Multiple services restart on their own hours apart. Why?

[dustindauncey]

Hi there,

I am new to running a fresh Vesta install with the services it brings, and I have proceeded far with hosting 3 small websites right now and their respective emails, etc.

I am curious, when I log into Vesta and click on Services, I notice that the uptime for many of the services seems to restart at together as it is far lower than the uptime of the system. I am not purposely rebooting any services, so what might be a cause I can look at? What files can I review that will assist in troubleshooting?

I've reviewed the messages log file for named as per the Vesta documentation, however it doesn't show any errors from my knowledge, just shows that it was told to "reload" the service and it does so successfully. I'm not seeing anything obvious that shows any problems.

I was also thinking memory because I see drastic changes in memory every couple of hours, but it never actually reaches the maximum (I have a 512 MB VPS) with a maximum used being 478 MB for today, with a minimum free of 12 MB. I have optimized MySQL for memory usage (was set to use up to ~400 MB and is now maxing at ~100 MB. Apache was also optimized for a low memory systems.

Copy of services and their values are below for your reference. Notice how many of the services share the same uptime. This uptime is low many times when I check throughout the day. Not sure why but they almost seem to not ever have an uptime of more than an couple hours.

running
<hostname>
CentOS 6.5 (x86_64)
LoadAverage: 0.01
Uptime: 20 hours

running
nginx
reverse proxy
CPU: 0%
Memory: 268 mb
Uptime: 10 minutes

running
httpd
web server
CPU: 2.5%
Memory: 3045 mb
Uptime: 10 minutes

running
named
dns server
CPU: 0%
Memory: 155 mb
Uptime: 20 minutes

running
exim
mail server
CPU: 0%
Memory: 91 mb
Uptime: 10 minutes

running
dovecot
pop/imap server
CPU: 0%
Memory: 18 mb
Uptime: 5 minutes

running
spamassassin
email antispam
CPU: 0%
Memory: 712 mb
Uptime: 10 minutes

running
mysqld
database server
CPU: 0%
Memory: 641 mb
Uptime: 10 minutes

running
vsftpd
ftp server
CPU: 0%
Memory: 51 mb
Uptime: 10 minutes

running
crond
job scheduler
CPU: 0%
Memory: 114 mb
Uptime: 10 minutes

[mopl]

Hello there,

I also have on my VPS (Debian 7 64 bit) the same problem with VestaCp.
I reboot my server there are no so long ago and returning the next day on the "Services" page, I noticed that the uptime of services "Bind9" and "Apache2" was lower than the uptime of the server. (The services have restarted so alone, why I do not know ...) No error message in the logs.

best regards

[erldcrtz]

according to my experience with vesta.. I dont rely stats and services info from the control panel because its not accurate and still buggy..

I suggest you guys install monit to monitor your server

http://www.tecmint.com/how-to-install-a ... g-program/

and somewhere in this forum has the proper config for monit... search it

[dustindauncey]

according to my experience with vesta.. I dont rely stats and services info from the control panel because its not accurate and still buggy..

I'm not sure I agree with this. Because I was able to find the following in the Apache error log which shows that it was being shutdown from something. Unfortunately I am not aware of how to find out what told it to restart, but it definitely was not me and I'm the only one with access to my server. Multiple services are shutting down at the same time. My memory graph shows that it has never reached its peak yet so I'm guessing it isn't a memory issue which is what my first thought was since I only have a 512 MB VPS.

Tail of my error log from today... (hope this helps somebody diagnose what might be going on. I'm still researching it myself as well.)

Code: Select all

[Mon Jul 28 08:33:21 2014] [notice] caught SIGTERM, shutting down
[Mon Jul 28 08:33:40 2014] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Mon Jul 28 08:33:40 2014] [notice] Digest: generating secret for digest authentication ...
[Mon Jul 28 08:33:40 2014] [notice] Digest: done
[Mon Jul 28 08:33:40 2014] [notice] mod_ruid2/0.9.4 enabled
[Mon Jul 28 08:33:40 2014] [notice] Apache/2.2.15 (Unix) DAV/2 PHP/5.4.31 mod_ssl/2.2.15 OpenSSL/1.0.0-fips configured -- resuming normal operations
[Mon Jul 28 09:02:05 2014] [notice] caught SIGTERM, shutting down
[Mon Jul 28 09:02:23 2014] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Mon Jul 28 09:02:23 2014] [notice] Digest: generating secret for digest authentication ...
[Mon Jul 28 09:02:23 2014] [notice] Digest: done
[Mon Jul 28 09:02:23 2014] [notice] mod_ruid2/0.9.4 enabled
[Mon Jul 28 09:02:23 2014] [notice] Apache/2.2.15 (Unix) DAV/2 PHP/5.4.31 mod_ssl/2.2.15 OpenSSL/1.0.0-fips configured -- resuming normal operations
[Mon Jul 28 10:13:26 2014] [error] [client 173.255.244.30] Invalid method in request \x16\x03\x02\x01o\x01
[Mon Jul 28 10:13:26 2014] [error] [client 173.255.244.30] File does not exist: /var/www/html/error
[Mon Jul 28 13:43:32 2014] [notice] caught SIGTERM, shutting down
[Mon Jul 28 13:43:51 2014] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Mon Jul 28 13:43:51 2014] [notice] Digest: generating secret for digest authentication ...
[Mon Jul 28 13:43:51 2014] [notice] Digest: done
[Mon Jul 28 13:43:51 2014] [notice] mod_ruid2/0.9.4 enabled
[Mon Jul 28 13:43:51 2014] [notice] Apache/2.2.15 (Unix) DAV/2 PHP/5.4.31 mod_ssl/2.2.15 OpenSSL/1.0.0-fips configured -- resuming normal operations
[Mon Jul 28 18:45:09 2014] [error] [client 74.207.244.187] Invalid method in request \x16\x03\x02\x01o\x01
[Mon Jul 28 18:45:09 2014] [error] [client 74.207.244.187] File does not exist: /var/www/html/error

[mopl]

I'm not sure I agree with this. Because I was able to find the following in the Apache error log which shows that it was being shutdown from something. Unfortunately I am not aware of how to find out what told it to restart, but it definitely was not me and I'm the only one with access to my server. Multiple services are shutting down at the same time. My memory graph shows that it has never reached its peak yet so I'm guessing it isn't a memory issue which is what my first thought was since I only have a 512 MB VPS.

Tail of my error log from today... (hope this helps somebody diagnose what might be going on. I'm still researching it myself as well.)

Hello, true, I have exactly the same problem with the same error and I have not been able to find or it comes.

[erldcrtz]

try to remove mod_fcgid from your server

(CentOS)
yum remove mod_fcgid

try not to use it for the time being and see if that helps ... suEXEC is from mod_fcgid and it might be causing troubles

[dustindauncey]

try to remove mod_fcgid from your server

(CentOS)
yum remove mod_fcgid

try not to use it for the time being and see if that helps ... suEXEC is from mod_fcgid and it might be causing troubles

Thank you. I will try this. For now, I noticed this... I had restarted my server this morning to align all the uptime, and everything was going fine up to about 4 hours and then all of a sudden named (which was originally disabled) was running again and that and nginx had an update of 48 minutes versus the other services at 4 hours.

Any thoughts on that?

This is why this is frustrating, because it seems so random. It's not like it's just named and nginx, it's more than usually almost all the services except perhaps crond and maybe vsftp. I do not understand why.

[alexcy]

I have exactly the same issue. I am using fastcgi though, so disabling it is not an option for me.

[erldcrtz]

if you want to disable named follow these steps

(CentOS)

yum remove bind

chkconfig bind off

and remove DNS_SYSTEM='named' in /usr/local/vesta/confvesta.conf

reboot

[dustindauncey]

Just removed named/bind, am hoping it works. But I have my doubts. I'll try to update here when I find out.