mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adam B" <a...@mesosphere.io>
Subject Re: Review Request 36281: Document per-container unique egress flows and network queueing statistics.
Date Wed, 08 Jul 2015 00:01:39 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/36281/#review90827
-----------------------------------------------------------


Thanks for writing this up! I had a few minor suggestions and some questions.


docs/network-isolation.md (line 11)
<https://reviews.apache.org/r/36281/#comment143970>

    "follow the following" is a little redundant.



docs/network-isolation.md (line 15)
<https://reviews.apache.org/r/36281/#comment143971>

    s/kernels versions/kernel versions/



docs/network-isolation.md (lines 24 - 29)
<https://reviews.apache.org/r/36281/#comment143974>

    Do these dependencies need to be added to getting-started.md?



docs/network-isolation.md (line 31)
<https://reviews.apache.org/r/36281/#comment143975>

    `### Build Configuration`?



docs/network-isolation.md (lines 40 - 42)
<https://reviews.apache.org/r/36281/#comment143976>

    I would put the parameter example between these two sentences. Where it is now, I expect
to see the "appropriate error" just described.



docs/network-isolation.md (line 48)
<https://reviews.apache.org/r/36281/#comment143978>

    s/so that service discovery/so that the service discovery/



docs/network-isolation.md (line 68)
<https://reviews.apache.org/r/36281/#comment143981>

    s/ephemerlal/ephemeral/



docs/network-isolation.md (lines 70 - 73)
<https://reviews.apache.org/r/36281/#comment143983>

    Is 1024 a reasonable example ephemeral_ports_per_container? Your current example only
allows 24 containers on that slave, which is extremely low. Maybe 16 ports per container would
make more sense, yielding 1536 possible containers.



docs/network-isolation.md (line 72)
<https://reviews.apache.org/r/36281/#comment143987>

    `s/\/\/`?



docs/network-isolation.md (line 79)
<https://reviews.apache.org/r/36281/#comment143986>

    Why is this a fixed constant limit? Seems like we might want to adjust this depending
on how many containers are running, or give some containers more bandwidth than others. Why
isn't this just another resource type (outbound network bandwidth) with a fixed total, where
subsets can be reserved/claimed by different frameworks/tasks.



docs/network-isolation.md (line 85)
<https://reviews.apache.org/r/36281/#comment143989>

    Is the `="true"` necessary, or would `--egress_unique_flow_per_container` work on its
own?



docs/network-isolation.md (line 93)
<https://reviews.apache.org/r/36281/#comment143988>

    Maybe leave out cpu/mem/disk to let the slave auto-detect them (and shorten this line)?



docs/network-isolation.md (line 178)
<https://reviews.apache.org/r/36281/#comment143991>

    lowercase?



docs/network-isolation.md (line 198)
<https://reviews.apache.org/r/36281/#comment143992>

    lowercase?



docs/network-isolation.md (line 204)
<https://reviews.apache.org/r/36281/#comment143990>

    This footnote `[1]` is never referenced above


- Adam B


On July 7, 2015, 2:54 p.m., Paul Brett wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/36281/
> -----------------------------------------------------------
> 
> (Updated July 7, 2015, 2:54 p.m.)
> 
> 
> Review request for mesos, Ian Downes, Jie Yu, and Cong Wang.
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Document per-container unique egress flows and network queueing statistics.
> 
> 
> Diffs
> -----
> 
>   docs/home.md bc277910907c381c08835b6e9d485b27d6da5002 
>   docs/network-isolation.md PRE-CREATION 
>   docs/network-monitoring.md 8889fb165cc70bc382be0c99de8d7748328abf57 
> 
> Diff: https://reviews.apache.org/r/36281/diff/
> 
> 
> Testing
> -------
> 
> Rendered at https://www.notehub.org/2015/7/7/network-isolation for review.
> 
> 
> Thanks,
> 
> Paul Brett
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message