On Wed, Dec 10, 2025 at 04:39:11PM -0600, Sami Imseih wrote:
> On 15 STABLE, I encountered the same issue [0] which was fixed in 6a20b04f0408
> starting from 16. It seems like a decision not to backpatch to lower
> versions at the time was due to a lack of reports [1], but I think since we have to
> support 14 and 15 for a few more years, it makes sense to.
The issue is that Sami has been able to trigger this problem
internally. :)
> I can prepare the patches if there are no objections.
If you could prepare a patch, that would be great! Let's do that if
v14 and v15 can also face the issue under a high-concurrent build.
The blast area of 6a20b04f0408 is not that bad.
--
Michael
