From bug 1168658, comment 26: " Yes, I'm surprised that we download so much. We need to investigate with a real profile, ie. not one that is reset all the time to run tests. " I've attached a logcat that is running test_cost_control_data_alert_mobile.py (before the pull request from bug 1177553). After that test has been running, after some while, the safebrowsing stuff seems to get downloaded, which was 4.80MB during this test. I'm not sure I can reproduce with a real profile, although I don't see what the difference between a real profile and a profile from Gaia UI test, except for the fact that it's starting from scratch. For me, it seems that FirefoxOS shouldn't download the safe browsing database while it's on a mobile data connection, because of the costs involved, no?

BTW, I've proposed a patch to switch Safe Browsing to OFF by default on B2G in bug 1180503.

Ok, I guess that would fix this issue.

So I guess this bug can be closed now that bug 1180503 is fixed? (I still don't really understand why it then was enabled in the first place).

Yes, I think the need for this is gone now that it's disabled by default.

Ok, according to comment 4, this is now fixed then (weird that it then had to enabled in the first place).