Show newer

Rebooting Banshee again to satisfy vendor in troubleshooting KVM connectivity.

False alarm on Banshee outage. Looks like it got bumped while plugging in KVM.

We've kicked into overdrive an effort to fight back against botnets that attack our customers. The number of IPs blocked that have no connection to customers or anything they could want is rising every second. Zero false positives, this is very surgical work.

We've limited exim connections to the Pixel server to 30 simultaneous connections per IP, to mitigate a DDOS attack from Amazon SES.

Eagle server maintenance window:
6/4/2021 Friday 10:00pm - 11:30pm Central Standard Time

Expected 30-60 minute downtime. Inbound email is expected to be held by sending servers and delivered after the maintenance. Server is being moved to a new rack a few feet away to obtain network upgrades. Not changing provider, this is Hivelocity maintenance.

Afterlogic webmail added to all non-reseller DA servers (not yet on cPanel boxes) at /afterlogic. Quick intro to calendar/contacts, more documentation to come:

We've purchased licensing for AfterLogic webmail which will include Cal/CardDAV in a simple, easy to use implementation that works with the webmail. Finally, a long requested feature is being checked off our list.

@pros You should get a great connection to our servers as a general rule. Being much further from the servers than you are, we still get great connectivity.

Breaking outbound mail briefly on Friday server to fix exim issue. Email sent during that brief period should be queued, fail to send, and then succeed a few minutes later. The only impact should be a delay for a handful of emails.

If you see any oddities on Roundcube on a DirectAdmin-based server, you may need to log out and back in. Running updates.

This is clearing up. It seems that adjustments to the backup system may have caused this. More in postmortems channel in

Show thread

The root disk partition ran out of space. This is 2x 512G NVMe and shouldn't be storing emails. There is nearly 9TB available on the box where this data should have been stored.

Show thread

Working on issue with Friday server, will explain more after fix.

MXRBL has quickly become a powerhouse at blocking spam. The volume of spam we're blocking each day, without false positives, is incredible. Even if you don't use MXroute, if you host your own mail servers, you are free to use it:

MXroute boosted

Coming soon: a day where no one can make any orders for service or upgrades on MXroute. Date not yet certain, but we will be migrating from WHMCS to HostBill and we will need to lock down changes to the database during the process. More as we know it.

Over 4,000 IPs added to MXRBL today, as a result of a log audit for a spam botnet trend.

Indeed the entire provider in Australia appears to be down. This should be up quickly.

Show older

Legal content allowed, don't be a dick.