In this patch we fill /proc/PID/uid_map and /proc/PID/gid_map for the
root task.
v2: initialize groups in a new namespace.
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>
v3: add a helper to initialize creds in a new userns
v4: initialize userns creds in prepare_namespaces()
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
For that we need to save per-namespace mappings of user and group IDs.
And all id-s for tasks and files are saved from the target user
namespace.
v2: move code into collect_namespaces()
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We are going to support user namespaces and uid-s will be converted
accoding with userns mappings.
v2: conver id-s for sockets too
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
It's a bad idea to a group of processes and only then check
rights for this operation. We need to check permissions a soon as
posible to reduce impacts in case of wrong permissions.
In addtion criu doesn't to parse /proc/pid/state and gets all required
infromation from /proc/pid/status.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
v2: don't leak FILE
CID 73423 (#1 of 1): Resource leak (RESOURCE_LEAK)
15. leaked_storage: Variable f going out of scope leaks the storage it points to.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We get sig and pgid from a parasite, because we need to get
them from a target pid namespace.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We have two reason for that:
* parsing of /proc/pid/status is slow
* parasite returns ids from a target userns
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Normally, one of the first steps after creating a new user namespace is
to define the mappings used for the user and group IDs of the processes
that will be created in that namespace. This is done by writing mapping
information to the /proc/PID/uid_map and /proc/PID/gid_map files
corresponding to one of the processes in the user namespace.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
When we create a new mntns in a userns, all inhereted mounts are marked
as locked. pivot_root() returns EINVAL if a new root is locked.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
page-server are not tested now. I suggest to add it in the pre-dump job.
v2: execute all tests with and without page server
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
All out processes are stopped in a moment, when file locks are
collected, so they can't to wait any locks.
Here is a proof of this theory:
[root@avagin-fc19-cr ~]# flock xxx sleep 1000 &
[1] 23278
[root@avagin-fc19-cr ~]# flock xxx sleep 1000 &
[2] 23280
[root@avagin-fc19-cr ~]# cat /proc/locks
1: FLOCK ADVISORY WRITE 23278 08:03:280001 0 EOF
1: -> FLOCK ADVISORY WRITE 23280 08:03:280001 0 EOF
[root@avagin-fc19-cr ~]# gdb -p 23280
(gdb) ^Z
[3]+ Stopped gdb -p 23280
[root@avagin-fc19-cr ~]# cat /proc/locks
1: FLOCK ADVISORY WRITE 23278 08:03:280001 0 EOF
Currently criu can dump nothing, if we have one process which is
waiting a lock. I don't see any reason to do this.
v2: typo fix
Cc: Qiang Huang <h.huangqiang@huawei.com>
Reported-by: Mr Jenkins
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Fixes: e47eccb188 ("page-server: don't forget to close a sever socket")
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CID 73370: Resource leak (RESOURCE_LEAK)
13. leaked_storage: Variable timer going out of scope leaks the storage it points to.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CID 73377: Resource leak (RESOURCE_LEAK)
10. leaked_storage: Variable values going out of scope leaks the storage
it points to.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CID 73347 (#7-4 of 7): Resource leak (RESOURCE_LEAK)
21. leaked_handle: Handle variable sk going out of scope leaks the handle.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
I think this version of code is a bit more readable.
It doesn't do memcpy and doesn't allocate FILE.
Everyone knows arguments for read(), but only a few of
us know arguments for fread().
CID 73345 (#1 of 1): String not null terminated (STRING_NULL)
2. string_null_argument: Function fread does not terminate string *buf. [Note: The source code implementation of the function has been overridden by a builtin model.]
Cc: Tycho Andersen <tycho.andersen@canonical.com>
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Acked-by: Tycho Andersen <tycho.andersen@canonical.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
list_for_each_entry_safe() should be used, you we are going to delete
something from a list.
CID 73383 (#1 of 1): Read from pointer after free (USE_AFTER_FREE)
4. deref_after_free: Dereferencing freed pointer prop.
Cc: Tycho Andersen <tycho.andersen@canonical.com>
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Acked-by: Tycho Andersen <tycho.andersen@canonical.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
This function doesn't allocate memory, so it should not release it.
CID 73380 (#1 of 1): Use after free (USE_AFTER_FREE)
14. deref_arg: Calling munmap dereferences freed pointer addr.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CID 73348 (#1 of 1): Resource leak (RESOURCE_LEAK)
19. leaked_handle: Handle variable sk going out of scope leaks the handle.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
When a process, which starts a service, exits, we know that we can
connect to the page server and we expect to find pidfile.
If we create pidfile from a page server process, we have a race window.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CID 73381 (#1 of 1): Out-of-bounds write (OVERRUN)
15. overrun-local: Overrunning array loc_buf of 4096 bytes at byte offset 4096 using index len (which evaluates to 4096).
CID 73355 (#1 of 1): Out-of-bounds write (OVERRUN)
6. overrun-local: Overrunning array loc_buf of 4096 bytes at byte offset 4096 using index ret (which evaluates to 4096)
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Introduced by eb214be2, the empty mnt_share list cannot
produce the list_first_entry element :)
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Andrew Vagin <avagin@parallels.com>
Some kernel modules such as pktgen runs kthred upon
new-net creation taking last_pid we were requested.
Lets workaround this problem using clone + unshare
bundle.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Acked-by: Andrew Vagin <avagin@parallels.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
| tty.c: In function ‘tty_restore_ctl_terminal’:
| include/criu-log.h:42:16: error: ‘index’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
Same time assign CONSOLE_INDEX for debug info.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
and return an error, if a proccess live in another userns,
because criu doesn't support it.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
CRIU reads /proc/pid/ns/[NS] and fails of a link is not exist.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
When creating helpers, we need to make sure their
PIDs do not intersect with threads' IDs.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: Andrew Vagin <avagin@parallels.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
This is for two reasons. First, validation can meet external mount
and will call plugins, which is not correct on pre-dump and actually
crashes on uninitilized plugins lists. Second, even if on pre-dump
mount tree is not "supported" this can be a temporary situation (yes,
yes, unlikely, but still).
On the other hand, it's better to fail earlier, but that's another
story.
Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Andrew Vagin <avagin@parallels.com>
The pty field here is only present with certain tty types, so we shouldn't
unconditionally dereference it, or criu will hang.
Signed-off-by: Tycho Andersen <tycho.andersen@canonical.com>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Now we check them for all test processes
v2: Busybox doesn't have truncate
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Reviewed-by: Christopher Covington <cov@codeaurora.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
In case if we meet virtualized devtmpfs on dump
(which means its s_dev is different from one obtained
during mountpoints dump procedure) we should dump it
with tar help. Thus on restore it get filled from the
image.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We will need devtmpfs as well so make it general.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
So we need to create a temporary private mount for the old root.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
This reverts commit 21d1b2fdb9.
pivot_root() can move the current root to a non-shared mount. So we are
going to create a temporary private mount in put_old.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We have two global enums with the member TUN. The first one is in fdinfo.proto and the
second one is in netdev.proto. protoc complains about having TUN in 2 global enums because
of the enum rules for C++(even though we're compiling .proto files for python).
So lets rename TUN into TUNF to avoid that conflict.
Signed-off-by: Ruslan Kuprieiev <kupruser@gmail.com>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
We simply check that major/minor didn't changed after restore.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
