downtime: August 13-20

We will have downtime sometime between August 13th and 20th, to re-cable the server. The exact date is not yet known. Downtime is expected to be approximately 6 hours.

mail problems -- poor reputation due to compromised user account

Today we have a poor IP address reputation because a bunch of spam was sent through via a compromised user account. The account password has been changed and the email queue cleared of the spam. However, it will be a little while before our email reputation improves. Until that happens, mail will be delayed (queued on or possibly rejected (bounced to the sender).

Sorry for the inconvenience.

UPDATE, 21-Feb 14:37. We now have a neutral senderbase reputation and are no longer listed on the blacklist. The mail queue has mostly drained.

Update on disk problems.

So it seems that bringing the system down to single user mode and letting fsck do its stuff has solved the immediate problem of terrible performance.

But the longer term problems still exist. Both of our RAID arrays are running degraded. Another disk failure and we could lose the data.

Current plan is to ditch the hardware and move the entire system to a VM on a server that being generously donated. Work is in progress on bringing up this VM and migrating everything over.

Stay tuned. More to come.


Twitter account for status information

To augment we recently created the
twitter account @EmployeesORG to report system status. You can follow
@EmployeesORG on twitter or see posts at

Web mail now supported (Roundcube)

We support web access to email, via the link Use your normal username and password to login.

migration complete

DNS records have been updated, and Banjo is the new server.

If there are any problems, send email to


During outages of, you can visit for status updates. is hosted on a different computer in a different datacenter.

Welcome to is now re-opened for logins.

We have done our best effort to recover user data off of /dev/ad4. I remind users we do NOT back up user data and the disaster recovery plan is to wipe the system and start anew. The latter almost happened today. If this is not acceptable to you, you may have outgrown what Employees.Org offers and should consider a commercial service.

There was some file corruption from /dev/ad4 that resulted in Mailman bouncing a lot of messages.

Syndicate content