General error: 2006 MySQL server has gone away

R

Rod Rayner

Member
Trying to run a campaign and its getting stuck at 31%.

I have checked the application logs and its full of messages saying General error: 2006 MySQL server has gone away.

I have recently upgraded from 1.5.1 to the latest version, other than that there are no changes other than added new subs to lists.

I have restarted the server and tried several times to restart the campaign. The server is a dedicated server running at AWS using SMTP2Go as a delivery server. Never had any issues before but i have upgraded.

Any idea's ?
 
just added the two innodb lines ( were they supposed to be there already ? ) and changed max connections to 500... re-started SQL and still getting the following error

2019/03/22 18:03:25 [error] [system.db.CDbCommand] CDbCommand::execute() failed: SQLSTATE[HY000]: General error: 2006 MySQL server has gone away. The SQL statement executed was: UPDATE `mw_campaign` SET `status`=:status, `last_updated`=NOW() WHERE campaign_id = :cid.
2019/03/22 18:03:25 [error] [application] CDbCommand failed to execute the SQL statement: SQLSTATE[HY000]: General error: 2006 MySQL server has gone away
2019/03/22 18:03:25 [error] [application] CDbCommand failed to execute the SQL statement: SQLSTATE[HY000]: General error: 2006 MySQL server has gone away

Any more idea's
 
What's your sending settings from backend > settings > cron ? (show a screenshot)
Also, how about your server resources, ram/cpu/storage?
 
here you go..

upload_2019-3-22_18-28-58.png

The server is a dedicated AWS EC2 T2.Medium, 4Gb RAM & 2 cores. Have only sent 1 campaign out today which is the one that is having the issue and max CPU usage has never gone over 25%. The server normally runs fine and the only thing i have done recently is upgraded to latest version of Mail Wizz. Our PHP version is 7.0.33-0ubuntu0.16.04.2
 
The sending settings are pretty low, they should not be a problem.
Did you enable pcntl sending in backend > settings > cron ?
 
Hmm, put it back on disabled, i asked because generally that will eat some resources.

Okay, in same page( backend > settings > cron ) what's the setting for "Delete delivery logs" ?
If it's set to No, can you set it to Yes? Then when the cron job runs at midnight, it will clean your database a lot, which should avoid this issue.

If you don't want to wait till midnight, you can manually run the cron job yourself from ssh.
If in backend > misc > cron jobs list, your cron job for daily looks like:
Code: 
0 0 * * * /usr/bin/php -q /var/www/vhosts/newsletter.onetwist.com/httpdocs/apps/console/console.php daily >/dev/null 2>&1
Then from ssh, run:
Code: 
/usr/bin/php -q /var/www/vhosts/newsletter.onetwist.com/httpdocs/apps/console/console.php daily --verbose=1

If none of those work, open a ticket with backend url and login and ssh access to the server and i'll take a look.
 
Ok so i have disabled the PCNTL again and checked the delete delivery log setting. This was set to no so i have changed to yes and ran the cron job manually. I resumed the campaign and still stuck, i checked the application log and still getting the error message saying SQL Server has gone away.

I create a ticket so you can check the backend config.

Thanks
 
Alexhere said:
I revise all the setting，But Still can't work ,Can i refund?
Click to expand...
You can refund. You can try to upgrade your server specifications, add more RAM to it and then this issue will go away, because this error happening when your mysql server cannot handle the requests and throw that error.
 
You must log in or register to reply here.
Back
Top