Mailcleaner cluster. Mail is in quarantaine but not shows in webinterface

Problems/questions regarding the MailCleaner configuration

Moderators: FlorianB, Pascal, bourgeois, mentor

jeroencoco
Posts: 8
Joined: Mon Oct 30, 2017 10:08 am
How did you hear about Mailcleaner: Google

Mailcleaner cluster. Mail is in quarantaine but not shows in webinterface

Postby jeroencoco » Tue Oct 31, 2017 1:19 pm

Hi All,

We are looking in to mailcleaner with clustering too see if it is something for us. For testing purpose i have setup 3 hosts (1 management and 2 'workers'). While testing them out i have discovered that if we disconnect the network from the management host and send mails to the 2 workers they have trouble to move the emails in to the quarantaine. To be more specific; it seems that if we disconnect the network from the management hosts the workers have trouble to connect to the local mysql-slave instance? As the management-host is also offline, the quarantaine e-mail is fanashid and cannot be traces again via the webinterface or found on any database on the workers.

We have the following trace from the log that's indicating this:

Code: Select all

2017-10-31 11:43:22 (4) 803258: message 1e9U0m-000EWn-PF R:<testbox@test.com> S:<number5@test.com> score: 1 status: want quarantine
2017-10-31 11:43:22 (4) Error while logging msg 1e9U0m-000EWn-PF to db slave, retrying, if no further message, it's ok
2017-10-31 11:43:22 (4) Connecting to database slave
2017-10-31 11:43:22 (4) Connecting to database realmaster
2017-10-31 11:43:38 (4) Error, could not connect to db realmaster
2017-10-31 11:43:38 (4) Error while executing log query (msgid=1e9U0m-000EWn-PF, db=slave): MySQL server has gone away
2017-10-31 11:43:38 (4) Error while logging msg 1e9U0m-000EWn-PF to db realmaster, retrying, if no further message, it's ok
2017-10-31 11:43:38 (4) Connecting to database realmaster
2017-10-31 11:43:41 (4) Error, could not connect to db realmaster
2017-10-31 11:43:41 (4) Error while executing log query (msgid=1e9U0m-000EWn-PF, db=realmaster): MySQL server has gone away


As far as i understand the whole-thing. Is that the worker-nodes or slave-nodes can work without a management-host and put the spam/quarantaine message in the slave-database on the server self. When the management-server comes back online the mc_spool database on the slave instance syncs back to the management-server. Or do i forget something?

Greetz,
Jeroen.

Return to “Configuration”

Who is online

Users browsing this forum: No registered users and 4 guests