since 1.260 upgrade, mysql is strange

[CiscoMike]

It's happened 3 nights in a row now and I'm baffled. Right at midnight, server time, MySQLd gets restarted. This scenario repeats itself every 5 minutes for an hour (12 restarts) and then I don't see it again for another 23 hours. At 10 minutes after the hour, httpd also gets restarted.

I have 3 cron jobs, one for bfd, one for prm and the other is directadmin's (and boy is it chatty). It runs a datatq script but nothing unusual that I had seen. I checked mysqld.log and it only had that the server was being restarted (stop, start, listening, repeat). Nothing in messages. The cron log only shows directadmin doing it's thing every minute and then "tally" running at 12:01am and again at 12:08am.

So I'm at a loss. Where would I go looking or changing whatever happened w/ the 1.260 update? It just seems really bizarre but it also bothers me because I have an hour of instability now too.

 

[CiscoMike]

some additional info, when in DA and looking at the processes, mysqld only has a start/stop and not a restart option like the other processes do however it had grabbed the correct pid.

I'm thinking about killing PRM to see if that helps but still, why would PRM freak out every night at midnight when it's running every hour as is?

 

[CiscoMike]

further looking into this, I notice that the PID that DA has is the child process. Example:

root 18966 1 0 11:23 ? 00:00:00 /bin/sh /usr/bin/mysqld_safe --datadir=/var/lib/mysql --pid-file=/var/lib/mysql/<servername>.pid

mysql 19726 18966 0 11:23 ? 00:00:00 /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/<servername>.pid

not sure if that's the problem. I disabled SIM and PRM because SIM was the one crying about mysql being down and my SIM log kept thinking MySQL was down even though I had the above two pids up and running. I'll see if that helps. Kinda bummed if it turns out to be SIM.

 

[CiscoMike]

hmmm

[mysqld]
#datadir=/var/lib/mysql
#socket=/var/lib/mysql/mysql.sock
skip-locking
skip-innodb
skip-networking
query_cache_limit=1M
query_cache_size=32M
query_cache_type=1
max_connections=500
interactive_timeout=100
wait_timeout=100
connect_timeout=10
thread_cache_size=128
key_buffer=16M
join_buffer=1M
max_allowed_packet=16M
table_cache=1024
record_buffer=1M
sort_buffer_size=2M
read_buffer_size=2M
max_connect_errors=10
# Try number of CPU's*2 for thread_concurrency
thread_concurrency=2
myisam_sort_buffer_size=64M
log-bin
server-id=1

[mysql.server]
user=mysql
basedir=/var/lib

[safe_mysqld]
err-log=/var/log/mysqld.log
#pid-file=/var/lib/mysql/mysql.pid
open_files_limit=8192

[mysqldump]
quick
max_allowed_packet=16M

[mysql]
no-auto-rehash
#safe-updates

[isamchk]
key_buffer=64M
sort_buffer=64M
read_buffer=16M
write_buffer=16M

[myisamchk]
key_buffer=64M
sort_buffer=64M
read_buffer=16M
write_buffer=16M

[mysqlhotcopy]
interactive-timeout

maybe I figured this out on my own but I'd still like a second opinion. notice I do the skip-networking part. Now when I look at conf.sim, it's looking for MySQL to be on port 3306 which isn't there because I told MySQL to stop listening externally. from conf.sim:

# The below values are used to check against the netstat output
# and make sure the given service is operating on the said port

FTP_PORT="21"                                           # TCP/IP port for FTP
HTTP_PORT="80"                                          # TCP/IP port for HTTP
DNS_PORT="53"                                           # TCP/IP port for DNS
SSH_PORT="22"                                           # TCP/IP port for SSH
MYSQL_PORT="3306"                                       # TCP/IP port for MySQL
EWP_PORT="19638"                                        # TCP/IP port for Ensim
XINET_PORT="143"                                        # TCP/IP port for an XINET service
SMTP_PORT="25"                                      # TCP/IP port for SMTP service

hmmm.

 

[DirectAdmin Support]

Hello,

Check:
Admin Level -> Full System Backup.

The sysbk script is the only cron that will ever touch the mysql processes. Try disabling the mysql portion of the sysbk backup to see if it stops the problem.

John

 

[CiscoMike]

thanks but it turned out to be System Integrity Monitor. Too bad too, I had to shut it off. Apache still bounces once a night @ 12:12am and I can't figure that one out. No crontab or cron jobs for that but every darn night, 12:12am, http received TERM signal. Then 5 seconds later, httpd successfully restarted. Go figure.

No DA backups, no SIM, no PRM. Oh well...

 

[harro]

Hi CiscoMike,

I was looking around for some info on a problem I have and after reading your post bumped into a remark elsewhere that might give a clue:

quote:"How often and when is Webalizer run?"

Webalizer is run nightly at 12:10 am. It can be changed in the /etc/cron.d/directadmin_cron file to do it more often. It's under "action=tally&value=all"

Maybe your webalizer is causing HTTPD to restart?

 

[Chrysalis]

apche bouncing nightly is normal for log rotation purposes it needs to be restarted.