[2.0] bump mongod to 3.2.9

[Mark Shuttleworth]

We're a little behind on point releases for Mongo and apparently the difference is all about production quality :)

Have noticed a couple of mongo-going-insane situations in testing, let's rev to 3.2.8 please and see how that fares.

[Test Case]
TBD

[Regression Potential]
Slight. juju-mongodb3.2 is only used by juju, and there have been several successful unit test runs with juju-mongodb3.2 3.2.8, e.g. http://data.vapour.ws/juju-ci/products/version-4299/run-unit-tests-yakkety-amd64/build-197/consoleText

[Mark Shuttleworth (sabdfl)]

https://docs.mongodb.com/manual/release-notes/3.2/

[Michael Hudson-Doyle (mwhudson)]

The update is mostly straightforward, a bunch of the patches we are carrying were included in 3.2.5 which is good to see. I'm doing a test build in https://launchpad.net/~mwhudson/+archive/ubuntu/devirt/+packages.

There is one patch that doesn't apply cleanly any more (skip-high-mem-test.patch). My test build will reveal if it's still needed.

There is one not-entirely trivial change, which is that the mmapv1 engine is now unsupported and not build by default on all non-intel architectures. You can build it on any arch, but it's definitely badly broken on big endian, so I've changed rules to enable it on all architectures but s390x to match the situation with the build currently in the archive.

[Launchpad Janitor (janitor)]

This bug was fixed in the package juju-mongodb3.2 - 3.2.8-0ubuntu1

---------------
juju-mongodb3.2 (3.2.8-0ubuntu1) yakkety; urgency=medium

  * New upstream release. (LP: #1605976)
  * Drop patches which are now included upstream:
    - offset-outside-of-bounds.patch
    - scons-detect-ppc64.patch
    - use-4kb-sector-size-for-direct-io.patch
    - fix-debug-non-jit-builds-of-spidermonkey.patch
    - processinfo_test-fails-on-ppc64.patch
    - dont-assume-char-is-unsigned-in-logging.patch
    - use-signed-char-explicitly.patch
    - warn-user-if-locked-memory-limit-is-too-low.patch
  * Drop d/patches/skip-high-mem-test.patch, that test now passes on Launchpad
    builders.
  * Regenerate s390x-support.patch.
  * Explicitly enable mmapv1 on all architectures except s390x.

 -- Michael Hudson-Doyle <email address hidden> Tue, 26 Jul 2016 11:41:46 +1200

[Michael Hudson-Doyle (mwhudson)]

I've uploaded 3.2.8 to the xenial queue, but in the mean time if someone wants to do some testing you can get 3.2.8-0ubuntu3~16.04~ppa1 from ppa:mwhudson/devirt.

[Nicholas Skaggs (nskaggs)]

@mwhudson, once mongo3.2.8 is in the proposed queue for xenial, we can ensure a CI test run or two before landing. Let us know.

[Mark Shuttleworth (sabdfl)]

MongoDB 3.2.*9* was released on August 16th. I did not intend the
version to be so literal ;)

Mark

[Michael Hudson-Doyle (mwhudson)]

And yakkety has had 3.2.8 for a month, but the SRU never happened :( Uploading 3.2.9 to yakkety now, and assigning the Xenial task to Nicholas to get the QA done before actually doing the SRU properly this time.

[Launchpad Janitor (janitor)]

This bug was fixed in the package juju-mongodb3.2 - 3.2.9-0ubuntu1

---------------
juju-mongodb3.2 (3.2.9-0ubuntu1) yakkety; urgency=medium

  * New upstream release. (LP: #1605976) (again)
  * Drop one part of s390x-support.patch that was included upstream.

 -- Michael Hudson-Doyle <email address hidden> Tue, 30 Aug 2016 11:40:44 +1200

[Mark Shuttleworth (sabdfl)]

Ah, thank you, I thought we had bundled the database in the Juju package
and didn't realise it needed all the independent machinery to be invoked.

Bring on the snap era :)

I do think we should track upstream releases of Mongo because they are
essentially free performance and resilience fixes of a layer we depend
on heavily. I would say we could afford to be slow (six months) to jump
to a new 3.x release but should grab 3.2.x minor point releases
aggressively.

Mark

[Michael Hudson-Doyle (mwhudson)]

On 31 August 2016 at 02:45, Mark Shuttleworth <email address hidden>
wrote:

> Ah, thank you, I thought we had bundled the database in the Juju package
> and didn't realise it needed all the independent machinery to be invoked.
>
> Bring on the snap era :)
>

Yeah.

I do think we should track upstream releases of Mongo because they are
> essentially free performance and resilience fixes of a layer we depend
> on heavily. I would say we could afford to be slow (six months) to jump
> to a new 3.x release

I *think* with 3.4 we could use a pure upstream snap, even. Currently we
have s390x support backported from 3.3.

> but should grab 3.2.x minor point releases
> aggressively.
>

Ack. I've subscribed to mongodb-announce now...

[Chris J Arges (arges)]

Hello Mark, or anyone else affected,

Accepted juju-mongodb3.2 into xenial-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/juju-mongodb3.2/3.2.9-0ubuntu1~16.04 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

[Nicholas Skaggs (nskaggs)]

Looks good.

/usr/lib/juju/mongo3.2/bin/mongo
MongoDB shell version: 3.2.9

[Michael Hudson-Doyle (mwhudson)]

Can we get a CI run or at least some kind of qa with the new mongo? The point of SRU verification is not only to verify that the bug is fixed but also that nothing new has broken :-) (I tend to forget this too!)

[Mark Shuttleworth (sabdfl)]

Am able to upgrade a bootstrapped Juju controller without issues here.
That's only lightweight QA but it's a data point for you.

Mark

[Nicholas Skaggs (nskaggs)]

@mwhudson, I utilized the tool locally and everything seemed fine. For now, bug 1564500 prevents a full CI run until it lands in xenial proper. I think we should discuss a set of autopkgtests to help bridge the gap in the interim.

That said, given 3.2.9 is in yakkety we've not seen any issues in the CI runs that occur within yakkety which I consider a useful datapoint in this case. It's been running for a couple weeks.

[Michael Hudson-Doyle (mwhudson)]

On 16 September 2016 at 01:34, Nicholas Skaggs <
<email address hidden>> wrote:

> @mwhudson, I utilized the tool locally and everything seemed fine. For
> now, bug 1564500 prevents a full CI run until it lands in xenial proper.
>

Oh well.

> I think we should discuss a set of autopkgtests to help bridge the gap
> in the interim.
>

We do have basic autopkgtests of course. I guess we could add one that runs
the unit tests too?

> That said, given 3.2.9 is in yakkety we've not seen any issues in the CI
> runs that occur within yakkety which I consider a useful datapoint in
> this case. It's been running for a couple weeks.

Oh yeah, that's a good point. Consider me satisfied :-)

Cheers,
mwh

[Launchpad Janitor (janitor)]

This bug was fixed in the package juju-mongodb3.2 - 3.2.9-0ubuntu1~16.04

---------------
juju-mongodb3.2 (3.2.9-0ubuntu1~16.04) xenial; urgency=medium

  * Build for Xenial. (LP: #1605976)

 -- Michael Hudson-Doyle <email address hidden> Thu, 01 Sep 2016 09:02:51 +1200

[Brian Murray (brian-murray)]

The verification of the Stable Release Update for juju-mongodb3.2 has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.