You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@mesos.apache.org by James Peach <jp...@apache.org> on 2017/04/06 18:32:29 UTC
Review Request 58242: Improve master robustness against duplicate
UPIDs.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/
-----------------------------------------------------------
Review request for mesos and Mesos Reviewbot.
Bugs: MESOS-7363
https://issues.apache.org/jira/browse/MESOS-7363
Repository: mesos
Description
-------
It is possible for a malicious client to send
libprocess SUBSCRIBE requests that will trigger the
!frameworks.principals.contains(...) CHECK. This can happen
if the client sends a subscribe with a framework ID, then a
second subscribe with a different framework ID but the same
UPID. The invariant in the master is that a UPID uniquely
identifies a given framework. This is violated if we allow
multiple frameworks with the same UPID.
Diffs
-----
src/linux/ldcache.cpp e93334465911d3ec37f38d51249486d5d317bdb3
src/master/master.cpp 6a6a570e52d21bfb2443f981e3d7faf8c36f74bc
Diff: https://reviews.apache.org/r/58242/diff/1/
Testing
-------
make checl (Fedora 25). Internal fuzzer run.
Thanks,
James Peach
Re: Review Request 58242: Improve master robustness against duplicate
UPIDs.
Posted by Mesos Reviewbot <re...@mesos.apache.org>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/#review171252
-----------------------------------------------------------
Patch looks great!
Reviews applied: [58242]
Passed command: export OS='ubuntu:14.04' BUILDTOOL='autotools' COMPILER='gcc' CONFIGURATION='--verbose' ENVIRONMENT='GLOG_v=1 MESOS_VERBOSE=1'; ./support/docker-build.sh
- Mesos Reviewbot
On April 6, 2017, 6:32 p.m., James Peach wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/58242/
> -----------------------------------------------------------
>
> (Updated April 6, 2017, 6:32 p.m.)
>
>
> Review request for mesos and Mesos Reviewbot.
>
>
> Bugs: MESOS-7363
> https://issues.apache.org/jira/browse/MESOS-7363
>
>
> Repository: mesos
>
>
> Description
> -------
>
> It is possible for a malicious client to send
> libprocess SUBSCRIBE requests that will trigger the
> !frameworks.principals.contains(...) CHECK. This can happen
> if the client sends a subscribe with a framework ID, then a
> second subscribe with a different framework ID but the same
> UPID. The invariant in the master is that a UPID uniquely
> identifies a given framework. This is violated if we allow
> multiple frameworks with the same UPID.
>
>
> Diffs
> -----
>
> src/linux/ldcache.cpp e93334465911d3ec37f38d51249486d5d317bdb3
> src/master/master.cpp 6a6a570e52d21bfb2443f981e3d7faf8c36f74bc
>
>
> Diff: https://reviews.apache.org/r/58242/diff/1/
>
>
> Testing
> -------
>
> make checl (Fedora 25). Internal fuzzer run.
>
>
> Thanks,
>
> James Peach
>
>
Re: Review Request 58242: Improve master robustness against duplicate
UPIDs.
Posted by James Peach <jp...@apache.org>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/
-----------------------------------------------------------
(Updated April 6, 2017, 11:22 p.m.)
Review request for mesos and Mesos Reviewbot.
Bugs: MESOS-7363
https://issues.apache.org/jira/browse/MESOS-7363
Repository: mesos
Description
-------
It is possible for a malicious client to send
libprocess SUBSCRIBE requests that will trigger the
!frameworks.principals.contains(...) CHECK. This can happen
if the client sends a subscribe with a framework ID, then a
second subscribe with a different framework ID but the same
UPID. The invariant in the master is that a UPID uniquely
identifies a given framework. This is violated if we allow
multiple frameworks with the same UPID.
Diffs
-----
src/master/master.cpp e547d2c410471f42bfeaae40a8b57ca30033e54f
Diff: https://reviews.apache.org/r/58242/diff/2/
Testing (updated)
-------
make check (Fedora 25). Internal fuzzer run.
Thanks,
James Peach
Re: Review Request 58242: Improve master robustness against duplicate
UPIDs.
Posted by James Peach <jp...@apache.org>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/
-----------------------------------------------------------
(Updated April 6, 2017, 11:21 p.m.)
Review request for mesos and Mesos Reviewbot.
Bugs: MESOS-7363
https://issues.apache.org/jira/browse/MESOS-7363
Repository: mesos
Description
-------
It is possible for a malicious client to send
libprocess SUBSCRIBE requests that will trigger the
!frameworks.principals.contains(...) CHECK. This can happen
if the client sends a subscribe with a framework ID, then a
second subscribe with a different framework ID but the same
UPID. The invariant in the master is that a UPID uniquely
identifies a given framework. This is violated if we allow
multiple frameworks with the same UPID.
Diffs (updated)
-----
src/master/master.cpp e547d2c410471f42bfeaae40a8b57ca30033e54f
Diff: https://reviews.apache.org/r/58242/diff/2/
Changes: https://reviews.apache.org/r/58242/diff/1-2/
Testing
-------
make checl (Fedora 25). Internal fuzzer run.
Thanks,
James Peach
Re: Review Request 58242: Improve master robustness against duplicate
UPIDs.
Posted by James Peach <jp...@apache.org>.
> On April 6, 2017, 11:18 p.m., Jie Yu wrote:
> > src/linux/ldcache.cpp
> > Line 120 (original), 120 (patched)
> > <https://reviews.apache.org/r/58242/diff/1/?file=1686218#file1686218line120>
> >
> > Why this change? Seems irrelevant?
I fat-fingered a rebase :-/
- James
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/#review171265
-----------------------------------------------------------
On April 6, 2017, 11:21 p.m., James Peach wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/58242/
> -----------------------------------------------------------
>
> (Updated April 6, 2017, 11:21 p.m.)
>
>
> Review request for mesos and Mesos Reviewbot.
>
>
> Bugs: MESOS-7363
> https://issues.apache.org/jira/browse/MESOS-7363
>
>
> Repository: mesos
>
>
> Description
> -------
>
> It is possible for a malicious client to send
> libprocess SUBSCRIBE requests that will trigger the
> !frameworks.principals.contains(...) CHECK. This can happen
> if the client sends a subscribe with a framework ID, then a
> second subscribe with a different framework ID but the same
> UPID. The invariant in the master is that a UPID uniquely
> identifies a given framework. This is violated if we allow
> multiple frameworks with the same UPID.
>
>
> Diffs
> -----
>
> src/master/master.cpp e547d2c410471f42bfeaae40a8b57ca30033e54f
>
>
> Diff: https://reviews.apache.org/r/58242/diff/2/
>
>
> Testing
> -------
>
> make checl (Fedora 25). Internal fuzzer run.
>
>
> Thanks,
>
> James Peach
>
>
Re: Review Request 58242: Improve master robustness against duplicate
UPIDs.
Posted by Jie Yu <yu...@gmail.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58242/#review171265
-----------------------------------------------------------
Fix it, then Ship it!
This LGTM.
src/linux/ldcache.cpp
Line 120 (original), 120 (patched)
<https://reviews.apache.org/r/58242/#comment244163>
Why this change? Seems irrelevant?
- Jie Yu
On April 6, 2017, 6:32 p.m., James Peach wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/58242/
> -----------------------------------------------------------
>
> (Updated April 6, 2017, 6:32 p.m.)
>
>
> Review request for mesos and Mesos Reviewbot.
>
>
> Bugs: MESOS-7363
> https://issues.apache.org/jira/browse/MESOS-7363
>
>
> Repository: mesos
>
>
> Description
> -------
>
> It is possible for a malicious client to send
> libprocess SUBSCRIBE requests that will trigger the
> !frameworks.principals.contains(...) CHECK. This can happen
> if the client sends a subscribe with a framework ID, then a
> second subscribe with a different framework ID but the same
> UPID. The invariant in the master is that a UPID uniquely
> identifies a given framework. This is violated if we allow
> multiple frameworks with the same UPID.
>
>
> Diffs
> -----
>
> src/linux/ldcache.cpp e93334465911d3ec37f38d51249486d5d317bdb3
> src/master/master.cpp 6a6a570e52d21bfb2443f981e3d7faf8c36f74bc
>
>
> Diff: https://reviews.apache.org/r/58242/diff/1/
>
>
> Testing
> -------
>
> make checl (Fedora 25). Internal fuzzer run.
>
>
> Thanks,
>
> James Peach
>
>