sources.list to look like this:

deb http://archive.debian.org/debian wheezy main contrib non-free
deb-src http://archive.debian.org/debian wheezy main contrib non-free
#deb http://archive.debian.org/debian wheezy-updates main contrib non-free
deb http://archive.debian.org/debian-security wheezy/updates main contrib non-free

2. to be able to do the updates "sudo apt-get -o Acquire::Check-Valid-Until=false update"

3. force install systemd-sysv using this:

apt-get install -y systemd-sysv --force-yes

4. because of the public key error, had to do this
sudo aptitude install debian-keyring debian-archive-keyring

After I finish upgrading to Jessie I get the message:
The requested URL /zabbix/ was not found on this server. Help?

[Markku]

The requested URL /zabbix/ was not found on this server. Help?

You need to fix your Apache configuration somehow. For example, in one system I have (set up by Zabbix, not by myself):

$ ls -l /etc/apache2/conf-available/
...
lrwxrwxrwx 1 root root 23 Dec 13 2016 zabbix.conf -> /etc/zabbix/apache.conf

$ ls -l /etc/apache2/conf-enabled/
...
lrwxrwxrwx 1 root root 29 Dec 13 2016 zabbix.conf -> ../conf-available/zabbix.conf

That causes Apache to load the /etc/zabbix/apache.conf configuration. Remember to restart Apache2 if you make any changes.

Markku

[vascoleiria]

I don't have either. I guess there is a dynamic link missing .Here's my list

root@zabbix:~# ls -l /etc/apache2/conf-available/
total 28
-rw-r--r-- 1 root root 221 Apr 3 11:43 apache2-doc.conf
-rw-r--r-- 1 root root 315 Apr 3 11:43 charset.conf
-rw-r--r-- 1 root root 127 Jul 29 2013 javascript-common.conf
-rw-r--r-- 1 root root 3224 Apr 3 11:43 localized-error-pages.conf
-rw-r--r-- 1 root root 189 Apr 3 11:43 other-vhosts-access-log.conf
lrwxrwxrwx 1 root root 28 Aug 1 08:22 phpmyadmin.conf -> ../../phpmyadmin/apache.conf
-rw-r--r-- 1 root root 2190 Apr 3 11:43 security.conf
-rw-r--r-- 1 root root 455 Apr 3 11:43 serve-cgi-bin.conf
root@zabbix:~# ls -l /etc/apache2/conf-enabled/
total 0
lrwxrwxrwx 1 root root 30 Jul 30 16:25 charset.conf -> ../conf-available/charset.conf
lrwxrwxrwx 1 root root 40 Jul 30 16:34 javascript-common.conf -> ../conf-available/javascript-common.conf
lrwxrwxrwx 1 root root 44 Jul 30 16:25 localized-error-pages.conf -> ../conf-available/localized-error-pages.conf
lrwxrwxrwx 1 root root 46 Jul 30 16:25 other-vhosts-access-log.conf -> ../conf-available/other-vhosts-access-log.conf
lrwxrwxrwx 1 root root 33 Aug 1 08:22 phpmyadmin.conf -> ../conf-available/phpmyadmin.conf
lrwxrwxrwx 1 root root 31 Jul 30 16:25 security.conf -> ../conf-available/security.conf
lrwxrwxrwx 1 root root 36 Jul 30 16:25 serve-cgi-bin.conf -> ../conf-available/serve-cgi-bin.conf

[vascoleiria]

I changed /etc/apache2/apache2.conf for include conf.d or config.d

Include conf.d/
Include sites-enabled/

And now I have access to the front end.

Now the "Zabbix Server not running" error appears below

There is not even a zabbix_server.log being created (I renamed the current to .old to check )

Regarding the service itself here is the ax |grep zabbix result

485 ? S 0:00 /usr/sbin/zabbix_agentd
486 ? S 0:00 /usr/sbin/zabbix_agentd: collector [idle 1 sec]
487 ? S 0:00 /usr/sbin/zabbix_agentd: listener #1 [waiting for connection]
488 ? S 0:00 /usr/sbin/zabbix_agentd: listener #2 [waiting for connection]
489 ? S 0:00 /usr/sbin/zabbix_agentd: listener #3 [waiting for connection]
490 ? S 0:00 /usr/sbin/zabbix_agentd: active checks #1 [idle 1 sec]
1341 pts/0 S+ 0:00 grep zabbix


Checked the services list and found this

zabbix-server.service loaded active exited LSB: Start zabbix-server daemon

[Markku]

systemctl stop zabbix-server
systemctl start zabbix-server

What do you get in zabbix_server.log or other logs?

Markku

[vascoleiria]

The server log is never written, only the agent log. Here is the output

root@zabbix:/var/log/zabbix# cat zabbix_agentd.log
485:20190802:084800.013 Got signal [signal:15(SIGTERM),sender_pid:7851,sender_uid:0,re ason:0]. Exiting ...
485:20190802:084800.018 Zabbix Agent stopped. Zabbix 3.4.15 (revision 86739).
480:20190802:084822.036 Starting Zabbix Agent [Zabbix server]. Zabbix 3.4.15 (revision 86739).
480:20190802:084822.038 **** Enabled features ****
480:20190802:084822.038 IPv6 support: YES
480:20190802:084822.038 TLS support: YES
480:20190802:084822.038 **************************
480:20190802:084822.038 using configuration file: /etc/zabbix/zabbix_agentd.conf
480:20190802:084822.039 agent #0 started [main process]
481:20190802:084822.040 agent #1 started [collector]
483:20190802:084822.042 agent #3 started[listener #2]
482:20190802:084822.044 agent #2 started[listener #1]
484:20190802:084822.048 agent #4 started[listener #3]
485:20190802:084822.052 agent #5 started [active checks #1]
485:20190802:084822.079 active check configuration update from [127.0.0.1:10051] started to fail (cannot connect to [[127.0.0.1]:10051]: [111] Connection refused)

[Markku]

Are you saying that when you stop/start "zabbix-server", the "zabbix-agent" service is stopped/started?

Markku

[vascoleiria]

Yes, only the agent seems to start

[Markku]

What is the error message you get when you try to start the Zabbix server with ”systemctl start zabbix-server”?

How about the other logs, journalctl, /var/log/messages, /var/log/daemon.log, /var/log/syslog, etc., what’s there when you try starting the server?

Markku