Delays when checking out GitHub repos via SSH
Incident Lifecycle
Incident Timeline
Monitoring
A fix has been implemented, and everything should be back to normal. The previously shared snippets can be removed.
May 20, 2026 at 3:02 PM UTC
Identified
We identified slow GitHub operations related to problems with the network, and we deployed a fix to re-route those operations to a healthier server. Some HTTPS operations started to fail because of that. To disable the fix, you should use the command:
sudo sed -i 's/^\([0-9].*github\.com\)/#\1/' /etc/hosts
And to re-enable, you can use:
sudo sed -i 's/^#\([0-9].*github\.com\)/\1/' /etc/hosts
sudo sed -i 's/^\([0-9].*github\.com\)/#\1/' /etc/hosts
And to re-enable, you can use:
sudo sed -i 's/^#\([0-9].*github\.com\)/\1/' /etc/hosts
May 20, 2026 at 1:59 PM UTC
Monitoring
A fix has been implemented and we are monitoring the results.
May 20, 2026 at 12:48 PM UTC
Identified
We identified slow communication between Semaphore jobs and GitHub. We are continuing to work with our upstream provider to resolve this. For impacted jobs, please replace the normal checkout command with the following:
retry --times 5 "timeout 40 bash -lc 'source ~/.toolbox/toolbox; checkout'"
And afterwards, use cd to change the directory to the cloned repository (replace with your project).
retry --times 5 "timeout 40 bash -lc 'source ~/.toolbox/toolbox; checkout'"
And afterwards, use cd to change the directory to the cloned repository (replace with your project).
May 20, 2026 at 8:44 AM UTC
Was your business affected by this Semaphore CI outage?
Set up instant alerts for Semaphore CI, be the first to know about outages via email, Slack, Teams, or Discord.