kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jay Kreps <jay.kr...@gmail.com>
Subject Re: Transactional writing
Date Fri, 26 Oct 2012 18:08:20 GMT
This is an important feature and I am interested in helping out in the
design and implementation, though I am working on 0.8 features for the next
month so I may not be of too much use. I have thought a little bit about
this, but I am not yet sure of the best approach.

Here is a specific use case I think is important to address: consider a
case where you are doing processing of one or more streams and producing an
output stream. This processing may involve some kind of local state (say
counters or other local aggregation intermediate state). This is a common
scenario. The problem is to give reasonable semantics to this computation
in the presence of failures. The processor effectively has a
position/offset in each of its input streams as well as whatever local
state. The problem is that if this process fails it needs to restore to a
state that matches the last produced messages. There are several solutions
to this problem. One is to make the output somehow idempotent, this will
solve some cases but is not a general solution as many things cannot be
made idempotent easily.

I think the two proposals you give outline a couple of basic approaches:
1. Store the messages on the server somewhere but don't add them to the log
until the commit call
2. Store the messages in the log but don't make them available to the
consumer until the commit call
Another option you didn't mention:

I can give several subtleties to these approaches.

One advantage of the second approach is that messages are in the log and
can be available for reading or not. This makes it possible to support a
kind of "dirty read" that allows the consumer to specify whether they want
to immediately see all messages with low latency but potentially see
uncommitted messages or only see committed messages.

The problem with the second approach at least in the way you describe it is
that you have to lock the log until the commit occurs otherwise you can't
roll back (because otherwise someone else may have appended their own
messages and you can't truncate the log). This would have all the problems
of remote locks. I think this might be a deal-breaker.

Another variation on the second approach would be the following: have each
producer maintain an id and generation number. Keep a schedule of valid
offset/id/generation numbers on the broker and only hand these out. This
solution would support non-blocking multi-writer appends but requires more
participation from the producer (i.e. getting a generation number and id).



On Thu, Oct 25, 2012 at 7:04 PM, Tom Brown <tombrown52@gmail.com> wrote:

> I have come up with two different possibilities, both with different
> trade-offs.
> The first would be to support "true" transactions by writing
> transactional data into a temporary file and then copy it directly to
> the end of the partition when the commit command is created. The
> upside to this approach is that individual transactions can be larger
> than a single batch, and more than one producer could conduct
> transactions at once. The downside is the extra IO involved in writing
> it and reading it from disk an extra time.
> The second would be to allow any number of messages to be appended to
> a topic, but not move the "end of topic" offset until the commit was
> received. If a rollback was received, or the producer timed out, the
> partition could be truncated at the most recently recognized "end of
> topic" offset. The upside is that there is very little extra IO (only
> to store the official "end of topic" metadata), and it seems like it
> should be easy to implement. The downside is that this the
> "transaction" feature is incompatible with anything but a single
> producer per partition.
> I am interested in your thoughts on these.
> --Tom
> On Thu, Oct 25, 2012 at 9:31 PM, Philip O'Toole <philip@loggly.com> wrote:
> > On Thu, Oct 25, 2012 at 06:19:04PM -0700, Neha Narkhede wrote:
> >> The closest concept of transaction on the publisher side, that I can
> >> think of, is using batch of messages in a single call to the
> >> synchronous producer.
> >>
> >> Precisely, you can configure a Kafka producer to use the "sync" mode
> >> and batch messages that require transactional guarantees in a
> >> single send() call. That will ensure that either all the messages in
> >> the batch are sent or none.
> >
> > This is an interesting feature -- something I wasn't aware of. Still it
> > doesn't solve the problem *completely*. As many people realise, it's
> still
> > possible for the batch of messages to get into Kafka fine, but the ack
> from
> > Kafka to be lost on its way back to the Producer. In that case the
> Producer
> > erroneously believes the messages didn't get in, and might re-send them.
> >
> > You guys *haven't* solved that issue, right? I believe you write about
> it on
> > the Kafka site.
> >
> >>
> >> Thanks,
> >> Neha
> >>
> >> On Thu, Oct 25, 2012 at 2:44 PM, Tom Brown <tombrown52@gmail.com>
> wrote:
> >> > Is there an accepted, or recommended way to make writes to a Kafka
> >> > queue idempotent, or within a transaction?
> >> >
> >> > I can configure my system such that each queue has exactly one
> producer.
> >> >
> >> > (If there are no accepted/recommended ways, I have a few ideas I would
> >> > like to propose. I would also be willing to implement them if needed)
> >> >
> >> > Thanks in advance!
> >> >
> >> > --Tom
> >
> > --
> > Philip O'Toole
> >
> > Senior Developer
> > Loggly, Inc.
> > San Francisco, Calif.
> > www.loggly.com
> >
> > Come join us!
> > http://loggly.com/company/careers/

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