Handle podman-remote run --rm#7224
Conversation
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: rhatdan The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
@edsantiago @mheon PTAL |
|
Is this intended to fix #7119? If so could you |
|
LGTM |
|
Flake in special_testing_rootless: $ podman-remote --url unix:/tmp/podman.Hu25Ai run --rm quay.io/libpod/alpine_labels:latest cat -v /proc/self/attr/current
[+0224s] # read unixpacket @->/run/user/14096/libpod/tmp/socket/32f534034d29cb66f3d3d341a7ddbec1bb13244ee426497a2ffed12579e4f1fb/attach: read: connection reset by peerI saw that on one CI run in #7111 but was unable to reproduce it (despite a lot of effort) hence did not file an issue. The socket in question is a conmon one, not the podman-remote socket. I suspect there is a troublesome interaction between conmon and podman-remote but I don't know how to even file an issue without a reproducer. |
|
LGTM but (if this really is intended as a fix for #7119) I'd really like those tests enabled |
|
FWIW I pulled the PR, removed the In case it helps: $ perl -i -ne 'print unless /7119/' test/system/*bats |
We need to remove the container after it has exited for podman-remote run --rm commands. If we don't remove this container at this step, we open ourselves up to race conditions. Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
|
Fixes: #7119 |
|
LGTM |
|
CI passed all in one go, no flakes. /lgtm |
We need to remove the container after it has exited for
podman-remote run --rm commands. If we don't remove this
container at this step, we open ourselves up to race conditions.
Signed-off-by: Daniel J Walsh dwalsh@redhat.com