runc: git bisect start v1.2.0-rc.2 v1.1.0 (manually)#48366
runc: git bisect start v1.2.0-rc.2 v1.1.0 (manually)#48366AkihiroSuda wants to merge 11 commits intomoby:masterfrom
git bisect start v1.2.0-rc.2 v1.1.0 (manually)#48366Conversation
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
git bisect start v1.2.0-rc.2 v1.1.0 (manually)git bisect start v1.2.0-rc.2 v1.1.0 (manually)
69afe3a to
b9b4d81
Compare
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
b9b4d81 to
ad38300
Compare
|
Failure in v1.1.0-753-g321aa20c is different from what was reported in #48160 (comment) https://github.com/moby/moby/actions/runs/10527465798/job/29171779480?pr=48366 https://github.com/moby/moby/actions/runs/10527465798/job/29171799694?pr=48366 |
Co-authored-by: Kir Kolyshkin <kolyshkin@gmail.com> Co-authored-by: Sebastiaan van Stijn <github@gone.nl> Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
|
I think you can work-around failures in runc_dmz using EXTRA_BUILDTAGS="runc_nodmz". This will never compile it (it is disabled by default, so you can just avoid compiling it too). |
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
9c07e6e to
6c34f5e
Compare
|
v1.1.0-843-g823636c3 w/ runc_nodmz fails with the same reasons as v1.1.0-753-g321aa20c, which is different from the OP https://github.com/moby/moby/actions/runs/10528984279/job/29175849263?pr=48366 |
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
c020b44 to
9489e1f
Compare
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
b203238 to
dbc4602
Compare
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
Signed-off-by: Akihiro Suda <akihiro.suda.cz@hco.ntt.co.jp>
|
v1.1.0-929-gf2f16213 (opencontainers/runc@f2f1621 ... Looks like stdout is closed too early on some condition. cc @cyphar @kolyshkin @rata @thaJeztah Its parent commit v1.1.0-928-g8e1cd2f5 (opencontainers/runc@8e1cd2f, ... |
|
Interesting that commit was part of a CVE fix that also went to the v1.1 branch (reading from my phone, so didn't check if the patch was the same there) |
|
Let's continue the discussion in: |
Trying to find out the regression commit that is causing the compatibility issue of runc v1.2 RC.
Related:
Originally posted by @kolyshkin in #48160 (comment)
runc_nodmzrunc_nodmzrunc_nodmz🔥 causing stdio regressionrunc_nodmz(failures are similar to v1.1.0-753-g321aa20c)runc_nodmz(failures are similar to v1.1.0-753-g321aa20c)runc_nodmz(failures are similar to v1.1.0-753-g321aa20c)runc_nodmz(failures are similar to v1.1.0-753-g321aa20c)git bisect start v1.2.0-rc.2 v1.1.0(manually) #48366 (comment))(🟢=Good, 🟡=Pending, 🔴=Bad for the same reason as @kolyshkin reported, ⚪=CI fails due to another reason)