A brief issue with lists.sr.ht caused a delay in mail delivery: A problem with a recent lists.sr.ht deployment went unnoticed for about an hour and a half, during which time incoming emails were not processed or forwarded. Service has been restored and all of the missed emails are being automatically processed now. No emails have been dropped; it is not necessary to send them again. (14:48 UTC — Oct 20)
A distributed denial of service (DDoS) attack on todo.sr.ht is underway: A denial of service attack against todo.sr.ht is currently underway. We suspect it is motivated by hate sentiments against a transgender maintainer on SourceHut. The attack is presently under control, but please let us know via IRC if you experience any problems. (11:00 UTC — Jul 23)
Planned maintenance is complete. Normal service is restored. Thank you for your patience. (16:30 UTC — Feb 28)
Planned maintenance is underway. All SourceHut services will be affected. Barring unforseen circumstances, the maintenance window is scheduled for two hours. (15:00 UTC — Feb 28)
Phase three of our Alpine Linux 3.15 roll-out will cause major service disruptions. We will be rolling out Alpine Linux 3.15 to more of our hosts on Monday the 28th at 15:00 UTC.
Maintenance complete. We have completed all maintenance operations planned for today and service is restored to normal. (17:15 UTC — Feb 21)
Planned maintenance may exceed the scheduled maintenance window. We are still working on the roll-out and now expect to complete the migration before 18:00 UTC. (17:00 UTC — Feb 21)
Planned maintenance is underway. Expect intermittent outages for all services. (15:00 UTC — Feb 21)
Phase two of our Alpine Linux 3.15 roll-out will cause major service disruptions. We will be rolling out Alpine Linux 3.15 to more of our hosts on Monday the 21st at 15:00 UTC. This upgrade is expected to cause major service disruptions lasting between 5 and 15 minutes each; generally one service at a time will be affected.
Today’s scheduled maintenance is complete. We have completed our maintenance work for today, and all services are restored to normal operation. Thank you for your patience. (10:45 UTC — Feb 14)
The migration is underway. We have started this migration work and expect to complete it no later than 12:00 UTC. (10:00 UTC — Feb 14)
Phase one of our Alpine Linux 3.15 roll-out will cause minor service disruptions. We will be rolling out Alpine Linux 3.15 to some of hosts on Monday the 14th at 10:00 UTC. This will disrupt blob storage (causing artifacts on builds.sr.ht and git.sr.ht to become unavailable) and will temporarily reduce build throughput (possibly causing longer waits for builds to start executing). Additionally, metrics.sr.ht will be briefly unavailable.
An incomplete deployment caused a brief pages.sr.ht outage: pages.sr.ht suffered 13 minutes of downtime following an incomplete deployment which failed to apply a database migration correctly. (10:46 UTC — Feb 3)
The issue has been resolved: The upstream network has corrected the problem. Normal service should be restored. Over the course of 19 minutes, we dropped 10 of our ~500 upstream chat.sr.ht connections. We are still experiencing internal issues, but they are not service impacting and we expect them to clear up shortly. (20:22 UTC — Dec 20)
An ongoing issue with our upstream network provider is causing chat.sr.ht reliability problems: An issue with our network provider is causing some unreliability for our services. The full extent of the problem is still not known, but the most important consequence of it so far is that chat.sr.ht connections to upstream networks are unreliable. You may experience disconnections and intermittent service. (20:03 UTC — Nov 16)
A deployment issue caused a brief outage on meta.sr.ht: An updated for one of our dependencies was not correctly verified before being deployed to production, resulting in a 6 minute outage. Only meta.sr.ht was affected; other services remained available for the duration of the outage. (08:13 UTC — Nov 16)
HTTP(s) clone access has been restored: The issue was narrowed down to a stale nginx configuration which directed authorization requests to the wrong port. The configuration has been corrected and normal service is restored. (08:35 UTC — Oct 27)
HTTP(s) clone is currently unavailable for hg.sr.ht: A planned deployment ran afoul and has caused some problems with cloning hg.sr.ht repositories via https. An investigation is ongoing. SSH clone is still working correctly. (08:20 UTC — Oct 27)
Planned maintenance is complete: We have finished upgrading all of the affected hosts and all services are restored to normal. (14:20 UTC — Aug 16)
Planned maintenance is ongoing: Our maintenance window has opened and we have started our work. (13:00 UTC — Aug 16)
Planned maintenance on August 16th will cause intermittent outages: We are planning the second (and last phase) of the maintenance which began on August 3rd. This will affect all services, causing intermittent outages that are expected to last between 15 and 30 minutes at most. The total maintenance period should last less than 2 hours. (13:00 UTC — Aug 16)