Cory Snider

Support for `SOURCE_DATE_EPOCH` has been discussed but not implemented. If I'm not mistaken, this issue is a duplicate of https://github.com/moby/buildkit/issues/1058 Related: https://github.com/moby/moby/issues/42417 https://github.com/moby/moby/issues/40754

Do you want _the daemon_ to wait indefinitely until the task has been deleted so as not to leak c8d tasks, or do you want the _delete API call_ to...

> * We also should not be processing that delete in the event loop, as this blocks other events from being processed. @cpuguy83 I checked into that -- only other...

Simple and straightforward, LGTM! @tych0 please update the PR description and commit message to reflect the updated PR content.

The shortlog for commit 345eb82cf6c3ebd3f0f448441235c208ee5cd5fa still reads `container: add a timeout for deletion`

Annoyingly, `net.ipv4.conf..arp_notify=1` is bugged in at least [the Ubuntu 20.04 kernel](https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1989187) and Docker Desktop's 6.6.16-linuxkit kernel. The gratuitous ARP is allegedly generated in the kernel on link up, but never...

I suspect that the kernel's support for generating unsolicited IPv6 neighbor advertisements is slightly less broken than the IPv4 equivalent. Linux implements Duplicate Address Detection in kernelspace, and when `net.ipv6.conf..ndisc_notify=1`...

@make-github-pseudonymous-again thank you, fixed! I had accidentally switched `HealthStartPeriod` with `HealthStartInterval` in the title and a couple places in the description.

- Reported upstream: https://github.com/golang/go/issues/65290 Only clang is affected; cross-compiling with gcc works fine.

The arm/v5 build failure looks to be due to libatomic not being included in the link. For reasons probably buried in mailing list archives, [gcc](https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81358) and clang don't bother to...