Hello Arnaldo,
I found, that you commit patches for perf. Could you review this part?
Arun Sharma said, that the second versions of patches works ok for him.
(Arun is the first user of this functionality after me.)
Thanks.
On 11/28/2011 01:03 PM, Andrew Vagin wrote:
Do you want to know where
Don't commit these patch, just look at them and send me comments.
This patches have not been tested on OpneVZ CT-s:)
Andrey Vagin (8):
vzctl: save a pid of init in a state file
vzctl: split ct_env_create
vzctl: transfer vps_res in env_stop
vzctl: save PID of init in a state file
CRIU requires a pid of the init.
Signed-off-by: Andrey Vagin
---
etc/vz.conf.in| 1 +
include/res.h | 1 +
include/types.h | 2 ++
include/util.h| 1 +
include/vzctl_param.h | 1 +
src/lib/config.c | 10 ++
src/lib/util.c| 6 ++
7
ct_env_create will able to execute CT and to resume CT
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 45 +++--
1 file changed, 27 insertions(+), 18 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index 6a5e766..184c26a 100644
It will be used to unlink a state file
Signed-off-by: Andrey Vagin
---
include/types.h| 3 ++-
src/lib/env.c | 18 +-
src/lib/hooks_ct.c | 4 ++--
src/lib/hooks_vz.c | 2 +-
4 files changed, 14 insertions(+), 13 deletions(-)
diff --git a/include/types.h b/include
Signed-off-by: Andrey Vagin
---
include/cpt.h | 2 +-
include/env.h | 2 +-
include/snapshot.h | 4 ++--
src/lib/cpt.c | 6 +++---
src/lib/hooks_vz.c | 2 +-
src/snapshot.c | 13 -
src/vzctl-actions.c | 6 +++---
7 files changed, 19 insertions(+), 16
It's a preparation for moving vz-specific code in hook_vz.c.
Signed-off-by: Andrey Vagin
---
src/lib/cpt.c | 43 +--
1 file changed, 33 insertions(+), 10 deletions(-)
diff --git a/src/lib/cpt.c b/src/lib/cpt.c
index 5501d92..24ba9a8 100644
--- a/sr
It will be used for checkpointing this CT
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 20 ++--
1 file changed, 18 insertions(+), 2 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index 0d4e439..3c82823 100644
--- a/src/lib/hooks_ct.c
+++ b/src/lib
Signed-off-by: Andrey Vagin
---
include/cpt.h | 13 ++
include/types.h| 24 ++--
src/lib/cpt.c | 403 +
src/lib/hooks_vz.c | 385 ++
4 files changed, 415 insertions(+), 410
It can be done by any third party tools.
sripts/ct-cpt and script/ct_rst are executed to dump and restore CT
This patch adds scripts for CRIU.
Signed-off-by: Andrey Vagin
---
scripts/ct-cpt | 19
scripts/ct-rst | 17 ++
src/lib/hooks_ct.c | 91
The project CRIU is used for dumping and restoring containers.
http://criu.org/
Known issues:
Network devices are not supported yet.
v2: The patch set was fixed according with Kir's comments.
Andrey Vagin (6):
vzctl: split ct_env_create
vzctl: save PID of init in a state
ct_env_create will able to execute CT and to resume CT
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 49 ++---
1 file changed, 30 insertions(+), 19 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index 6a5e766..83ef20a 100644
It's a preparation for moving vz-specific code in hook_vz.c.
Signed-off-by: Andrey Vagin
---
src/lib/cpt.c | 43 +--
1 file changed, 33 insertions(+), 10 deletions(-)
diff --git a/src/lib/cpt.c b/src/lib/cpt.c
index 362c646..e4213a2 100644
--- a/sr
Signed-off-by: Andrey Vagin
---
include/cpt.h | 13 ++
include/types.h| 25 ++--
src/lib/cpt.c | 403 +
src/lib/hooks_vz.c | 385 ++
4 files changed, 416 insertions(+), 410
CRIU requires a pid of the init.
Signed-off-by: Andrey Vagin
---
include/types.h| 2 ++
include/util.h | 1 +
src/lib/hooks_ct.c | 20 ++--
src/lib/util.c | 5 +
4 files changed, 26 insertions(+), 2 deletions(-)
diff --git a/include/types.h b/include/types.h
Signed-off-by: Andrey Vagin
---
include/env.h | 4 ++--
src/lib/hooks_vz.c | 6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/env.h b/include/env.h
index 1628bbf..dd5c2f8 100644
--- a/include/env.h
+++ b/include/env.h
@@ -34,8 +34,8 @@
-typedef int
It can be done by any third party tools.
sripts/ct-cpt and script/ct_rst are executed to dump and restore CT
This patch adds scripts for CRIU.
v2: add the GPL header and descriptions in scripts
describe and check variables in scripts
Signed-off-by: Andrey Vagin
---
scripts/Makefile.am
The project CRIU is used for dumping and restoring containers.
http://criu.org/
Known issues:
Network devices are not supported yet.
v2: The patch set was fixed according with Kir's comments.
v3: env_create_FN returns PID or a negative error code
Andrey Vagin (6):
vzctl:
ct_env_create will able to execute CT and to resume CT
v2: A return code contains one of known codes with inverted sign.
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 79 +++---
1 file changed, 45 insertions(+), 34 deletions(-)
diff --git
CRIU requires a pid of the init.
Signed-off-by: Andrey Vagin
---
include/types.h| 2 ++
include/util.h | 1 +
src/lib/hooks_ct.c | 18 +-
src/lib/util.c | 5 +
4 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/include/types.h b/include/types.h
It's a preparation for moving vz-specific code in hook_vz.c.
Signed-off-by: Andrey Vagin
---
src/lib/cpt.c | 43 +--
1 file changed, 33 insertions(+), 10 deletions(-)
diff --git a/src/lib/cpt.c b/src/lib/cpt.c
index 626c055..cca15b6 100644
--- a/sr
Signed-off-by: Andrey Vagin
---
include/cpt.h | 13 ++
include/types.h| 25 ++--
src/lib/cpt.c | 404 +
src/lib/hooks_vz.c | 386 ++
4 files changed, 417 insertions(+), 411
Signed-off-by: Andrey Vagin
---
include/env.h | 4 ++--
src/lib/hooks_vz.c | 6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/env.h b/include/env.h
index 4fef438..7426d63 100644
--- a/include/env.h
+++ b/include/env.h
@@ -34,8 +34,8 @@
-typedef int
It can be done by any third party tools.
sripts/ct-cpt and script/ct_rst are executed to dump and restore CT
This patch adds scripts for CRIU.
v2: add the GPL header and descriptions in scripts
describe and check variables in scripts
Signed-off-by: Andrey Vagin
---
scripts/Makefile.am
Andrey Vagin (3):
vzctl: pass vps_res in env_create_FN
vzctl: add ability to skip creation of veth devices
cpt: restore veth devices with correct names
include/env.h| 2 +-
include/types.h | 2 ++
scripts/vps-netns_dev_add.in | 5 -
scripts/vps-rst.in
Signed-off-by: Andrey Vagin
---
include/env.h | 2 +-
src/lib/hooks_ct.c | 4 ++--
src/lib/hooks_vz.c | 4 ++--
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/env.h b/include/env.h
index 7426d63..67ab3ef 100644
--- a/include/env.h
+++ b/include/env.h
@@ -34,7 +34,7
It will be used for resuming CT with help CRIU.
CRIU restores veth devices and configures them inside CT,
so vzctl should configures them on the host side.
Signed-off-by: Andrey Vagin
---
include/types.h | 2 ++
scripts/vps-netns_dev_add.in | 5 -
src/lib/hooks_ct.c
transmit pair of veth names to criu via the option --veth-pair
Signed-off-by: Andrey Vagin
---
scripts/vps-rst.in | 14 --
src/lib/hooks_ct.c | 18 ++
2 files changed, 26 insertions(+), 6 deletions(-)
diff --git a/scripts/vps-rst.in b/scripts/vps-rst.in
index
transmit pair of veth names to criu via the option --veth-pair
v2: unset IFS and delete eval from vps-rst
Signed-off-by: Andrey Vagin
---
scripts/vps-rst.in | 15 +--
src/lib/hooks_ct.c | 18 ++
2 files changed, 27 insertions(+), 6 deletions(-)
diff --git a/scripts
It will be used for resuming CT with help CRIU.
CRIU restores veth devices and configures them inside CT,
so vzctl should configures them on the host side.
Signed-off-by: Andrey Vagin
---
include/types.h | 2 ++
scripts/vps-netns_dev_add.in | 5 -
src/lib/hooks_ct.c
transmit pair of veth names to criu via the option --veth-pair
v2: unset IFS and delete eval from vps-rst
v3: fix comments from Kir
Signed-off-by: Andrey Vagin
---
scripts/vps-rst.in | 12 ++--
src/lib/hooks_ct.c | 16 +---
2 files changed, 23 insertions(+), 5 deletions
transmit pair of veth names to criu via the option --veth-pair
v2: unset IFS and delete eval from vps-rst
v3: fix comments from Kir
v4: vzcheckvar VE_VETH_DEVS
Signed-off-by: Andrey Vagin
---
scripts/vps-rst.in | 11 ++-
src/lib/hooks_ct.c | 16 +---
2 files changed, 23
nly a tools, which checkpoints CT, knows how to delete own data,
so lets do that in the script vzt-rst.
Signed-off-by: Andrey Vagin
---
include/cpt.h | 3 ++-
scripts/vps-rst.in | 11 ++-
src/lib/env.c | 10 +-
src/lib/hooks_ct.c | 31 +--
sr
vzctl provides two descriptors signalfd and waitfd, it's used for apply
host-side configuration after creating environment.
A signal is send to signalfd after creating environment and an answer is
recieved from waitfd.
Signed-off-by: Andrey Vagin
---
scripts/Makefile.am| 3 ++-
script
The most part of vzctl functions return positive error code.
This two functions were exceptions, because they return PID of init
to create netns file. Currently each of these functions creates
netns file by itself.
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 40
vzctl provides two descriptors signalfd and waitfd, it's used for apply
host-side configuration after creating environment.
A signal is send to signalfd after creating environment and an answer is
recieved from waitfd.
v2: fix comments from Kir
Signed-off-by: Andrey Vagin
---
sc
-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index 4972dac..aff9cee 100644
--- a/src/lib/hooks_ct.c
+++ b/src/lib/hooks_ct.c
@@ -103,6 +103,25 @@ int ct_chroot(const char *root
and /sys/fs/cgroup/ that exist explicitly
for other filesystems to mount on top of.
Signed-off-by: Andrey Vagin
---
src/lib/env.c | 9 -
src/lib/hooks_ct.c | 26 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git a/src/lib/env.c b/src/lib
User namespaces was restricted in the mainstream kernel and
vzctl must follow to the current rules.
ps: I didn't test these patches on OpenVZ kernels.
Andrey Vagin (2):
vzctl: bind-mount the root in itself after entering in a user
namespace
hooks_ct: mount /proc and /sys before umou
and /sys/fs/cgroup/ that exist explicitly
for other filesystems to mount on top of.
Signed-off-by: Andrey Vagin
---
src/lib/env.c | 9 -
src/lib/hooks_ct.c | 26 ++
2 files changed, 34 insertions(+), 1 deletion(-)
diff --git a/src/lib/env.c b/src/lib
this configuration doesn't
supported by CRIU.
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 48 ++--
1 file changed, 2 insertions(+), 46 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index a1b91d9..b0a3e7b 100644
---
this configuration doesn't
supported by CRIU.
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 48 ++--
1 file changed, 2 insertions(+), 46 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
index 5ec4a93..2a0b54c 100644
---
All modern distributions require devtmpfs in /dev. devtmpfs can't
be mounted from userns. This patch bind-mounts the host /dev.
It's secure, because permissions are handled according with uid and
gid maps for the user namespace.
Signed-off-by: Andrey Vagin
---
src/lib/hooks
t required any more.
Signed-off-by: Andrey Vagin
---
etc/dists/scripts/prestart.sh | 4 ---
src/lib/hooks_ct.c| 66 +++
2 files changed, 16 insertions(+), 54 deletions(-)
diff --git a/etc/dists/scripts/prestart.sh b/etc/dists/scripts/prestar
This script must be started before the init will be executed,
so it must not be executed in background.
Probably the author want to write && instead of &.
I don't think, that we need to check errors here,
because the next script is independent from the previous one.
Signed-off
for that we must remount root with MS_PRIVATE before doing anything else.
https://bugzilla.openvz.org/show_bug.cgi?id=2879
Signed-off-by: Andrey Vagin
---
src/lib/hooks_ct.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/src/lib/hooks_ct.c b/src/lib/hooks_ct.c
ith the patch applied I can checkpoint/restore container
> with venet configured.
Is it posibale to unload the venet module? Is it ok with this patch?
>
> Signed-off-by: Cyrill Gorcunov
> CC: Vladimir Davydov
> CC: Konstantin Khorenko
> CC: Pavel Emelyanov
> CC: Andrey Vagi
ith the patch applied I can checkpoint/restore container
> with venet configured.
>
Acked-by: Andrey Vagin
> Signed-off-by: Cyrill Gorcunov
> CC: Vladimir Davydov
> CC: Konstantin Khorenko
> CC: Pavel Emelyanov
> CC: Andrey Vagi
Now CT starts in a new user namespace. This allows us
* to remove our capabilities (CAP_VE_*)
* to improve security of our containers, because a process doesn't have
privileges outside the container
Here is a good article about user namespaces https://lwn.net/Articles/532593/
https://jira.sw.ru
On Tue, Feb 09, 2016 at 01:02:09PM +0300, Vladimir Davydov wrote:
> On Mon, Feb 08, 2016 at 11:32:12AM +0300, Cyrill Gorcunov wrote:
> > +static int ve_ts_read(struct cgroup *cg, struct cftype *cft, struct
> > seq_file *m)
> > +{
> > + struct ve_struct *ve = cgroup_ve(cg);
> > + struct timespe
On Tue, Apr 26, 2016 at 12:36:12PM +0300, Cyrill Gorcunov wrote:
> On Mon, Apr 25, 2016 at 06:22:10PM +0300, Evgenii Shatokhin wrote:
> > https://jira.sw.ru/browse/PSBM-46199
> >
> > Allowing the privileged processes in the containers to set leases on
> > arbitrary files seems to make no harm. Let
Acked-by: Andrey Vagin
On Mon, May 16, 2016 at 11:28:51AM +0300, Cyrill Gorcunov wrote:
> This is a backport of commit
>
> ML: bdb4d100afe9818aebd1d98ced575c5ef143456c
>
> From: Calvin Owens
>
> Currently, /proc//map_files/ is restricted to CAP_SYS_ADMIN, and
> reasons as well.
>
> --
> Best regards,
>
> Konstantin Khorenko,
> Virtuozzo Linux Kernel Team
>
> On 05/16/2016 10:42 PM, Andrey Vagin wrote:
> > Acked-by: Andrey Vagin
> >
> > On Mon, May 16, 2016 at 11:28:51AM +0300, Cyrill Gorcunov wrote:
>
On Mon, Jun 06, 2016 at 03:45:11PM +0300, Vasily Averin wrote:
> In mainline rootfs is marked always as MNT_LOCKED,
> sys_umount checks this flag and fails its processing.
> Our kernels lacks for MNT_LOCKED flag, so we use another kind of check
> to prevent incorrect operation.
>
> https://jira.sw
>
> v2: use mnt_has_parent()
>
> https://jira.sw.ru/browse/PSBM-46437
>
Acked-by: Andrey Vagin
> Signed-off-by: Vasily Averin
> diff --git a/fs/namespace.c b/fs/namespace.c
> index 988320b..4fb935a 100644
> --- a/fs/namespace.c
> +++ b/fs/namespace.c
> @@ -1355,6 +
This allows us to read socket's queue without removing skbs from it.
The same logic was implemented for unix and inet sockets and we use this
to dump and restore sockets in CRIU.
Here is a question whether sk_peek_off has to be protected by locks.
Currently it isn't protected and an user who uses
ke we do for tcp sockets.
* When we can dump data from a receive queue.
I think we can do this only if a socket doesn't have a running callback.
Andrey Vagin (3):
netlink: allow to set peeking offset for sockets
netlink: add an ability to restore messages in a receive queue
netlink/di
This patch adds an repair mode for netlink sockets. sendmsg queues
messages into a receive queue if a socket is in the repair mode.
---
include/uapi/linux/netlink.h | 19 ++---
net/netlink/af_netlink.c | 51 +++-
2 files changed, 47 insertion
We need to know flags for dumping and restoring netlink sockets.
All flags except NDIAG_FLAG_CB_RUNNING can be received with help of
getsockopt(), but in this case we need a socket descriptor and we need
to call getsockopt() to get each flag.
With this chages we will be able to show netlink socke
easons why we need to restore snd_wnd was described above.
Cc: Pavel Emelyanov
Cc: "David S. Miller"
Cc: Alexey Kuznetsov
Cc: James Morris
Cc: Hideaki YOSHIFUJI
Cc: Patrick McHardy
Signed-off-by: Andrey Vagin
Signed-off-by: David S. Miller
---
include/uapi/linux/tcp.h | 10 +
From: Tycho Andersen
ML: f8e529ed941ba2bbcbf310b575d968159ce7e895
https://jira.sw.ru/browse/PSBM-55322
This patch adds support for dumping a process' (classic BPF) seccomp
filters via ptrace.
PTRACE_SECCOMP_GET_FILTER allows the tracer to dump the user's classic BPF
seccomp filters. addr shoul
From: Andrei Vagin
Signed-off-by: Andrei Vagin
---
criu/mount.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/criu/mount.c b/criu/mount.c
index fcec92f..76d1505 100644
--- a/criu/mount.c
+++ b/criu/mount.c
@@ -2493,7 +2493,8 @@ static int propagate_mount(struct mount_inf
From: Andrei Vagin
A widest mount of a shared group has to be mounted first and
then other mounts of this group has to be mounted from the first mount
to inherite the shared group.
https://jira.sw.ru/browse/PSBM-55494
Signed-off-by: Andrei Vagin
---
criu/mount.c | 13 -
1 file chan
From: Andrei Vagin
pivot_root requires a place where to move an old root. Currently
a temporary directory is created for that, but it doesn't
work if the / directory is read-only.
Actually we can use any existing directory. In this patch,
criu tries to use /tmp and only if it doesn't exist,
criu
On Tue, Dec 06, 2016 at 05:59:30PM +0300, Cyrill Gorcunov wrote:
> In commit 42b65fd18057d64410a0519962cd0650c762c99f there
> is a typo: we need to copy the complete filter chain
> not only first number.
>
> https://jira.sw.ru/browse/PSBM-55593
>
Acked-by: Andrey Vagin
ht now,
> lets rather do a trick for seccomp sake and simply carry a copy inside
> struct seccomp_filter. The socket filters are decoded into userspace form
> anyway so this area is safe.
>
> https://jira.sw.ru/browse/PSBM-55593
>
> CC: Andrey Vagin
> Signed-off-by: Cyrill
From: Andrei Vagin
If we can't create a temporary directory for a detached mount,
we can clone a whole mount namespace, open a mount and release
the created namespace. The result will be the same.
https://jira.sw.ru/browse/PSBM-57135
Signed-off-by: Andrei Vagin
---
criu/mount.c | 18 ++
From: Andrei Vagin
Only sockets in the repair mode can be bound to the same
port a few time, what is required to restore tcp connects.
Signed-off-by: Andrei Vagin
---
criu/sk-inet.c | 10 +-
soccr/soccr.c | 10 ++
2 files changed, 15 insertions(+), 5 deletions(-)
diff --git a
ubsequent recvmsg
> calls. Instead lets exit early with error.
>
> In sake of https://jira.sw.ru/browse/PSBM-57921
>
> CC: Andrey Vagin
> Signed-off-by: Cyrill Gorcunov
> ---
> net/netlink/af_netlink.c |4
> 1 file changed, 4 insertions(+)
>
&
On Thu, Dec 22, 2016 at 03:34:15PM -0800, Andrey Vagin wrote:
> On Thu, Dec 22, 2016 at 06:41:42PM +0300, Cyrill Gorcunov wrote:
> > When skb_copy_datagram_iovec called to fetch queued data
> > it may fail with EFAULT and if MSG_PEEK set by a caller
> > the position get
On Thu, Dec 22, 2016 at 04:12:13PM -0800, Andrey Vagin wrote:
> On Thu, Dec 22, 2016 at 03:34:15PM -0800, Andrey Vagin wrote:
> > On Thu, Dec 22, 2016 at 06:41:42PM +0300, Cyrill Gorcunov wrote:
> > > When skb_copy_datagram_iovec called to fetch queued data
> > > it
On Wed, Feb 08, 2017 at 12:05:37AM +0300, Cyrill Gorcunov wrote:
> The main idea is to be able to operate with container
> at the moment where its processess and resources are
> already restored but the processes are not yet in
> running state, ie just before we kick them.
>
> Beside the need of t
On Thu, Feb 09, 2017 at 12:01:20AM +0300, Cyrill Gorcunov wrote:
> On Wed, Feb 08, 2017 at 12:37:18PM -0800, Andrey Vagin wrote:
> > On Wed, Feb 08, 2017 at 12:05:37AM +0300, Cyrill Gorcunov wrote:
> > > The main idea is to be able to operate with container
> > > at the
From: Andrei Vagin
Send a netlink message with ucred and check that it received
with the same ucred.
Signed-off-by: Andrei Vagin
---
test/zdtm/static/sk-netlink.c | 55 ++-
1 file changed, 54 insertions(+), 1 deletion(-)
diff --git a/test/zdtm/static/sk
From: Andrei Vagin
ucred in a netlink message contains the same pid and
it doesn't metter from which pidns it is read.
https://jira.sw.ru/browse/PSBM-59339
Cc: Kirill Tkhai
Signed-off-by: Andrei Vagin
---
criu/include/sk-queue.h | 5 ++-
criu/sk-netlink.c | 2 +-
criu/sk-queue.c
From: Andrei Vagin
We need to know a type of unsupported control messages.
https://jira.sw.ru/browse/PSBM-60643
Signed-off-by: Andrei Vagin
---
criu/sk-queue.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/criu/sk-queue.c b/criu/sk-queue.c
index be33ca1..f68d26f 100644
--- a/criu/sk-qu
From: Andrei Vagin
https://jira.sw.ru/browse/PSBM-60643
Andrei Vagin (5):
netlink: restore queues and only then restore opts
netlink: dump/restore netlink options
socket: don't restore ucred from kernel messages
netlink: skip pktinfo scm if a sender group is less than 32
netlink: avoid
From: Andrei Vagin
There are a few netlink specific opotions, which have to be dumped.
Signed-off-by: Andrei Vagin
---
criu/sk-netlink.c | 60 +
images/sk-netlink.proto | 9
2 files changed, 69 insertions(+)
diff --git a/criu/sk-
From: Andrei Vagin
nl_pktinfo control messages is used to get the extended
destination group number.
Signed-off-by: Andrei Vagin
---
criu/sk-queue.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/criu/sk-queue.c b/criu/sk-queue.c
index 88e47b8..8124f75 100644
--- a/criu/sk
From: Andrei Vagin
ucred.pid is zero for kernel messages and in this case it is
impossible to restore it.
Signed-off-by: Andrei Vagin
---
criu/sk-queue.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/criu/sk-queue.c b/criu/sk-queue.c
index e72ca45..88e47b8 100644
--
From: Andrei Vagin
When we restore queues, we set limits to unlimited values,
so then we need to restore required values for them.
Signed-off-by: Andrei Vagin
---
criu/sk-netlink.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/criu/sk-netlink.c b/criu/sk-netlink.c
ind
From: Andrei Vagin
If a socket buffer is full and some messaged have been
dropped, the first recvmsg will return -ENOBUF.
We can avoid this if we set the NETLINK_NO_ENOBUFS opt for a socket.
Signed-off-by: Andrei Vagin
---
criu/sk-netlink.c | 19 ++-
1 file changed, 18 inserti
Acked-by: Andrey Vagin
On Fri, Dec 23, 2016 at 10:19:01AM +0300, Cyrill Gorcunov wrote:
> On Fri, Dec 23, 2016 at 09:59:28AM +0300, Cyrill Gorcunov wrote:
> > On Thu, Dec 22, 2016 at 04:45:10PM -0800, Andrey Vagin wrote:
> > >
> > > Actually, this patch breaks
On Mon, Mar 27, 2017 at 03:39:40PM +0300, Konstantin Khorenko wrote:
> Andrey, are you going to send your patch to mainstream?
> (Kirill's patch is based on your patch, AFAIK)
>
> commit 523b88d0b87acbc7af9ae1677e993ed20f55dc5e
> Author: Andrey Vagin
> Date: Fri Jun
From: Andrei Vagin
run_scripts can execute notifications via rpc, so WEXITSTATUS()
can't be applied to return codes in this case.
And an action script can be killed and it is an error too.
This patch may fix on of errors of
https://jira.sw.ru/browse/PSBM-62583
Signed-off-by: Andrei Vagin
---
On Wed, Jul 19, 2017 at 11:31:28AM -0700, Stanislav Kinsburskiy wrote:
>
>
> 19 июля 2017 г. 9:14 PM пользователь Andrey Vagin
> написал:
>
> On Wed, Jul 19, 2017 at 08:04:22PM +0300, Andrey Ryabinin wrote:
> >
> >
> > On 07/19/2017 04:14 AM
On Mon, Jul 24, 2017 at 07:12:50PM +0300, Cyrill Gorcunov wrote:
> On Mon, Jul 24, 2017 at 07:09:39PM +0300, Stanislav Kinsburskiy wrote:
> >
> >Well, no. I don't like this defines.
> >And I don't think that pid has to be explicitly provided. At least for
> >ve0.
> >But frankly, I
On Mon, Jul 24, 2017 at 11:37:52PM -0700, Stanislav Kinsburskiy wrote:
>
>
> 24 июля 2017 г. 7:12 PM пользователь Kirill Gorkunov
> написал:
>
> On Mon, Jul 24, 2017 at 07:09:39PM +0300, Stanislav Kinsburskiy wrote:
> >
> >Well, no. I don't like this defines.
> >And I do
PSBM-68062
>
> v2:
> - make sure the ids is inherited if zombie met
>
Acked-by: Andrey Vagin
> Signed-off-by: Cyrill Gorcunov
> ---
> criu/mount.c | 11 +--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/criu/mount.c b/criu/mount.c
>
On Sat, Jul 22, 2017 at 01:27:08PM +0300, Cyrill Gorcunov wrote:
> In case if our parent is a dead task (zombie) we should lookup
> for parent ids which will be inherited on restore. Otherwise
> parent->ids may be nil and SIGSEGV produced.
>
> https://jira.sw.ru/browse/PSBM-68062
>
> Signed-off-b
Applied, thanks!
On Tue, Jul 18, 2017 at 10:57:54PM +0300, Cyrill Gorcunov wrote:
> We will need them to handle tty inheritance.
>
> Signed-off-by: Cyrill Gorcunov
> ---
> images/core.proto | 4
> 1 file changed, 4 insertions(+)
>
> In sake of https://jira.sw.ru/browse/PSBM-67671
>
> diff
Do we need this patch for the upstream criu?
On Tue, Sep 12, 2017 at 05:49:43PM +0300, Cyrill Gorcunov wrote:
> On Tue, Sep 12, 2017 at 5:43 PM, Cyrill Gorcunov wrote:
> > On Sun, Jul 23, 2017 at 6:15 PM, Cyrill Gorcunov wrote:
> >> We are carrying uids/gids in per-thread way since v1.8,
> >> so
Reviewed-by: Andrey Vagin
On Wed, Oct 18, 2017 at 01:01:40PM +0300, Cyrill Gorcunov wrote:
> Andrew, Kirill, take a look please. I think we should use %lu
> for all images names instead of %d, but this will require more
> code changes which I'm planning to do for vanilla criu firs
ks,
Andrei
>
> https://jira.sw.ru/browse/PSBM-76490
>
> CC: Andrey Vagin
> CC: Andrey Ryabinin
> CC: Konstantin Khorenko
> CC: "Denis V. Lunev"
> Signed-off-by: Cyrill Gorcunov
> ---
> Guys, take a look please. Note the patch on its own is useless
>
94 matches
Mail list logo
