accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From <dlmar...@comcast.net>
Subject RE: [DISCUSS] Should we support upgrading 1.4 -> 1.6 w/o going through 1.5?
Date Wed, 30 Jul 2014 22:55:23 GMT
Please add a check that instance.dfs.uri, although deprecated, is configured in the upgrade
and that instance.volumes is NOT configured, see the discussion in ACCUMULO-3006.

-----Original Message-----
From: Mike Drob [mailto:madrob@cloudera.com] 
Sent: Wednesday, July 30, 2014 6:12 PM
To: Accumulo Dev List
Subject: Re: [DISCUSS] Should we support upgrading 1.4 -> 1.6 w/o going through 1.5?

The upgrade test looks something like:

* Start up 1.4 (or 1.5)
* Create tables with a variety of configurations (LZO/Snappy/GZ, Bloom Filter On/Off, Block
Cache On/Off)
* Load some data into these tables, enough such that flushes and compactions start to occur.
* Abruptly kill all of the servers. This ensures that there are WALs around. Unfortunately,
we can't have any compactions in progress at this point, since that causes a Fate Operation,
and that would prevent the upgrade from completing.
* Upgrade the bits, start 1.6, and wait for the upgrade steps to finish.
* Check for data loss
* Trigger some compactions and wait for those to finish. In a previous iteration, we discovered
an issue where the relative path'd files and WALs were not being properly deleted from !0.
* Check for data loss.
* Restart the tablet servers to force tablets to reload (test for orphaned files).
* Check for data loss.

We did not try using custom iterators.

Mike


On Wed, Jul 30, 2014 at 4:58 PM, Christopher <ctubbsii@apache.org> wrote:

> Do you have a list of test cases you've tried?
> (like with/without walogs, custom iterators on tables, delete markers 
> for the metadata table, etc.)?
>
> If we could consolidate some of those into a comprehensive list, 
> that'd be pretty useful, I'd think, to give confidence that we didn't 
> miss anything in the review. It might also help focus testing for 
> 1.6.1 release after you've pushed.
>
>
> --
> Christopher L Tubbs II
> http://gravatar.com/ctubbsii
>
>
> On Wed, Jul 30, 2014 at 4:39 PM, Sean Busbey <busbey@cloudera.com> wrote:
>
> > Friendly reminder that direct upgrades from 1.4 to 1.6 is under review:
> >
> > https://reviews.apache.org/r/23413/
> >
> > Pending any additional concerns, I'll be pushing this soon.
> >
> >
> > On Fri, Jul 11, 2014 at 2:22 PM, Sean Busbey <busbey@cloudera.com>
> wrote:
> >
> > > An initial port of the 1.4 -> 1.6 upgrade code for the current 
> > > 1.6.1-SNAPSHOT branch is now up:
> > >
> > > * https://issues.apache.org/jira/browse/ACCUMULO-2988
> > > * https://reviews.apache.org/r/23413/
> > >
> > >
> > > On Wed, Jun 25, 2014 at 2:36 PM, Christopher <ctubbsii@apache.org>
> > wrote:
> > >
> > >> Cool. Thanks, Sean!
> > >>
> > >>
> > >> --
> > >> Christopher L Tubbs II
> > >> http://gravatar.com/ctubbsii
> > >>
> > >>
> > >> On Wed, Jun 25, 2014 at 3:25 PM, Sean Busbey 
> > >> <busbey@cloudera.com>
> > wrote:
> > >>
> > >> > I'll be creating a ticket and posting a patch this week.
> > >> >
> > >> >
> > >> > On Wed, Jun 25, 2014 at 2:21 PM, Christopher 
> > >> > <ctubbsii@apache.org>
> > >> wrote:
> > >> >
> > >> > > So, just to revisit this conversation, it seems like there is
> > >> interest in
> > >> > > supporting this. Is there already a ticket for it and/or 
> > >> > > somebody interested in doing the necessary work for 1.6.1?
> > >> > >
> > >> > >
> > >> > > --
> > >> > > Christopher L Tubbs II
> > >> > > http://gravatar.com/ctubbsii
> > >> > >
> > >> > >
> > >> > > On Thu, Jun 19, 2014 at 11:07 AM, Mike Drob 
> > >> > > <madrob@cloudera.com>
> > >> wrote:
> > >> > >
> > >> > > > In a nutshell: stop 1.4, install 1.6, copy the WALs to HDFS

> > >> > > > (ACCUMULO-2770), start 1.6
> > >> > > >
> > >> > > > Mike
> > >> > > >
> > >> > > >
> > >> > > > On Wed, Jun 18, 2014 at 5:54 PM, Drew Farris <
> > drew.farris@gmail.com
> > >> >
> > >> > > > wrote:
> > >> > > >
> > >> > > > > Mike,
> > >> > > > >
> > >> > > > > So works just like upgrading from 1.5?
> > >> > > > >
> > >> > > > > (After 1.4 shutdown, install 1.6 and restart?)
> > >> > > > >
> > >> > > > > That sounds entirely reasonable.
> > >> > > > >
> > >> > > > > Drew
> > >> > > > > On Jun 17, 2014 10:52 PM, "Mike Drob" 
> > >> > > > > <madrob@cloudera.com>
> > >> wrote:
> > >> > > > >
> > >> > > > > > We initially tried to set it up as a stand-alone

> > >> > > > > > utility but
> > >> > > eventually
> > >> > > > > > gave up. In order to properly do the upgrade,
you
> concurrently
> > >> need
> > >> > > to
> > >> > > > > run
> > >> > > > > > whatever upgrade code concurrently with a tablet
server
> > hosting
> > >> > > > !METADATA
> > >> > > > > > and a tablet server that can replay WALs. We ended
up
> > >> duplicating a
> > >> > > lot
> > >> > > > > of
> > >> > > > > > logic already present in master before scrapping
that plan.
> An
> > >> > > > > alternative
> > >> > > > > > would have been to try to build on MAC, but that
was 
> > >> > > > > > also
> > >> > non-trivial
> > >> > > > to
> > >> > > > > > deploy, so we spliced the code into the existing

> > >> > > > > > upgrade
> path.
> > >> How
> > >> > do
> > >> > > > you
> > >> > > > > > feel about that, Drew?
> > >> > > > > >
> > >> > > > > >
> > >> > > > > > On Tue, Jun 17, 2014 at 8:57 PM, Drew Farris <
> > >> > drew.farris@gmail.com>
> > >> > > > > > wrote:
> > >> > > > > >
> > >> > > > > > > I'm +1 for a utility that would allow us
to go 
> > >> > > > > > > directly
> from
> > >> 1.4
> > >> > to
> > >> > > > > 1.6.
> > >> > > > > > >
> > >> > > > > > > In terms of a general policy, I suggest we
make this 
> > >> > > > > > > sort
> of
> > >> > > decision
> > >> > > > > on
> > >> > > > > > a
> > >> > > > > > > case by case basis. My unreasonably self-centered
> intuition
> > >> > > suggests
> > >> > > > > that
> > >> > > > > > > there may be some folks that want to go from
1.4 to 
> > >> > > > > > > 1.6
> now
> > >> due
> > >> > to
> > >> > > a
> > >> > > > > > > relatively short 1.5 cycle. The need to jump
multiple
> > versions
> > >> > like
> > >> > > > > might
> > >> > > > > > > not exist in the future.
> > >> > > > > > >
> > >> > > > > > >
> > >> > > > > > >
> > >> > > > > > > On Mon, Jun 16, 2014 at 5:24 PM, Sean Busbey
<
> > >> > busbey@cloudera.com>
> > >> > > > > > wrote:
> > >> > > > > > >
> > >> > > > > > > > In an effort to get more users off of
our now
> unsupported
> > >> 1.4
> > >> > > > > release,
> > >> > > > > > > > should we support upgrading directly
to 1.6 without
> going
> > >> > > through a
> > >> > > > > 1.5
> > >> > > > > > > > upgrade?
> > >> > > > > > > >
> > >> > > > > > > > More directly for those on user@: would
you be more
> > likely
> > >> to
> > >> > > > > upgrade
> > >> > > > > > > off
> > >> > > > > > > > of 1.4 if you could do so directly to
1.6?
> > >> > > > > > > >
> > >> > > > > > > > We have this working locally at Cloudera
as a part 
> > >> > > > > > > > of
> our
> > >> CDH
> > >> > > > > > integration
> > >> > > > > > > > (we shipped 1.4 and we're planning to
ship 1.6 next).
> > >> > > > > > > >
> > >> > > > > > > > We can get into implementation details
on a jira if
> > there's
> > >> > > > positive
> > >> > > > > > > > consensus, but the changes weren't very
complicated.
> > They're
> > >> > > mostly
> > >> > > > > > > >
> > >> > > > > > > > * forward porting and consolidating
some upgrade 
> > >> > > > > > > > code
> > >> > > > > > > > * additions to the README for instructions
> > >> > > > > > > >
> > >> > > > > > > > Personally, I can see the both sides
of the 
> > >> > > > > > > > argument. On
> > the
> > >> > plus
> > >> > > > > side,
> > >> > > > > > > > anything to get more users off of 1.4
is a good thing.
> On
> > >> the
> > >> > > > > negative
> > >> > > > > > > > side, it means we have the 1.4 related
upgrade code
> > sitting
> > >> in
> > >> > a
> > >> > > > > > > supported
> > >> > > > > > > > code branch longer.
> > >> > > > > > > >
> > >> > > > > > > > Thoughts?
> > >> > > > > > > >
> > >> > > > > > > > --
> > >> > > > > > > > Sean
> > >> > > > > > > >
> > >> > > > > > >
> > >> > > > > >
> > >> > > > >
> > >> > > >
> > >> > >
> > >> >
> > >> >
> > >> >
> > >> > --
> > >> > Sean
> > >> >
> > >>
> > >
> > >
> > >
> > > --
> > > Sean
> > >
> >
> >
> >
> > --
> > Sean
> >
>


Mime
View raw message