<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/4.6.6">
</HEAD>
<BODY>
<BR>
Hi Geoff!<BR>
<BR>
As I stated in an earlier message, moved your reply from my thread "FrontEnd will connect if feels like it" to this one as with help from this forum got the third Frontend to consistently connect, tried to fix a channel problem (KQIN RF 34.99 being used/selected instead of RF 34.1 [Virtual 36.1]). Your suggestions were to go back to the Backend, and since I managed to kill that off figured may as well move it over to this thread.<BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
<FONT COLOR="#000000">>> Does systemctl status mythbackend.service report that the backend is</FONT><BR>
<FONT COLOR="#000000">>> running?</FONT><BR>
<FONT COLOR="#000000">> Hmm: no, and before I go on some times this Frontend works/connects and</FONT><BR>
<FONT COLOR="#000000">> some times it doesn't. The other Frontends don't seem to have a problem.</FONT><BR>
<FONT COLOR="#000000">> The response to the command was:</FONT><BR>
<FONT COLOR="#000000">> Loaded : not-found (Reason: No such file or directory)</FONT><BR>
<FONT COLOR="#000000">> Active: inactive (dead)</FONT><BR>
OK, all the way back to basics.<BR>
1) You need network services running on the master box. Looks like you <BR>
have that: I am presuming you are posting from that box, otherwise all<BR>
bets are off!<BR>
</BLOCKQUOTE>
Naaa! I'm emailing from a system running Ubuntu 14.04! I can face forward and type instead of emailing with my knees jammed against the desk drawers! <g> Pseudo nerd-ville: real system to the front, monitor for the Backend to my right, monitor for the 3rd Frontend that was ready to go to that right. Looooong HDMI cables and wireless keyboard!<BR>
<BR>
Back to little more serious answer, the failed Backend (MythBE1) does connect to other computers (just pinged the printer server), as well as the outside world (Firefox connects to my usual home page).<BR>
<BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
2) You need mysql running on a computer with the mythconverg database. <BR>
It may be mysqld, mysqld.service or mariadb.service depending on <BR>
Ubuntu/Fedora/obselete-any-of-the-above.<BR>
</BLOCKQUOTE>
<BR>
The Backend and Frontends are running 16.04.<BR>
<BR>
The Backend is running/has installed mysql.service. It doesn't find anything with mysqld.servcie. (Used the command "locate -w 'mysqld.service'".)<BR>
<BR>
(BTW, I may give too much information some times to be sure I'm doing things correctly.)<BR>
<BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
Check on the "master" computer.<BR>
You must be able to do a 'mysql -umythtv -pmythtv mythconverg' on that<BR>
machine. (The "usual" setup parameters).<BR>
</BLOCKQUOTE>
<BR>
Bzzt! <g> Also tried with 'sudo' just in case. Both gave:<BR>
ERROR 2002 (HY200): Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)<BR>
<BR>
Know from previous weeks' fiddlings before I started posting that's a problem with MySQL not being installed correctly, possibly something with the repositories. Found an uninstall/re-install sequence which corrected. At this point I haven't done that as it seems like it's not the end-fix.<BR>
<BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
And 'systemctl status mysqld.service' (or systemctl status <BR>
mariadb.service) should return something like:<BR>
</BLOCKQUOTE>
<BR>
systemctl status mysqld.service ==> <BR>
<A HREF="mailto:barry@MythBE1">barry@MythBE1</A>:~$ systemctl status mysqld.service<BR>
● mysqld.service<BR>
Loaded: not-found (Reason: No such file or directory)<BR>
Active: inactive (dead)<BR>
<BR>
[Bet we anticipated that from the previous test!]<BR>
<BR>
Now FWIW based on the previous test I tried "systemctl status mysql.service" (no 'd') and got this:<BR>
<BR>
<A HREF="mailto:barry@MythBE1">barry@MythBE1</A>:~$ systemctl status mysql.service<BR>
● mysql.service - MySQL Community Server<BR>
Loaded: loaded (/lib/systemd/system/mysql.service; enabled; vendor preset: en<BR>
Active: activating (start-post) (Result: exit-code) since Wed 2016-07-13 15:3<BR>
Process: 12285 ExecStart=/usr/sbin/mysqld (code=exited, status=1/FAILURE)<BR>
Process: 12281 ExecStartPre=/usr/share/mysql/mysql-systemd-start pre (code=exi<BR>
Main PID: 12285 (code=exited, status=1/FAILURE); : 12286 (mysql-systemd<BR>
CGroup: /system.slice/mysql.service<BR>
└─control<BR>
├─12286 /bin/bash /usr/share/mysql/mysql-systemd-start post<BR>
└─12354 sleep 1<BR>
<BR>
Jul 13 15:32:59 MythBE1 systemd[1]: Starting MySQL Community Server...<BR>
Jul 13 15:33:01 MythBE1 systemd[1]: mysql.service: Main process exited, code=exi<BR>
lines 1-13/13 (END)<BR>
<BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
If necessary try each of the ExecStartPre and ExecStart lines in <BR>
/lib/systemd/system/mariadb.service file and fix the errors <BR>
individually. You will often see permissions errors or missing folders<BR>
on a just-installed box.<BR>
This is a bit 'belt and suspenders' but I have the following in my <BR>
/etc/rc.d/rc.local file,<BR>
The /etc/rc.d/rc.local file is 'chmod 777' and I have run 'systemctl <BR>
enable rc-local.service' to make systemd run the file.<BR>
</BLOCKQUOTE>
<Snip><BR>
<BR>
OK, looks like the rest is something I need to test so will send this off. <BR>
<BR>
Thanks for the tips!
</BODY>
</HTML>