I'm going to -1 (non binding) for the same reason as David Robinson.

I would classify the FD leak as serious and a violation of the isolation
that the agent provides.

It should be back ported to 1.1.0 just like how it was backported to 1.0.2

On Mon, Oct 24, 2016 at 5:37 PM, David Robinson <drobin...@twitter.com>
wrote:

> -1
>
> Can the fix for MESOS-6420 be backported? The Mesos agent leaks sockets
> when the port mapping network isolator is enabled, the leaked sockets are
> passed to the executor (the close-on-exec flag is not set) and that can
> cause problems for certain frameworks. The Aurora executor uses Kazoo (the
> python ZooKeeper library) for service announcement, Kazoo uses Python's
> select() call for polling its file descriptors and Python's select() chokes
> when there's > 1024 file descriptors. The end result for Aurora is that
> after an agent runs > 1024 tasks any new tasks will fail to announce (will
> not be registered in ZooKeeper) and will therefore be unknown to other
> services.
>
> On Tue, Oct 18, 2016 at 1:01 PM, Till Toenshoff <toensh...@me.com> wrote:
>
>> Hi all,
>>
>> Please vote on releasing the following candidate as Apache Mesos 1.1.0.
>>
>>
>> 1.1.0 includes the following:
>> ------------------------------------------------------------
>> --------------------
>>   * [MESOS-2449] - **Experimental** support for launching a group of tasks
>>     via a new `LAUNCH_GROUP` Offer operation. Mesos will guarantee that
>> either
>>     all tasks or none of the tasks in the group are delivered to the
>> executor.
>>     Executors receive the task group via a new `LAUNCH_GROUP` event.
>>
>>   * [MESOS-2533] - **Experimental** support for HTTP and HTTPS health
>> checks.
>>     Executors may now use the updated `HealthCheck` protobuf to implement
>>     HTTP(S) health checks. Both default executors (command and docker)
>> leverage
>>     `curl` binary for sending HTTP(S) requests and connect to `127.0.0.1`,
>>     hence a task must listen on all interfaces. On Linux, For BRIDGE and
>> USER
>>     modes, docker executor enters the task's network namespace.
>>
>>   * [MESOS-3421] - **Experimental** Support sharing of resources across
>>     containers. Currently persistent volumes are the only resources
>> allowed to
>>     be shared.
>>
>>   * [MESOS-3567] - **Experimental** support for TCP health checks.
>> Executors
>>     may now use the updated `HealthCheck` protobuf to implement TCP health
>>     checks. Both default executors (command and docker) connect to
>> `127.0.0.1`,
>>     hence a task must listen on all interfaces. On Linux, For BRIDGE and
>> USER
>>     modes, docker executor enters the task's network namespace.
>>
>>   * [MESOS-4324] - Allow access to persistent volumes as read-only or
>> read-write
>>     by tasks. Mesos doesn't allow persistent volumes to be created as
>> read-only
>>     but in 1.1 it starts allow tasks to use the volumes as read-only.
>> This is
>>     mainly motivated by shared persistent volumes but applies to regular
>>     persistent volumes as well.
>>
>>   * [MESOS-5275] - **Experimental** support for linux capabilities.
>> Frameworks
>>     or operators now have fine-grained control over the capabilities that
>> a
>>     container may have. This allows a container to run as root, but not
>> have all
>>     the privileges associated with the root user (e.g., CAP_SYS_ADMIN).
>>
>>   * [MESOS-5344] -- **Experimental** support for partition-aware Mesos
>>     frameworks. In previous Mesos releases, when an agent is partitioned
>> from
>>     the master and then reregisters with the cluster, all tasks running
>> on the
>>     agent are terminated and the agent is shutdown. In Mesos 1.1,
>> partitioned
>>     agents will no longer be shutdown when they reregister with the
>> master. By
>>     default, tasks running on such agents will still be killed (for
>> backward
>>     compatibility); however, frameworks can opt-in to the new
>> PARTITION_AWARE
>>     capability. If they do this, their tasks will not be killed when a
>> partition
>>     is healed. This allows frameworks to define their own policies for
>> how to
>>     handle partitioned tasks. Enabling the PARTITION_AWARE capability also
>>     introduces a new set of task states: TASK_UNREACHABLE, TASK_DROPPED,
>>     TASK_GONE, TASK_GONE_BY_OPERATOR, and TASK_UNKNOWN. These new states
>> are
>>     intended to eventually replace the TASK_LOST state.
>>
>>   * [MESOS-6077] - **Experimental** A new default executor is introduced
>> which
>>     frameworks can use to launch task groups as nested containers. All the
>>     nested containers share resources likes cpu, memory, network and
>> volumes.
>>
>>   * [MESOS-6014] - **Experimental** A new port-mapper CNI plugin, the
>>     `mesos-cni-port-mapper` has been introduced. For Mesos containers,
>> with the
>>     CNI port-mapper plugin, users can now expose container ports through
>> host
>>     ports using DNAT. This is especially useful when Mesos containers are
>>     attached to isolated CNI networks such as private bridge networks,
>> and the
>>     services running in the container needs to be exposed outside these
>>     isolated networks.
>>
>>
>> The CHANGELOG for the release is available at:
>> https://git-wip-us.apache.org/repos/asf?p=mesos.git;a=blob_p
>> lain;f=CHANGELOG;hb=1.1.0-rc1
>> ------------------------------------------------------------
>> --------------------
>>
>> The candidate for Mesos 1.1.0 release is available at:
>> https://dist.apache.org/repos/dist/dev/mesos/1.1.0-rc1/mesos-1.1.0.tar.gz
>>
>> The tag to be voted on is 1.1.0-rc1:
>> https://git-wip-us.apache.org/repos/asf?p=mesos.git;a=commit;h=1.1.0-rc1
>>
>> The MD5 checksum of the tarball can be found at:
>> https://dist.apache.org/repos/dist/dev/mesos/1.1.0-rc1/mesos
>> -1.1.0.tar.gz.md5
>>
>> The signature of the tarball can be found at:
>> https://dist.apache.org/repos/dist/dev/mesos/1.1.0-rc1/mesos
>> -1.1.0.tar.gz.asc
>>
>> The PGP key used to sign the release is here:
>> https://dist.apache.org/repos/dist/release/mesos/KEYS
>>
>> The JAR is up in Maven in a staging repository here:
>> https://repository.apache.org/content/repositories/orgapachemesos-1158
>>
>> Please vote on releasing this package as Apache Mesos 1.1.0!
>>
>> The vote is open until Fri Oct 21 21:57:02 CEST 2016 and passes if a
>> majority of at least 3 +1 PMC votes are cast.
>>
>> [ ] +1 Release this package as Apache Mesos 1.1.0
>> [ ] -1 Do not release this package because ...
>>
>> Thanks,
>> Alex & Till
>>
>>
>
>
> --
> David Robinson
> SRE - Mesos
> @daverobinson
>
> --
> Zameer Manji
>

Reply via email to