Commit Graph

12 Commits

Author SHA1 Message Date
Phil Estes
06e04bc5a9
Merge pull request #2830 from Ace-Tang/support_cr_without_image
cr: support checkpoint/restore without image
2018-12-20 13:24:37 -05:00
Ace-Tang
6593399e9f cr: support checkpoint/restore without image
support checkpoint without committing a checkpoint dir into a
checkpoint image and restore without untar image into checkpoint
directory. support for both v1 and v2 runtime

Signed-off-by: Ace-Tang <aceapril@126.com>
2018-11-29 10:19:39 +08:00
Michael Crosby
3eae8b9c3f
Merge pull request #2631 from masters-of-cats/shim-io-redirect
Use named pipes for shim logs
2018-11-27 10:44:00 -05:00
Lifubang
e76a8879eb fix pipe in broken may cause shim lock forever for runtime v1
Signed-off-by: Lifubang <lifubang@acmcoder.com>
2018-11-19 09:25:43 +08:00
Julia Nedialkova
1d4105cacf Use named pipes for shim logs
Relating to issue [#2606](https://github.com/containerd/containerd/issues/2606)

Co-authored-by: Oliver Stenbom <ostenbom@pivotal.io>
Co-authored-by: Georgi Sabev <georgethebeatle@gmail.com>
Co-authored-by: Giuseppe Capizzi <gcapizzi@pivotal.io>
Co-authored-by: Danail Branekov <danailster@gmail.com>

Signed-off-by: Oliver Stenbom <ostenbom@pivotal.io>
Signed-off-by: Georgi Sabev <georgethebeatle@gmail.com>
Signed-off-by: Giuseppe Capizzi <gcapizzi@pivotal.io>
Signed-off-by: Danail Branekov <danailster@gmail.com>
2018-11-16 16:11:43 +02:00
Lantao Liu
c524b9ce41 Partially revert the event discard change in #2748.
Signed-off-by: Lantao Liu <lantaol@google.com>
2018-11-08 14:47:32 -08:00
Wei Fu
38d7d59e8a enhance: update v1/v2 runtime
1. avoid dead lock during kill, fetch allProcesses before handle events
2. use argu's ctx instead of context.Backgroud() in openlog

Signed-off-by: Wei Fu <fuweid89@gmail.com>
2018-11-06 22:48:43 +08:00
Michael Crosby
232a063496 Increase reaper buffer size and non-blocking send
Fixes #2709

This increases the buffer size for process exit subscribers. It also
implements a non-blocking send on the subscriber channel.  It is better
to drop an exit even than it is to block a shim for one slow subscriber.

Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2018-10-29 16:46:58 -04:00
Ace-Tang
c206da7957 optimize shim lock in runtime v1
apply lock only around process map of shim service, avoid lock affect
other procs operations.

Signed-off-by: Ace-Tang <aceapril@126.com>
2018-10-25 15:27:46 +08:00
Ace-Tang
079292e3fc fix: modify lock location of exec delete
func (e *execProcess) delete(ctx context.Context) error {
    e.wg.Wait()
...
}
delete exec process will wait for io copy finish, if wait here,
other process can not get lock of shim service.

1. apply lock around s.transition() calls in the Delete methods.
2. put lock after wait io copy in exec Delete.

Signed-off-by: Ace-Tang <aceapril@126.com>
2018-09-11 13:22:59 +08:00
Tom Godkin
b5ccc66c2c Do not kill all on task delete by default
- Still KillAll if the task uses the hosts pid namespace
 - Test for both host pid namespace and normal cases

Co-authored-by: Oliver Stenbom <ostenbom@pivotal.io>
Co-authored-by: Georgi Sabev <georgethebeatle@gmail.com>
Signed-off-by: Oliver Stenbom <ostenbom@pivotal.io>
2018-08-30 15:58:33 +01:00
Michael Crosby
da1b5470cd Runtime v2
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2018-07-17 10:21:29 -04:00