Last 10 major outages posts, sorted by last updated
Order posts by limited to posts

MAJOR Closed Routing - General
AFFECTED
Routing - General
STARTED
Apr 26, 09:30 AM (5 months ago)
CLOSED
Apr 26, 10:00 AM (5 months ago)
DESCRIPTION
One of our routers has restarted and is causing some routing problems. Investigations underway...
Resolution:

MAJOR Closed General
AFFECTED
General
STARTED
Mar 30, 10:00 AM (6 months ago)
CLOSED
Mar 30, 11:00 AM (6 months ago)
DESCRIPTION
We're investigating connectivity problems.
Resolution:

MAJOR Closed L2TP and Routing
AFFECTED
L2TP and Routing
STARTED
Jan 26, 08:57 AM (8 months ago)
CLOSED
Jan 26, 09:40 AM (8 months ago)
DESCRIPTION
We say some L2TP sessions drop at around 08:57 - we are investigating this and a possible routing problem.
Resolution:

MAJOR Closed Line Drops
AFFECTED
Line Drops
STARTED
Dec 15, 01:00 AM (9½ months ago)
CLOSED
Dec 15, 01:58 AM (9½ months ago)
DESCRIPTION
We're investigating dropping connections and routing problems that started at 1AM (15th December)
Resolution: Many of our routers lost their BGP connection to our internal 'route servers' - this caused some of our LNSs and edge routers to loose connectivity and their connection. This resulted in customers being unable to reach many places on the internet for a short while. Logs and other data have been captured and will be further investigated to find the cause.

MAJOR Closed General
AFFECTED
General
STARTED
Nov 21, 07:25 PM (10¼ months ago)
CLOSED
Nov 21, 07:45 PM (10¼ months ago)
DESCRIPTION
We're investigating a network incident that occured at around 19:30, updates to follow.
Resolution: Service restored and back to normal at around 19:50. This was due to a network issue that affected some of our BGP sessions which caused routing problems to certain destinations.

MAJOR Closed Broadband
AFFECTED
Broadband
STARTED
Oct 20, 10:20 AM (11¼ months ago)
CLOSED
Oct 20, 10:45 AM (11¼ months ago)
DESCRIPTION
Lines connected on the witless LNS dropped and are in process of reconnecting.
Resolution: The Witless LNSs suffered a crash which was caused by a bug in our CQM graphing code. This bug has been fixed and is in the process of being tested and we will then roll out the new software to the the Witless LNSs overnight as 'emergency' planned work.

MAJOR Closed TalkTalk
AFFECTED
TalkTalk
STARTED
Sep 30, 03:24 PM (12 months ago)
CLOSED
Sep 30, 04:36 PM (12 months ago)
DESCRIPTION
at 15:24 we saw most of our TalkTalk circuits drop. They are in the process of reconnecting.
Resolution: Lines reconnected within a couple of minutes and we're investigating this further with TalkTalk. Further updates about this will be on https://aastatus.net/42468

MAJOR Closed TalkTalk
AFFECTED
TalkTalk
STARTED
Sep 27, 10:00 PM (1 year ago)
CLOSED
Sep 28, 08:00 AM (1 year ago)
DESCRIPTION
Following on from the problems this morning, TalkTalk are performing some emergency work overnight, tonight. The work will start at 10PM and will carry on through to 6AM. They say that their work before midnight should not impact circuits, but the work after midnight may cause some reconnects.
Resolution: TalkTalk have emergency overnight work that failed and turned in to an outage causing circuits to drop. We took our Telehouse connection out of service and lines reconnected via our Equinix link and have been stable since 08:30. We'll leave the Telehouse link down for the moment in case TalkTalk schedule further work on it.

MAJOR Closed Broadband
AFFECTED
Broadband
STARTED
Jul 20, 01:00 AM (1 year ago)
CLOSED
Jul 20, 07:29 AM (1 year ago)
DESCRIPTION
We're investigating a problem causing some lines to be reconnecting.
Resolution: A bug has been found in the LNS code which erroneously limited the number of connections that would successfully connect. This has been fixed. We will update the software on our test LNS (B.Doubtless) overnight. We'll then reschedule an upgrade of the production LNS (X.Witless) and bulk-move of customers for later in the week.

MAJOR Closed Internet
AFFECTED
Internet
STARTED
Jul 06, 02:45 PM (1 year ago)
CLOSED
Jul 06, 03:08 PM (1 year ago)
DESCRIPTION
at around 14:45 there was some sort of problem that affected traffic, we're investigating the cause.
Resolution: We caused a network loop between two of our switches in one of our London datacentres. This was caused by loading wrong software variant on to one of our test FireBrick FB9000 routers; the software variant that was loaded didn't have the LACP feature and when the FireBrick booted it caused a network loop. This caused a broadcast storm and general chaos until the switch disabled the ports. We do apologise for the inconvenience this caused customers this afternoon, lesson learnt and software builds have been changed to include LACP in all variants.