+1 (binding). tried it on AWS and left running for about a day. Everything 
works great 

25.02.2017, 14:50, "David Lyle" <[email protected]>:
> 16GB. It's a MBP from a few years back. I'm getting the same result as most
> it seems, currently oversubscribed but doable with care, works way better
> with the Ansible/Ambari integration that I'm doing for METRON-671.
>
> -D...
>
> On Sat, Feb 25, 2017 at 2:13 PM, Matt Foley <[email protected]> wrote:
>
>>  David, how much RAM do you have in the test system where you run full dev?
>>
>>  On 2/25/17, 8:46 AM, "David Lyle" <[email protected]> wrote:
>>
>>      +1 (binding)
>>      Checked signatures
>>      Tests passed
>>      Full dev ran sucessfully-ish (bit more below, tl;dr- it's ugly, but
>>  much
>>      better after METRON-671)
>>      Data flowed into the index/dashboard/hdfs
>>
>>      On full-dev: Monit got into a state where it started killing topologies
>>      because the Storm cli was non-responsive due to memory pressure. That
>>      caused cascading failures. I let it be for about 20 minutes, and it
>>  seemed
>>      to calm down. I then killed pcap-service and Monit to make some
>>  headroom. I
>>      haven't seen this in my work with METRON-671 as only sensor probes and
>>      pcap-service are managed by Monit. Once the topologies are monitored by
>>      Ambari, we no longer use the Storm CLI for status so that frees up some
>>      additional memory.
>>
>>      -D...
>>
>>      On Sat, Feb 25, 2017 at 10:42 AM, David Lyle <[email protected]>
>>  wrote:
>>
>>      > I'm running it up now. I don't think it should affect the release,
>>  but
>>      > it's extremely important that full-dev work. It's how quick-dev is
>>  created.
>>      > Any notion that it's been deprecated is incorrect.
>>      >
>>      > Fwiw, I've been working with it for the last few weeks as part of
>>      > METRON-671 and it's been (knocking loudly on wood) reliable.
>>      > I think we should leave full-dev in as the release verification step,
>>      > particularly in the case that a release changes the Hadoop versions.
>>      > There's a bit of a chicken-egg problem when we bump versions and
>>  quick dev
>>      > will lag.
>>      >
>>      > I'll update with run results shortly.
>>      >
>>      > -D...
>>      >
>>      >
>>      > On Sat, Feb 25, 2017 at 9:24 AM, Otto Fowler <
>>  [email protected]>
>>      > wrote:
>>      >
>>      >> I see this with full dev on my laptop pretty much every time.
>>      >> I think it is a resource issue. I see not enough memory errors
>>  trying to
>>      >> start things.
>>      >>
>>      >>
>>      >>
>>      >> On February 25, 2017 at 09:15:15, Ryan Merriman (
>>  [email protected])
>>      >> wrote:
>>      >>
>>      >> When I go to Ambari to ensure the services are all up, HDFS is
>>  down. I
>>      >> tried it 4 or 5 times and got the same result each time. I've seen
>>  others
>>      >> validate with quick dev so I assumed full dev was not used anymore.
>>  I'll
>>      >> spin it up this morning and get a more detailed error.
>>      >>
>>      >> Is anyone else able to validate it in full dev?
>>      >>
>>      >> > On Feb 25, 2017, at 7:21 AM, Casey Stella <[email protected]>
>>  wrote:
>>      >> >
>>      >> > What exactly are the errors that you saw, Ryan?
>>      >> >> On Sat, Feb 25, 2017 at 07:31 David Lyle <[email protected]>
>>  wrote:
>>      >> >>
>>      >> >> Is there any reason full dev shouldn't be working?
>>      >> >>
>>      >> >>> On Fri, Feb 24, 2017 at 9:19 PM, Casey Stella <
>>  [email protected]>
>>      >> wrote:
>>      >> >>>
>>      >> >>> Sounds like a good idea to me; thanks Ryan!
>>      >> >>>> On Fri, Feb 24, 2017 at 21:11 Ryan Merriman <
>>  [email protected]>
>>      >> wrote:
>>      >> >>>>
>>      >> >>>> +1 binding
>>      >> >>>>
>>      >> >>>> Verified the signature
>>      >> >>>> Passed maven tests
>>      >> >>>> Started quick-dev, verified data in ES, kibana, and checked the
>>      >> >>> topologies
>>      >> >>>> for errors (bro topology has parsing errors but I think a
>>  couple bad
>>      >> >>>> messages in bro data set is normal)
>>      >> >>>> Tested REPL
>>      >> >>>> RPMs built fine
>>      >> >>>>
>>      >> >>>> The recommended build validation wiki page (
>>      >> https://cwiki.apache.org/
>>      >> >>>> confluence/display/METRON/Verifying+Builds
>>      >> >>>> <https://cwiki.apache.org/confluence/display/METRON/
>>  Verifying+Builds
>>      >> >)
>>      >> >>>> has some mistakes. This did
>>      >> >>>> not run successfully in full-dev-platform and the HDFS paths
>>  look
>>      >> like
>>      >> >>> they
>>      >> >>>> are old. I am happy to update the wiki page if everyone agrees
>>  these
>>      >> >> are
>>      >> >>>> legitimate mistakes.
>>      >> >>>>
>>      >> >>>> On Fri, Feb 24, 2017 at 9:22 AM, Justin Leet <
>>  [email protected]>
>>      >> >>>> wrote:
>>      >> >>>>
>>      >> >>>>> +1 (non-binding)
>>      >> >>>>>
>>      >> >>>>> Verified signature
>>      >> >>>>> Ran build and tests in maven
>>      >> >>>>> Ran up in quick-dev and saw data flow through topologies into
>>  the UI
>>      >> >>>>> Ensured the REPL spun up and performed some basic tasks
>>      >> >>>>> Built rpms
>>      >> >>>>>
>>      >> >>>>> Justin
>>      >> >>>>>
>>      >> >>>>>> On Thu, Feb 23, 2017 at 11:18 AM, Casey Stella <
>>  [email protected]
>>      >> >
>>      >> >>>>> wrote:
>>      >> >>>>>
>>      >> >>>>>> This is a call to vote on releasing Apache Metron 0.3.1-RC5
>>      >> >>> incubating
>>      >> >>>>>>
>>      >> >>>>>> Full list of changes in this release:
>>      >> >>>>>> https://dist.apache.org/repos/dist/dev/incubator/metron/0.3.
>>      >> >>>>>> 1-RC5-incubating/CHANGES
>>      >> >>>>>>
>>      >> >>>>>> The tag/commit to be voted upon is apache-metron-0.3.1-rc5-
>>      >> >>> incubating:
>>      >> >>>>>> https://git-wip-us.apache.org/repos/asf?p=incubator-metron.
>>      >> >>>>>> git;a=shortlog;h=refs/tags/apache-metron-0.3.1-rc5-
>>  incubating
>>      >> >>>>>>
>>      >> >>>>>> The source archive being voted upon can be found here:
>>      >> >>>>>> https://dist.apache.org/repos/dist/dev/incubator/metron/0.3.
>>      >> >>>>>> 1-RC5-incubating/apache-metron-0.3.1-rc5-incubating.tar.gz
>>      >> >>>>>>
>>      >> >>>>>> Other release files, signatures and digests can be found
>>  here:
>>      >> >>>>>> https://dist.apache.org/repos/dist/dev/incubator/metron/0.3.
>>      >> >>>>>> 1-RC5-incubating/
>>      >> >>>>>>
>>      >> >>>>>> The release artifacts are signed with the following key:
>>      >> >>>>>> https://git-wip-us.apache.org/repos/asf?p=incubator-metron.
>>      >> >>>>>> git;a=blob;f=KEYS;h=8381e96d64c249a0c1b489bc0c234d
>>      >> >>>>> 9c260ba55e;hb=refs/tags/
>>      >> >>>>>> apache-metron-0.3.1-rc5-incubating
>>      >> >>>>>>
>>      >> >>>>>> The book associated with this RC is located at
>>      >> >>>>>> https://dist.apache.org/repos/dist/dev/incubator/metron/0.3.
>>      >> >>>>>> 1-RC5-incubating/book-site/index.html
>>      >> >>>>>>
>>      >> >>>>>> Please vote on releasing this package as Apache Metron
>>  0.3.1-RC5
>>      >> >>>>> incubating
>>      >> >>>>>>
>>      >> >>>>>> When voting, please list the actions taken to verify the
>>  release.
>>      >> >>>>>>
>>      >> >>>>>> Recommended build validation and verification instructions
>>  are
>>      >> >> posted
>>      >> >>>>> here:
>>      >> >> https://cwiki.apache.org/confluence/display/METRON/
>>  Verifying+Builds
>>      >> >>>>>>
>>      >> >>>>>>
>>      >> >>>>>> This vote will be open for at least 72 hours.
>>      >> >>>>>>
>>      >> >>>>>> [ ] +1 Release this package as Apache Metron 0.3.1-RC5
>>  incubating
>>      >> >>>>>>
>>      >> >>>>>> [ ] 0 No opinion
>>      >> >>>>>>
>>      >> >>>>>> [ ] -1 Do not release this package because...
>>      >> >>
>>      >>
>>      >
>>      >

------------------- 
Thank you,

James Sirota
PPMC- Apache Metron (Incubating)
jsirota AT apache DOT org

Reply via email to