Bug 1025823 - race condition between kinit and rpcgssd
race condition between kinit and rpcgssd
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: krb5 (Show other bugs)
x86_64 Linux
high Severity high
: rc
: ---
Assigned To: Nalin Dahyabhai
BaseOS QE Security Team
Depends On:
Blocks: 994246 1061410
  Show dependency treegraph
Reported: 2013-11-01 12:49 EDT by Rui Gouveia
Modified: 2016-01-04 00:10 EST (History)
8 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2014-02-20 13:18:11 EST
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Rui Gouveia 2013-11-01 12:49:46 EDT
Description of problem:
In a system with Kerberized NFSv4, when the ticket expires, Kerberos tries to renew the ticket, but at the same time so does rpcgssd, it appears. Customer systems are experiencing this race condition and then mount fails.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
Given the very tight timings involved its difficult to trigger at just the right times.
Active Directory based domain (no special tunings):

1. start rpcgssd with -vvv debug
2. start rpcdebug -m rpc -s auth
3. allow rpcgssd to renew machine cred
4. trigger nfs mount to krb5 auth share at same moment.
5. review /var/log/messages for rpcdebug/rcpgssd log

alternatively you could reproduce by doing this.
1. start rpcgssd with -vvv debug
2. start rpcdebug -m rpc -s auth
3. do a kinit of the machine credentials to /tmp/krb5cc_machine_DOMAIN (same file as rpcgssd uses)
4. at same moment trigger a mount of a krb5 authenticated share.
5. review /var/log/messages for rpcdebug/rcpgssd log

For step 5 in either case , if you get the timings just right, you should see a debug trace similar to what i have shown on this case, ie:
1. machine credential refresh starts
2. mount starts
3. rpc errors with a GSS_S_CONTEXT_EXPIRED 0x000c0000 , mount fails with error -5
4. machine credential refresh finishes
5. from this point on mount will work.

NOTE: step 1-4 all happen in fractions of a second.

Actual results:
Race condition experienced, and mount fails.

Expected results:
Not the above.

Additional info:
Comment 2 RHEL Product and Program Management 2013-11-01 13:20:15 EDT
This request was evaluated by Red Hat Product Management for
inclusion in the current release of Red Hat Enterprise Linux.
Because the affected component is not scheduled to be updated
in the current release, Red Hat is unable to address this
request at this time.

Red Hat invites you to ask your support representative to
propose this request, if appropriate, in the next release of
Red Hat Enterprise Linux.
Comment 7 Dmitri Pal 2014-02-20 13:18:11 EST
This issue is addressed in RHEL7. This is a major change so it can't be ported back.

Note You need to log in before you can comment on or make changes to this bug.