Ubuntu users left waiting after Canonical's servers take weekend off
- Reference: 1757343548
- News link: https://www.theregister.co.uk/2025/09/08/canonical_server_outage/
- Source link:
Canonical's [1]status page shows that both security.ubuntu.com and archive.ubuntu.com experienced brief issues on September 5 and 7. The incidents appeared short-lived, ending with the reassuring "All components are Operational" message. Case closed, right?
Not exactly. While Canonical's servers came back online quickly, the real problems were just beginning. Users flooded the company's forums throughout the weekend, reporting failed installations and frozen updates. The brief server outages had created a processing backlog that left Ubuntu's repositories effectively broken for days.
[2]
"They say the outage was only 36 minutes, but two days later it still isn't working, a frustrated user told The Register yesterday.
[3]
[4]
"I'm trying to install Ubuntu Server 24.04.2 LTS on a machine right now, and I just can't because it freezes in the middle of the install because it can't download some of the packages."
Our source wasn't alone. A look at Canonical's forums indicated plenty of users encountering the same issue, prompting terse responses from Canonical representatives who eventaully shut down the discussion.
[5]
The Ubuntu Studio Project Leader, Erich Eickmeyer, [6]posted : "We don't need a bunch of 'Can Confirm' and 'me too' posts. Instructions were given as to what needs to happen as 1) a workaround, and 2) what you need to do to get the repos working (you can't)."
So there.
[7]Ubuntu 24.04.3: Noble Numbat point release slips out quietly
[8]Linux is about to lose a feature – over a personality clash
[9]Linux Lite relief: 7.6 keeps it simple, shiny, and mostly slim
[10]The plan for Linux after Torvalds has a kernel of truth: There isn't one
The underlying problem was that while the servers themselves recovered quickly, they couldn't process the accumulated backlog of requests. Users found themselves unable to install fresh Ubuntu systems or download security updates until Monday, September 8. Canonical's [11]official advice , buried in forum posts, boiled down to "wait it out" until everything synced.
Canonical did not respond to our request for comment, but as one source put it, losing security.ubuntu.com could be a serious issue if millions of systems were unable to download a critical update.
A status page claiming all is well when the customer experience is quite different is becoming all too familiar and frustrating for users.
[12]
Uptime percentages that reflect reality might look a little less rosy, though they'd almost certainly cut down on the number of "me too" posts clogging up the forums when the next outage strikes. ®
Get our [13]Tech Resources
[1] https://status.canonical.com/
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/oses&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aL79ngwV9xTAIU8KkPPP9QAAAMk&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/oses&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aL79ngwV9xTAIU8KkPPP9QAAAMk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/oses&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aL79ngwV9xTAIU8KkPPP9QAAAMk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/oses&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aL79ngwV9xTAIU8KkPPP9QAAAMk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://discourse.ubuntu.com/t/is-there-a-problem-with-archive-ubuntu-com/66686/77
[7] https://www.theregister.com/2025/08/08/ubuntu_24043_quietly_slips_out/
[8] https://www.theregister.com/2025/08/15/sad_end_of_bcachefs/
[9] https://www.theregister.com/2025/09/04/linux_lite_76/
[10] https://www.theregister.com/2025/08/14/the_plan_for_linux_after/
[11] https://discourse.ubuntu.com/t/current-status-of-ubuntu-update-errors-please-read/66809
[12] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/oses&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aL79ngwV9xTAIU8KkPPP9QAAAMk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[13] https://whitepapers.theregister.com/
Just for fun, I let a laptop install run with bytes dripping in pretty much one at a time. The connection dropped around the 3-hour mark.
A linux-firmware package seemed to be the biggest problem for many, including me. I was able to do the install by grabbing the firmware package off a local mirror, doing the install but opting out of the usual "install updates during the install" then installing the firmware update, rebooting, and updating per usual.
I saw reports that security.ubuntu.com is intentionally not mirrored. True? Common practice? If it's true, seems like you need a plan.
It looks like they reported down time on the main server but not on the mirrors which would have been more realistic. It also looks as if some (?all) the mirrors were mid-sync. Maybe they needed to completely re-sync to ensure consistency. It's something the .deb world needs to look at in case it's a problem that affects other server/mirror systems.
If you are a company, you really should be hosting your own internal mirror so that you don't experience these types of outages / dependencies. If you have multiple ubuntu servers, its only a couple of TB for the repos of the last 3 releases. Will save bandwidth, increase update speeds, reduce load for canonical, and remove you external dependencies.
Explains why I couldn't get a machine running Friday. But it was a Friday so for me it boiled down to "I'll look at it Monday"