From: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
To: Oren Laadan <orenl@cs.columbia.edu>
Cc: Serge Hallyn <serge@hallyn.com>,
Matt Helsley <matthltc@us.ibm.com>, Dan Smith <danms@us.ibm.com>,
John Stultz <johnstul@us.ibm.com>,
Matthew Wilcox <matthew@wil.cx>,
Jamie Lokier <jamie@shareable.org>,
Steven Whitehouse <swhiteho@redhat.com>,
<linux-fsdevel@vger.kernel.org>,
Containers <containers@lists.linux-foundation.org>
Subject: [PATCH 04/17][cr][v4]: Restore file_owner info
Date: Mon, 16 Aug 2010 12:43:08 -0700 [thread overview]
Message-ID: <1281987801-1293-5-git-send-email-sukadev@linux.vnet.ibm.com> (raw)
In-Reply-To: <1281987801-1293-1-git-send-email-sukadev@linux.vnet.ibm.com>
Restore the file-owner information for each 'struct file'. This is
essentially is like a new fcntl(F_SETOWN) and fcntl(F_SETSIG) calls,
except that the pid, uid, euid and signum values are read from the
checkpoint image.
Changelog[v4]:
- [Oren Laadan]: Sanitize the pid-type field read from checkpoint image.
Changelog[v3]:
- [Oren Laadan]: Ensure find_vpid() found a valid pid before
making it the file owner.
Changelog[v2]:
- [Matt Helsley, Serge Hallyn]: Don't trust uids in checkpoint image.
(added CAP_KILL check)
- Check that signal number read from the checkpoint image is valid.
(not sure it is required, since its an incomplete check for tampering)
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
---
fs/checkpoint.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 78 insertions(+), 0 deletions(-)
diff --git a/fs/checkpoint.c b/fs/checkpoint.c
index ce1b4af..be9d39a 100644
--- a/fs/checkpoint.c
+++ b/fs/checkpoint.c
@@ -618,6 +618,80 @@ static int attach_file(struct file *file)
return fd;
}
+static int restore_file_owner(struct ckpt_ctx *ctx, struct ckpt_hdr_file *h,
+ struct file *file)
+{
+ int ret;
+ struct pid *pid;
+ uid_t uid, euid;
+
+ uid = h->f_owner_uid;
+ euid = h->f_owner_euid;
+
+ ckpt_debug("restore_file_owner(): uid %u, euid %u, pid %d, type %d\n",
+ uid, euid, h->f_owner_pid, h->f_owner_pid_type);
+ /*
+ * We can't trust the uids in the checkpoint image and normally need
+ * CAP_KILL. But if the uids match our ids, should be fine since we
+ * have access to the file.
+ *
+ * TODO: Move this check to __f_setown() ?
+ */
+ ret = -EACCES;
+ if (!capable(CAP_KILL) &&
+ (uid != current_uid() || euid != current_euid())) {
+ ckpt_err(ctx, ret, "image uids [%d, %d] don't match current "
+ "process uids [%d, %d] and no CAP_KILL\n",
+ uid, euid, current_uid(), current_euid());
+ return ret;
+ }
+
+ ret = -EINVAL;
+ if (!valid_signal(h->f_owner_signum)) {
+ ckpt_err(ctx, ret, "Invalid signum %d\n", h->f_owner_signum);
+ return ret;
+ }
+ file->f_owner.signum = h->f_owner_signum;
+
+ switch(h->f_owner_pid_type) {
+ case PIDTYPE_PID:
+ case PIDTYPE_PGID:
+ case PIDTYPE_SID:
+ break;
+ default:
+ ckpt_err(ctx, ret, "Invalid pid-type %d\n",
+ h->f_owner_pid_type);
+ return ret;
+
+ }
+
+ rcu_read_lock();
+
+ /*
+ * If file had a non-NULL owner and we can't find the owner after
+ * restart, return error.
+ */
+ pid = find_vpid(h->f_owner_pid);
+ if (h->f_owner_pid && !pid)
+ ret = -ESRCH;
+ else {
+ /*
+ * TODO: Do we need 'force' to be 1 here or can it be 0 ?
+ * 'force' is used to modify the owner, if one is
+ * already set. Can it be set when we restart an
+ * application ?
+ */
+ ret = __f_setown(file, pid, h->f_owner_pid_type, uid, euid, 1);
+ }
+
+ rcu_read_unlock();
+
+ if (ret < 0)
+ ckpt_err(ctx, ret, "__fsetown_uid() failed\n");
+
+ return ret;
+}
+
#define CKPT_SETFL_MASK \
(O_APPEND | O_NONBLOCK | O_NDELAY | FASYNC | O_DIRECT | O_NOATIME)
@@ -651,6 +725,10 @@ int restore_file_common(struct ckpt_ctx *ctx, struct file *file,
if (ret < 0)
return ret;
+ ret = restore_file_owner(ctx, h, file);
+ if (ret < 0)
+ return ret;
+
/*
* Normally f_mode is set by open, and modified only via
* fcntl(), so its value now should match that at checkpoint.
--
1.6.0.4
next prev parent reply other threads:[~2010-08-16 19:37 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-16 19:43 [PATCH 00/17][cr][v4]: C/R file owner, locks, leases Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 01/17][cr][v4]: Add uid, euid params to f_modown() Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 02/17][cr][v4]: Add uid, euid params to __f_setown() Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 03/17][cr][v4]: Checkpoint file-owner information Sukadev Bhattiprolu
[not found] ` <1281987801-1293-4-git-send-email-sukadev-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
2010-09-16 23:34 ` Oren Laadan
2010-08-16 19:43 ` Sukadev Bhattiprolu [this message]
2010-09-16 23:45 ` [PATCH 04/17][cr][v4]: Restore file_owner info Oren Laadan
2010-08-16 19:43 ` [PATCH 05/17][cr][v4]: Move file_lock macros into linux/fs.h Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 06/17][cr][v4]: Checkpoint file-locks Sukadev Bhattiprolu
2010-09-17 0:03 ` Oren Laadan
2010-08-16 19:43 ` [PATCH 07/17][cr][v4]: Define flock_set() Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 08/17][cr][v4]: Define flock64_set() Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 09/17][cr][v4]: Restore file-locks Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 10/17][cr][v4]: Initialize ->fl_break_time to 0 Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 11/17][cr][v4]: Add ->fl_type_prev field Sukadev Bhattiprolu
2010-09-17 0:06 ` Oren Laadan
2010-08-16 19:43 ` [PATCH 12/17][cr][v4]: Add ->fl_break_notified field Sukadev Bhattiprolu
2010-09-17 0:07 ` Oren Laadan
2010-08-16 19:43 ` [PATCH 13/17][cr][v4]: Add jiffies_begin field to ckpt_ctx Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 14/17][cr][v4]: Checkpoint file-leases Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 15/17][cr][v4]: Define do_setlease() Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 16/17][cr][v4]: Restore file-leases Sukadev Bhattiprolu
2010-08-16 19:43 ` [PATCH 17/17][cr][v4]: Document design of C/R of file-locks and leases Sukadev Bhattiprolu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1281987801-1293-5-git-send-email-sukadev@linux.vnet.ibm.com \
--to=sukadev@linux.vnet.ibm.com \
--cc=containers@lists.linux-foundation.org \
--cc=danms@us.ibm.com \
--cc=jamie@shareable.org \
--cc=johnstul@us.ibm.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=matthew@wil.cx \
--cc=matthltc@us.ibm.com \
--cc=orenl@cs.columbia.edu \
--cc=serge@hallyn.com \
--cc=swhiteho@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).