Live data from Hacker News

Technical Details on the Recent Firefox Add-On Outage

hacks.mozilla.org

131–140 of 279 posts

Re: Technical Details on the Recent Firefox Add-On Outage

#131

Earlier quoted context omitted.

> The notion that this was a deliberate ploy to get more people to turn on Studies is surely conspiracy-theorist mumbo-jumbo, but nonsense like this makes me wonder. This is not the case. Please see my response downthread: https://news.ycombinator.com/item?id=19872490

I know full well it's not actually the case, but that doesn't make it not feel like it could be the case. It feels scummy, and I'd expect Mozilla to be above that scumminess. Like, just link to the XPI. Not that hard. The unexplained reluctance to do so is suspicious.

Complexity of test scenarios. There were already ~6 user states on ~20 supported operating systems with ~10 add-on types. There has already been public announcement of a solution for older versions of Firefox without studies including a manual install option.

> For users who cannot update to the latest version of Firefox or Firefox ESR, we plan to distribute an update that automatically applies the fix to versions 52 through 60. This fix will also be available as a user-installable extension. - https://blog.mozilla.org/addons/2019/05/02/add-on-policy-and...

Re: Technical Details on the Recent Firefox Add-On Outage

#132

Earlier quoted context omitted.

So that means going forward if Mozilla don't want you to have an add-on then you won't be able to enable it?

We do not take an editorial stance on add-on content with regard to signing, but we do have the ability to block add-ons that are malicious or which violate user privacy and security. We hope those choices will be few and far between, and that our users will agree with them. Our Add-on Policy is discussed further at https://blog.mozilla.org/addons/2019/05/02/add-on-policy-and...

FWIW, I was genuinely shocked that this could happen, and it has severely damaged my trust in Firefox.

I am extremely wary, having been bitten more than once, about software that automatically updates itself or is otherwise subject to remote interference. I don't run Windows 10. I don't use Chrome for anything important. I avoid subscription-based or activation-required software as much as possible. And in Firefox, I specifically chose to be prompted to install updates (which I usually do immediately, but it's my choice and I can do a quick search first in case there ever is a problem being widely reported).

I find the argument that it's impossible to make this configurable because malware could then circumvent it very weak. If we're talking about that level of interference, anyone with access to the Firefox executable could in theory replace it, and given the open source nature of the Firefox codebase this wouldn't be particularly difficult technically for anyone willing to go to such lengths in the first place.

In any case, even if the argument about hard-coding protection into the executable did stand up to scrutiny, there are alternatives possible instead of retrospectively disabling addons with no possible workaround. Perhaps most obviously, you could show a warning message and require explicit user approval at startup before activating the addon, for example, as is already done with various other useful features that are also potentially open to abuse.

As things stand, far from protecting our privacy and security from malicious addons, the current system in fact deactivated all of our privacy- and security-protecting addons, without warning, right in the middle of browsing sessions. One of these seems to be very much worse than the other in terms of the risks created, and I urge you to consider that when deciding how addons are treated in the future.

Re: Technical Details on the Recent Firefox Add-On Outage

#133
post #30

Earlier quoted context omitted.

As someone who opted out of Studies after the last abuse, I felt betrayed that my addons were, effectively, held hostage behind enabling both telemetry and Studies. I decided to wait. It’s bad optics at the very least. Users who opted in for the update were in fact entered into studies they explicitly wouldn’t have wanted to be in without the lure of an earlier update.

I'm sorry that you felt like you were held hostage by Telemetry/Studies. With the exception of the hotfix, we disabled rolling out new Studies during the incident, and will not be re-enabling them until some time after Monday next week. We are also completely deleting all Telemetry and Studies data received in the week following the incident to ensure we respect people who had concerns like yours, but enabled Studies…

"I'm sorry that you felt like you were..." is the worst form of apology, because it admits no guilt or responsibility. "I'm sorry that you were..." or "I'm sorry that we..." would be a legitimate apology.

That said, nuking this data is the first good thing Mozilla has done in this whole fiasco. It's a small but real act of contrition, so kudos for that.

Re: Technical Details on the Recent Firefox Add-On Outage

#134
post #87

It might not have helped here, but updating certs every several years always leads to problems from my experience - people move on, processes get lost or outdated, etc. I'll accept the yearly annoyance to avoid those issues every time.

Usually organisations of size of Mozilla have metrics and alarms for various things, cert expiry should have been on top priority.

I’ve never worked in a company so large that certificate expiration was a completely solved problem. I’ve seen it bite every company I’ve worked for, and I’ve been part of ones much larger than Mozilla.

Re: Technical Details on the Recent Firefox Add-On Outage

#135
post #11

Looks like a good read. I haven't finished reading it yet, but there's something I still don't get ... Windows and macOS both have a signing infrastructure for apps. The rules of that infrastructure dictate only that apps must have been signed by a valid certificate at the time they were signed. That way old app downloads don't need to be periodically re-signed just to account for expiring certificates. I can downloa…

Technically I think one must be clear. The cert (client and ca) all have dates and that you are right. But what your miss is that the revocation measure is the key here. It is not just sign it for the period. The CA and the root CA list has to maintain those cert and revoke them. You should not just trust the date and signature. The revocation list is key.

In other words you check everytime you use it basically. That is how stolen Microsoft cert stopped.

All old app are checked for validity. It is not as you implied.

Re: Technical Details on the Recent Firefox Add-On Outage

#136
post #66

My point of view as a long-time Firefox user that cares about privacy but also knows we live in an imperfect world: It obviously sucks that this happened but I think they handled it very well. The bug was fixed so quickly that I wouldn't even have realized it had happened if it hadn't been for the thread here on HN. My extensions hadn't even been disabled yet by the time the patch came out. And pushing out the hotfix…

It was a couple of days before the Android fix was released. For many users this was a multiday affair. Some lost data (containers). You were lucky you extentions were not disabled before the fix, but for many people this was a major problem.

I don't think containers are supported on Android.

But it did take a few days for the Android fix to be released on Google Play, so it was more annoying than on the desktop.

Re: Technical Details on the Recent Firefox Add-On Outage

#139

Earlier quoted context omitted.

One option would've been to use Nightly and set xpinstall.signatures.required = false in about:config. That's exactly what I did.

I view using nightly to be as much of a barrier to entry that I might as well switch to a totally different browser. If the issue had lasted longer, I would have found (someone on the internet who had found) a solution like that.

On Android, you already had the xpinstall.signatures.required option without having to install nightly. Linux too. Took me maybe 30 second to fix all my devices.

Re: Technical Details on the Recent Firefox Add-On Outage

#140

My point of view as a long-time Firefox user that cares about privacy but also knows we live in an imperfect world: It obviously sucks that this happened but I think they handled it very well. The bug was fixed so quickly that I wouldn't even have realized it had happened if it hadn't been for the thread here on HN. My extensions hadn't even been disabled yet by the time the patch came out. And pushing out the hotfix…

Handled it very well? That must be the worst hyperbole I have ever seen in a while. It broke extensions suddenly for millions of users. If you went on Firefox.com there was absolutely no news about it. Completely awful lack of communication. You had to find a minuscule banner on the "find a fix" page in order to hear about the problem.

Add to that that the recommended fix was to activate the backdoor of Firefox, thats just a horror story from beginning to the end.

And no admission of guilt anywhere.

Post reply on HN