Bugzilla

(In reply to Chris Manchester (:chmanchester) from comment #9)

> My only knowledge of this file is that the updater treats it specially, which can cause problems when updating its contents (as seen in bug 1431342).

That is by request of QA and RelEng. I've often wanted to remove this special handling but it would require more work especially by RelEng to support updating beta users to RC's.

(In reply to :Gijs (he/him) from comment #5)
> If there is some reason that I don't yet see why that can't work, we could perhaps move the reading of this file off the main thread and off the startup path, and delay starting the update service until we have read the relevant value... but that seems like more work to me.

The update service already only loads a very small stub unless there is an update in progress in which case it is likely needed. Telemetry on the other hand add this to the environment.

I've also considered using a plain file to read this value since it doesn't have to be a pref file but I don't think that would help much here though it would remove it from the read prefs path .

(In reply to Chris Manchester (:chmanchester) from comment #9)

> My only knowledge of this file is that the updater treats it specially, which can cause problems when updating its contents (as seen in bug 1431342).

That is by request of QA and RelEng. I've often wanted to remove this special handling but it would require more work especially by RelEng to support updating beta users to RC's.

(In reply to :Gijs (he/him) from comment #5)
> If there is some reason that I don't yet see why that can't work, we could perhaps move the reading of this file off the main thread and off the startup path, and delay starting the update service until we have read the relevant value... but that seems like more work to me.

The update service already only loads a very small stub unless there is an update in progress in which case the channel is likely needed. Telemetry on the other hand add this to the environment.

I've also considered using a plain file to read this value since it doesn't have to be a pref file but I don't think that would help much here though it would remove it from the read prefs path .

(In reply to Chris Manchester (:chmanchester) from comment #9)

> My only knowledge of this file is that the updater treats it specially, which can cause problems when updating its contents (as seen in bug 1431342).

That is by request of QA and RelEng. I've often wanted to remove this special handling but it would require more work especially by RelEng to support updating beta users to RC's.

(In reply to :Gijs (he/him) from comment #5)
> If there is some reason that I don't yet see why that can't work, we could perhaps move the reading of this file off the main thread and off the startup path, and delay starting the update service until we have read the relevant value... but that seems like more work to me.

The update service already only loads a very small stub unless there is an update in progress in which case the channel is likely needed. Telemetry on the other hand add this to the environment.

I've also considered using a plain file to read this value since it doesn't have to be a pref file but I don't think that would help much here though it would remove it from the read prefs path.