Logins/actions failing for some users for the last hour

Status (by robm at Mon Apr 14 03:13 UTC)
Due to an error while rolling out a change, some web servers were incorrectly rejecting all actions/logins for the last hour. This would have affected about 15% of users trying to use the web interface. This has now been fixed.

Posted in Status. Comments Off

FastMail changelog update

The following changes have been rolled out to production:

  • New keyboard shortcuts in the rich text editor:
    * Ctrl-Shift-8 – make bullet list
    * Ctrl-Shift-9 – make numbered list
    * Ctrl-[ - decrease quote level
    * Ctrl-] – increase quote level
  • New! Use multiple levels of bullet points and numbers to structure your message. Just hit ‘tab’ on a blank list item to increase the indentation, or ‘enter’ on a blank list item to decrease the indentation.
Posted in Status. Comments Off

SSL connections failing for some users

Status (by brong at Thu Apr 10 04:52 UTC)
A failed update has caused missing intermediate certificates on our servers.

We’re working on a fix now.

Update (by brong at Thu Apr 10 04:59 UTC)
All servers are now updated and serving the intermediate certificates correctly again

Posted in Status. Comments Off

One server down

Status (by brong at Mon Mar 31 22:03 UTC)
One IMAP server is having difficulty. Some users will see slow responses or an outage.

Update (by brong at Mon Mar 31 22:14 UTC)
Update: everything is running fine again

Posted in Status. Comments Off

XMPP server unreliability

Status (by robn at Fri Mar 14 04:59 UTC)
The XMPP server has been crashing and restarting every few minutes for a few hours now. The problem has been traced to unusual data received from a client. A workaround has been implemented while we work to understand and fix the problem properly.

Posted in Status. Comments Off

FastMail unavailable

It looks like there’s some network problems at our primary data centre in New York, rendering FastMail unavailable to most users. We’re investigating further.

Update 13:34 AEDT:

The problem was caused by a denial-of-service attack against our network in New York. The data centre staff have implemented workarounds to get us back online and we’re working with them to restore services properly.

Posted in Status. Comments Off

One server running slowly

Status (by brong at Fri Feb 28 22:16 UTC)
After the server failure a couple of days ago, we are re-replicating data to bring every server back to fully replicated status. Unfortunately, this has caused high load on one server, which is running slowly. We are slowing down the replication rate to improve performance again.

Posted in Status. Comments Off
Follow

Get every new post delivered to your Inbox.

Join 3,975 other followers