namespaces: log namespaces per task

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

Quoted 27 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Quoted 5 lines

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

I'm undecided if I'd rather see this as a separate NS_INFO record type. It would mean we could filter them out of the logs...

Do we print out lots of pidns=0 for tasks not in a newly created NS? Do we want to?

Quoted 6 lines

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

I'd love to be able to grep for netns=20 and find both the NS_INIT and the SYSCALL/NS_INFO records, instead of having them named different things. So basically I think you want to translate the type= into a string for the old_X= and X=...

Quoted 3 lines · Eric Paris

Quoted 2 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

Quoted 10 lines

Quoted 9 lines

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

Quoted 2 lines

I'm undecided if I'd rather see this as a separate NS_INFO record type. It would mean we could filter them out of the logs...

I don't have a strong opinion either way. Steve G.'s opinion would be helpful here.

Quoted 2 lines

Do we print out lots of pidns=0 for tasks not in a newly created NS? Do we want to?

There is no "pidns=0", but I understand your point. This would come back to Steve G.'s point about disappearing fields, and the value of having it as a seperate record that could be filtered.

Quoted 7 lines

Quoted 6 lines

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

Quoted 4 lines

I'd love to be able to grep for netns=20 and find both the NS_INIT and the SYSCALL/NS_INFO records, instead of having them named different things. So basically I think you want to translate the type= into a string for the old_X= and X=...

That actually makes a bit more sense, and we could do away with the "type=" field since the "Xns=" fields are self-describing.

- RGB

Quoted 4 lines · Richard Guy Briggs

Quoted 3 lines · Eric Paris

Quoted 2 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

(Including the Linux API list due to the additions to /proc/<pid>/ns/. Please see http://www.kernelhub.org/?p=2&msg=477668 and in particular
http://www.kernelhub.org/?msg=477678&p=2 )

<snip>

Quoted 10 lines

Quoted 10 lines

Quoted 9 lines

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

Quoted 5 lines

Quoted 2 lines

I'm undecided if I'd rather see this as a separate NS_INFO record type. It would mean we could filter them out of the logs...

I don't have a strong opinion either way. Steve G.'s opinion would be helpful here.

Breaking this out into a seperate record type would mean calling audit_log_namespace_info() from the callers of audit_log_task_info() (presently audit_log_link_denied(), ima_audit_measurement(), audit_log_exit() ), but would make it easier to emit with other record types.

Quoted 14 lines

Quoted 2 lines

Do we print out lots of pidns=0 for tasks not in a newly created NS? Do we want to?

There is no "pidns=0", but I understand your point. This would come back to Steve G.'s point about disappearing fields, and the value of having it as a seperate record that could be filtered.

Quoted 7 lines

Quoted 6 lines

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

Quoted 7 lines

Quoted 4 lines

I'd love to be able to grep for netns=20 and find both the NS_INIT and the SYSCALL/NS_INFO records, instead of having them named different things. So basically I think you want to translate the type= into a string for the old_X= and X=...

That actually makes a bit more sense, and we could do away with the "type=" field since the "Xns=" fields are self-describing.

Steve, how do you feel about this since the NS_INIT/NS_DEL record type would have changing fields amongst the 6 namespace types. Only one of the 6 would be present in each message. I suppose we could have an NS_INIT_XXX/NS_DEL_XXX for each namespace type.

Quoted 1 line

- RGB

- RGB

--
Richard Guy Briggs <[email protected]> Senior Software Engineer, Kernel Security, AMER ENG Base Operating Systems, Red Hat Remote, Ottawa, Canada
Voice: +1.647.777.2635, Internal: (81) 32635, Alt: +1.613.693.0684x3545

Quoted 9 lines · Richard Guy Briggs

Quoted 4 lines · Richard Guy Briggs

Quoted 3 lines · Eric Paris

Quoted 2 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

(Including the Linux API list due to the additions to /proc/<pid>/ns/. Please see http://www.kernelhub.org/?p=2&msg=477668 and in particular
http://www.kernelhub.org/?msg=477678&p=2 )

Sigh if you have to use something like this use the proc inode number. It is the same thing.

I hate to claim it is unique absent of the proc superblock but it is and will be for the forseable future.

It would be better to include the block device number that appears in proc of 3h of the primary mount of to qualify the number. But it is not particularly important. Coming up with an additional unique number that needs to be maintained seems stronlgy silly.

Eric

Quoted 60 lines

<snip>

Quoted 10 lines

Quoted 10 lines

Quoted 9 lines

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

Quoted 5 lines

Quoted 2 lines

I'm undecided if I'd rather see this as a separate NS_INFO record type. It would mean we could filter them out of the logs...

I don't have a strong opinion either way. Steve G.'s opinion would be helpful here.

Breaking this out into a seperate record type would mean calling audit_log_namespace_info() from the callers of audit_log_task_info() (presently audit_log_link_denied(), ima_audit_measurement(), audit_log_exit() ), but would make it easier to emit with other record types.

Quoted 14 lines

Quoted 2 lines

Do we print out lots of pidns=0 for tasks not in a newly created NS? Do we want to?

There is no "pidns=0", but I understand your point. This would come back to Steve G.'s point about disappearing fields, and the value of having it as a seperate record that could be filtered.

Quoted 7 lines

Quoted 6 lines

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

Quoted 7 lines

Quoted 4 lines

I'd love to be able to grep for netns=20 and find both the NS_INIT and the SYSCALL/NS_INFO records, instead of having them named different things. So basically I think you want to translate the type= into a string for the old_X= and X=...

That actually makes a bit more sense, and we could do away with the "type=" field since the "Xns=" fields are self-describing.

Steve, how do you feel about this since the NS_INIT/NS_DEL record type would have changing fields amongst the 6 namespace types. Only one of the 6 would be present in each message. I suppose we could have an NS_INIT_XXX/NS_DEL_XXX for each namespace type.

Quoted 1 line

- RGB

- RGB

--
Richard Guy Briggs <[email protected]> Senior Software Engineer, Kernel Security, AMER ENG Base Operating Systems, Red Hat Remote, Ottawa, Canada
Voice: +1.647.777.2635, Internal: (81) 32635, Alt: +1.613.693.0684x3545

Quoted 15 lines · Eric W. Biederman

Quoted 9 lines · Richard Guy Briggs

Quoted 4 lines · Richard Guy Briggs

Quoted 3 lines · Eric Paris

Quoted 2 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

(Including the Linux API list due to the additions to /proc/<pid>/ns/. Please see http://www.kernelhub.org/?p=2&msg=477668 and in particular
http://www.kernelhub.org/?msg=477678&p=2 )

Sigh if you have to use something like this use the proc inode number. It is the same thing.

Quoted 7 lines

I hate to claim it is unique absent of the proc superblock but it is and will be for the forseable future.

It would be better to include the block device number that appears in proc of 3h of the primary mount of to qualify the number. But it is not particularly important. Coming up with an additional unique number that needs to be maintained seems stronlgy silly.

I am reading a contradiction here:

	https://www.redhat.com/archives/linux-audit/2013-March/msg00032.html

and this posting went completely ignored:

	https://www.redhat.com/archives/linux-audit/2014-January/msg00180.html

And then there was this patchset and thread where there was some good discussion to clarify the use case:

	https://lkml.org/lkml/2014/4/22/662

Then V2:

	https://lkml.org/lkml/2014/5/9/637

Then V3 3 months ago:

	https://www.redhat.com/archives/linux-audit/2014-May/msg00071.html

I'm about to post another version of the patchset addressing Eric Paris' concerns about record types, field naming...

Quoted 4 lines · Richard Guy Briggs

Quoted 3 lines · Eric Paris

Quoted 2 lines · Richard Guy Briggs

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

(Including the Linux API list due to the additions to /proc/<pid>/ns/. Please see http://www.kernelhub.org/?p=2&msg=477668 and in particular
http://www.kernelhub.org/?msg=477678&p=2 )

<snip>

Quoted 10 lines

Quoted 10 lines

Quoted 9 lines

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

	type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

Quoted 5 lines

Quoted 2 lines

I'm undecided if I'd rather see this as a separate NS_INFO record type. It would mean we could filter them out of the logs...

I don't have a strong opinion either way. Steve G.'s opinion would be helpful here.

Breaking this out into a seperate record type would mean calling audit_log_namespace_info() from the callers of audit_log_task_info() (presently audit_log_link_denied(), ima_audit_measurement(), audit_log_exit() ), but would make it easier to emit with other record types.

Quoted 14 lines

Quoted 2 lines

Do we print out lots of pidns=0 for tasks not in a newly created NS? Do we want to?

There is no "pidns=0", but I understand your point. This would come back to Steve G.'s point about disappearing fields, and the value of having it as a seperate record that could be filtered.

Quoted 7 lines

Quoted 6 lines

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

	type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1

Quoted 7 lines

Quoted 4 lines

I'd love to be able to grep for netns=20 and find both the NS_INIT and the SYSCALL/NS_INFO records, instead of having them named different things. So basically I think you want to translate the type= into a string for the old_X= and X=...

That actually makes a bit more sense, and we could do away with the "type=" field since the "Xns=" fields are self-describing.

Steve, how do you feel about this since the NS_INIT/NS_DEL record type would have changing fields amongst the 6 namespace types. Only one of the 6 would be present in each message. I suppose we could have an NS_INIT_XXX/NS_DEL_XXX for each namespace type.

Quoted 1 line

- RGB

- RGB

--
Richard Guy Briggs <[email protected]> Senior Software Engineer, Kernel Security, AMER ENG Base Operating Systems, Red Hat Remote, Ottawa, Canada
Voice: +1.647.777.2635, Internal: (81) 32635, Alt: +1.613.693.0684x3545

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

        type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

        type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1
        type=NS_DEL msg=audit(1400217435.730:95): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 snum=a1 res=1

v2 -> v3:
Use atomic64_t in ns_serial to simplify it.
Avoid funciton duplication in proc, keying on dentry. Squash down audit patch to avoid rcu sleep issues. Add tracking for creation and deletion of namespace instances.

v1 -> v2:
Avoid rollover by switching from an int to a long long. Change rollover behaviour from simply avoiding zero to raising a BUG. Expose serial numbers in /proc/<pid>/ns/*_snum.
Expose ns_entries and use it in audit.

Notes:
There has been some progress made for audit in net namespaces and pid namespaces since this previous thread. net namespaces are now served as peers by one auditd in the init_net namespace with processes in a non-init_net namespace being able to write records if they are in the init_user_ns and have CAP_AUDIT_WRITE. Processes in a non-init_pid_ns can now similarly write records. As for CAP_AUDIT_READ, I just posted a patchset to check capabilities of userspace processes that try to join netlink broadcast groups.

This set does not try to solve the non-init namespace audit messages and auditd problem yet. That will come later, likely with additional auditd instances running in another namespace with a limited ability to influence the master auditd. I echo Eric B's idea that messages destined for different namespaces would have to be tailored for that namespace with references that make sense (such as the right pid number reported to that pid namespace, and not leaking info about parents or peers).

Bugs:
Patch 6/6 has a timing bug such that mnt and net namespace initial namespaces never get logged, I suspect because they are initialized before the audit subsystem. I've tried moving audit from __initcall to subsys_initcall, but that doesn't help.

Questions:
Is there a way to link serial numbers of namespaces involved in migration of a container to another kernel? It sounds like what is needed is a part of a mangement application that is able to pull the audit rcords from constituent hosts to build an audit trail of a container.

What additional events should list this information?

Does this present any problematic information leaks? Only CAP_AUDIT_CONTROL (and proposed CAP_AUDIT_READ) in init_user_ns can get to this information in the init namespace at the moment from audit. *However*, the addition of the proc/<pid>/ns/*_snum does make it available to other processes now.

Richard Guy Briggs (6):
namespaces: assign each namespace instance a serial number
namespaces: expose namespace instance serial number in proc_ns_operations
namespaces: expose ns instance serial numbers in proc
namespaces: expose ns_entries
audit: log namespace serial numbers
audit: log creation and deletion of namespace instances

 fs/mount.h                     |    1 +
 fs/namespace.c                 |   12 +++++++++
 fs/proc/namespaces.c           |   35 +++++++++++++++++++-------
 include/linux/audit.h          |   15 +++++++++++
 include/linux/ipc_namespace.h  |    1 +
 include/linux/nsproxy.h        |    8 ++++++
 include/linux/pid_namespace.h  |    1 +
 include/linux/proc_ns.h        |    2 +
 include/linux/user_namespace.h |    1 +
 include/linux/utsname.h        |    1 +
 include/net/net_namespace.h    |    1 +
 include/uapi/linux/audit.h     |    2 +
 init/version.c                 |    1 +
 ipc/msgutil.c                  |    1 +
 ipc/namespace.c                |   20 +++++++++++++++
 kernel/audit.c                 |   53 +++++++++++++++++++++++++++++++++++++++-
 kernel/nsproxy.c               |   17 +++++++++++++
 kernel/pid.c                   |    1 +
 kernel/pid_namespace.c         |   19 ++++++++++++++
 kernel/user.c                  |    1 +
 kernel/user_namespace.c        |   18 +++++++++++++
 kernel/utsname.c               |   20 +++++++++++++++
 net/core/net_namespace.c       |   20 ++++++++++++++-
 23 files changed, 240 insertions(+), 11 deletions(-)

_______________________________________________
Containers mailing list Containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
https://lists.linuxfoundation.org/mailman/listinfo/containers

Richard,

Quoted 20 lines

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

Quoted 2 lines

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

Since the 3 and 4 change the ABI, please CC iterations of this patch
series to [email protected], as per Documentation/SubmitChecklist.

Michael

Quoted 102 lines

5/6 provides an example of usage for audit_log_task_info() which is used by syscall audits, among others. audit_log_task() and audit_common_recv_message() would be other potential use cases.

Proposed output format:
This differs slightly from Aristeu's patch because of the label conflict with "pid=" due to including it in existing records rather than it being a seperate record. The serial numbers are printed in hex.

        type=SYSCALL msg=audit(1399651071.433:72): arch=c000003e syscall=272 success=yes exit=0 a0=40000000 a1=ffffffffffffffff a2=0 a3=22 items=0 ppid=1 pid=483 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="(t-daemon)" exe="/usr/lib/systemd/systemd" netns=97 utsns=2 ipcns=1 pidns=4 userns=3 mntns=5 subj=system_u:system_r:init_t:s0 key=(null)

6/6 tracks the creation and deletion of of namespaces, listing the type of namespace instance, related namespace id if there is one and the newly minted serial number.

Proposed output format:

        type=NS_INIT msg=audit(1400217435.706:94): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 old_snum=0 snum=a1 res=1
        type=NS_DEL msg=audit(1400217435.730:95): pid=524 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:mount_t:s0 type=20000 snum=a1 res=1

v2 -> v3:
Use atomic64_t in ns_serial to simplify it.
Avoid funciton duplication in proc, keying on dentry. Squash down audit patch to avoid rcu sleep issues. Add tracking for creation and deletion of namespace instances.

v1 -> v2:
Avoid rollover by switching from an int to a long long. Change rollover behaviour from simply avoiding zero to raising a BUG. Expose serial numbers in /proc/<pid>/ns/*_snum.
Expose ns_entries and use it in audit.

Notes:
There has been some progress made for audit in net namespaces and pid namespaces since this previous thread. net namespaces are now served as peers by one auditd in the init_net namespace with processes in a non-init_net namespace being able to write records if they are in the init_user_ns and have CAP_AUDIT_WRITE. Processes in a non-init_pid_ns can now similarly write records. As for CAP_AUDIT_READ, I just posted a patchset to check capabilities of userspace processes that try to join netlink broadcast groups.

This set does not try to solve the non-init namespace audit messages and auditd problem yet. That will come later, likely with additional auditd instances running in another namespace with a limited ability to influence the master auditd. I echo Eric B's idea that messages destined for different namespaces would have to be tailored for that namespace with references that make sense (such as the right pid number reported to that pid namespace, and not leaking info about parents or peers).

Bugs:
Patch 6/6 has a timing bug such that mnt and net namespace initial namespaces never get logged, I suspect because they are initialized before the audit subsystem. I've tried moving audit from __initcall to subsys_initcall, but that doesn't help.

Questions:
Is there a way to link serial numbers of namespaces involved in migration of a container to another kernel? It sounds like what is needed is a part of a mangement application that is able to pull the audit rcords from constituent hosts to build an audit trail of a container.

What additional events should list this information?

Does this present any problematic information leaks? Only CAP_AUDIT_CONTROL (and proposed CAP_AUDIT_READ) in init_user_ns can get to this information in the init namespace at the moment from audit. *However*, the addition of the proc/<pid>/ns/*_snum does make it available to other processes now.

Richard Guy Briggs (6):
namespaces: assign each namespace instance a serial number
namespaces: expose namespace instance serial number in proc_ns_operations
namespaces: expose ns instance serial numbers in proc
namespaces: expose ns_entries
audit: log namespace serial numbers
audit: log creation and deletion of namespace instances

 fs/mount.h                     |    1 +
 fs/namespace.c                 |   12 +++++++++
 fs/proc/namespaces.c           |   35 +++++++++++++++++++-------
 include/linux/audit.h          |   15 +++++++++++
 include/linux/ipc_namespace.h  |    1 +
 include/linux/nsproxy.h        |    8 ++++++
 include/linux/pid_namespace.h  |    1 +
 include/linux/proc_ns.h        |    2 +
 include/linux/user_namespace.h |    1 +
 include/linux/utsname.h        |    1 +
 include/net/net_namespace.h    |    1 +
 include/uapi/linux/audit.h     |    2 +
 init/version.c                 |    1 +
 ipc/msgutil.c                  |    1 +
 ipc/namespace.c                |   20 +++++++++++++++
 kernel/audit.c                 |   53 +++++++++++++++++++++++++++++++++++++++-
 kernel/nsproxy.c               |   17 +++++++++++++
 kernel/pid.c                   |    1 +
 kernel/pid_namespace.c         |   19 ++++++++++++++
 kernel/user.c                  |    1 +
 kernel/user_namespace.c        |   18 +++++++++++++
 kernel/utsname.c               |   20 +++++++++++++++
 net/core/net_namespace.c       |   20 ++++++++++++++-
 23 files changed, 240 insertions(+), 11 deletions(-)

Quoted 4 lines

_______________________________________________
Containers mailing list Containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
https://lists.linuxfoundation.org/mailman/listinfo/containers

Signature · 3 lines

Michael Kerrisk Linux man-pages maintainer;
http://www.kernel.org/doc/man-pages/
Author of "The Linux Programming Interface", http://blog.man7.org/

Quoted 1 line · Michael Kerrisk

Richard,

Hi Michael,

Quoted 21 lines · Richard Guy Briggs

Quoted 20 lines

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

Quoted 5 lines

Quoted 2 lines

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

Since the 3 and 4 change the ABI, please CC iterations of this patch
series to [email protected], as per Documentation/SubmitChecklist.

Neither patch 3/6 nor 4/6 changes the syscall interface.

Patch 3/6 adds /proc/<pid>/ns/ entries, which looks more like #16 in that document (for which /proc/<pid>/ns/<nstype> was never added).

Hi Richard

Quoted 26 lines · Richard Guy Briggs

Quoted 1 line · Michael Kerrisk

Richard,

Hi Michael,

Quoted 21 lines · Richard Guy Briggs

Quoted 20 lines

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

Quoted 7 lines

Quoted 5 lines

Quoted 2 lines

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

Since the 3 and 4 change the ABI, please CC iterations of this patch
series to [email protected], as per Documentation/SubmitChecklist.

Neither patch 3/6 nor 4/6 changes the syscall interface.

(Agreed.)

Quoted 2 lines

Patch 3/6 adds /proc/<pid>/ns/ entries, which looks more like #16 in that document (for which /proc/<pid>/ns/<nstype> was never added).

But, that's a change to the surface that the kernel exposes to user space, right? If so, it is best CCed to linux-api.

Thanks,

Michael

Hi Richard

Quoted 26 lines · Richard Guy Briggs

Quoted 1 line · Michael Kerrisk

Richard,

Hi Michael,

Quoted 21 lines · Richard Guy Briggs

Quoted 20 lines

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

Quoted 7 lines

Quoted 5 lines

Quoted 2 lines

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

Since the 3 and 4 change the ABI, please CC iterations of this patch
series to [email protected], as per Documentation/SubmitChecklist.

Neither patch 3/6 nor 4/6 changes the syscall interface.

(Agreed.)

Quoted 2 lines

Patch 3/6 adds /proc/<pid>/ns/ entries, which looks more like #16 in that document (for which /proc/<pid>/ns/<nstype> was never added).

But, that's a change to the surface that the kernel exposes to user space, right? If so, it is best CCed to linux-api.

Thanks,

Michael

Quoted 1 line · Michael Kerrisk

Richard,

Hi Michael,

Quoted 21 lines · Richard Guy Briggs

Quoted 20 lines

The purpose is to track namespaces in use by logged processes from the perspective of init_*_ns.

1/6 defines a function to generate them and assigns them.

Use a serial number per namespace (unique across one boot of one kernel) instead of the inode number (which is claimed to have had the right to change reserved and is not necessarily unique if there is more than one proc fs). It could be argued that the inode numbers have now become a defacto interface and can't change now, but I'm proposing this approach to see if this helps address some of the objections to the earlier patchset.

2/6 adds access functions to get to the serial numbers in a similar way to inode access for namespace proc operations.

3/6 implements, as suggested by Serge Hallyn, making these serial numbers available in /proc/self/ns/{ipc,mnt,net,pid,user,uts}_snum. I chose "snum" instead of "seq" for consistency with inum and there are a number of other uses of "seq" in the namespace code.

Quoted 5 lines

Quoted 2 lines

4/6 exposes proc's ns entries structure which lists a number of useful operations per namespace type for other subsystems to use.

Since the 3 and 4 change the ABI, please CC iterations of this patch
series to [email protected], as per Documentation/SubmitChecklist.

Neither patch 3/6 nor 4/6 changes the syscall interface.

Patch 3/6 adds /proc/<pid>/ns/ entries, which looks more like #16 in that document (for which /proc/<pid>/ns/<nstype> was never added).