You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Packaging securedrop for Tails workstations (freedomofpress/securedrop#3502) -- and maybe later for Qubes admin workstations -- could let us avoid using git on Tails workstations altogether. In the meantime, admins and journalists alike are interacting with git repositories to keep their workstations up-to-date. That means they may also encounter git index corruption issues sooner or later, e.g., after an unclean shutdown. This is especially likely in the context of a live USB operating system like Tails, as USB devices can become unplugged accidentally.
As an interim measure until we switch to full packaging, I'd suggest adding at least some basic troubleshooting steps for recovering a securedrop directory to the docs. A method that has worked for me when I've encountered this issue (at least 2-3 times over the years) is to create a fresh clone, remove the broken .git directory and use the one from the new clone to replace it.
How will this impact users?
It may simplify troubleshooting or rebuild operations for admins significantly.
Related evidence
We've had occasional support tickets related to git index corruption on Tails workstations
I've sporadically encountered the issue myself
The text was updated successfully, but these errors were encountered:
Describe the change
Packaging securedrop for Tails workstations (freedomofpress/securedrop#3502) -- and maybe later for Qubes admin workstations -- could let us avoid using
git
on Tails workstations altogether. In the meantime, admins and journalists alike are interacting with git repositories to keep their workstations up-to-date. That means they may also encounter git index corruption issues sooner or later, e.g., after an unclean shutdown. This is especially likely in the context of a live USB operating system like Tails, as USB devices can become unplugged accidentally.As an interim measure until we switch to full packaging, I'd suggest adding at least some basic troubleshooting steps for recovering a
securedrop
directory to the docs. A method that has worked for me when I've encountered this issue (at least 2-3 times over the years) is to create a fresh clone, remove the broken.git
directory and use the one from the new clone to replace it.How will this impact users?
Related evidence
The text was updated successfully, but these errors were encountered: