@Ronn has made exactly this argument in defense of Fram filters, arguing that "
filtering delayed is not filtering denied."
I personally don't know if this is a reasonable argument or not, but if you look at the responses to that statement, you can see there are definitely those that disagree with this idea-- even to the point of
tossing filters to avoid the risk.
This is my take as well. I'm sticking with non-leakers unless/until there is good data that leaking is not a significant problem, and that will likely require better testing results than we have at the moment.
Totally get that, and I would tend to have the exact same inclination were I to be in your position. If you decide to use the potential leakers you have, you might be comforted by the fact that:
1) Per
@ZeeOSix 's responses to me (just above), there may be a decent chance that the ones you have will not leak even if they are suspected to be prone to leaking.
2) If it does leak, you and
@Ronn may be right that filtering delayed is not filtering denied.
3) Even if there is a leak, and even if some amount of filtering is denied, per what
@Glenda W. just responded, it's possible that there still would be no significant long-term implications, especially if it was just 1-2 OCI's.