From: Karl O. P. <ko...@me...> - 2015-08-15 01:32:52
|
I prefer to stay far away from MySQL/MariaDB. Having said that it seems like the first thing you should do is instrument the db so that you can tell when connections might be lost for some reason. You get the message when a db connection goes bad for some reason or another. (For me, it's when my Postgres DB restarts due to security update.) You could ask the MariaDB people for help with this. If you followed the install directions your db was created the same way as everybody else's. You could re-create it the same way, but I don't see a point until you know what's wrong. I'd expect sqlgrey to be configured to log to the mail log, although I suppose yours could be different. This will give you timestamps to use when looking through the db log. On Fri, 14 Aug 2015 12:58:14 -0400 Alex <mys...@gm...> wrote: > Hi guys, > > I sent the message below a few days ago, and haven't received any > responses. I'm still having difficulty with this and really don't know > how to resolve it. It appears to happen to all three systems, two of > which are on the same gigabit network, and the other is on a remote > cable connection (which is the one that least frequently fails). > > Would rebuilding the database help? I haven't even been able to find > an error in the logs related to this, only the email that is sent > every time it happens. > > Thanks, > Alex > > > On Tue, Aug 11, 2015 at 7:13 PM, Alex <mys...@gm...> wrote: > > Hi, > > > > I've been running sqlgrey on fedora for quite a while, set up to use > > DB_Cluster with a single master set up with mysql (mariadb). > > Recently, I've started to receive a large number of "sqlgrey > > encountered an SQL error and triggered a reconnection" messages, > > followed immediately by a reconnection message. > > > > It seems to happen at all times during the day, not just when the > > network or server is congested or busy. > > > > There are three read_hosts, including the master itself, but one of > > the hosts has multiple DNS names, so it may be listed twice. > > > > I don't exactly remember how the database schema was set up, which > > is kind of a problem if I ever had to set it up again. > > > > I've included the sqlgrey.conf config below, in hopes someone had > > some ideas how to troubleshoot this. This is the config that's been > > working fine for more than a year. > > > > # cat sqlgrey.conf |grep '^[a-z|A-Z]' > > loglevel = 3 > > log_override = whitelist:1,grey:3,spam:2 > > reconnect_delay = 5 > > awl_age = 31 > > group_domain_level = 10 > > db_type = mysql > > db_name = sqlgrey > > db_host = ns1.example.com;mysql_connect_timeout=1 > > db_port = default > > db_user = sqlgrey > > db_pass = mypass > > db_cleanup_hostname=ns1.example.com > > db_cleandelay = 1800 > > clean_method = sync > > db_cluster = on > > read_hosts=localhost,ns1.example.com,mail02.example.com,mail03.example.com,mail01.example.com > > prepend = 1 > > admin_mail = 212...@me... > > > > Thanks, > > Alex > > ------------------------------------------------------------------------------ > _______________________________________________ > Sqlgrey-users mailing list > Sql...@li... > https://lists.sourceforge.net/lists/listinfo/sqlgrey-users > Karl <ko...@me...> Free Software: "You don't pay back, you pay forward." -- Robert A. Heinlein |