Come on, I’ve done at least 2 talks on this one :) Producing counts to a topic is part of it, but that’s only part. So you count you have 100 messages in topic A. When you mirror topic A to another cluster, you have 99 messages. Where was your problem? Or worse, you have 100 messages, but one producer duplicated messages and another one lost messages. You need details about where the message came from in order to pinpoint problems when they happen. Source producer info, where it was produced into your infrastructure, and when it was produced. This requires you to add the information to the message.
And yes, you still need to maintain your clients. So maybe my original example was not the best. My thoughts on not wanting to be responsible for message formats stands, because that’s very much separate from the client. As you know, we have our own internal client library that can insert the right headers, and right now inserts the right audit information into the message fields. If they exist, and assuming the message is Avro encoded. What if someone wants to use JSON instead for a good reason? What if user X wants to encrypt messages, but user Y does not? Maintaining the client library is still much easier than maintaining the message formats. -Todd On Thu, Dec 1, 2016 at 6:21 PM, Gwen Shapira <g...@confluent.io> wrote: > Based on your last sentence, consider me convinced :) > > I get why headers are critical for Mirroring (you need tags to prevent > loops and sometimes to route messages to the correct destination). > But why do you need headers to audit? We are auditing by producing > counts to a side topic (and I was under the impression you do the > same), so we never need to modify the message. > > Another thing - after we added headers, wouldn't you be in the > business of making sure everyone uses them properly? Making sure > everyone includes the right headers you need, not using the header > names you intend to use, etc. I don't think the "policing" business > will ever go away. > > On Thu, Dec 1, 2016 at 5:25 PM, Todd Palino <tpal...@gmail.com> wrote: > > Got it. As an ops guy, I'm not very happy with the workaround. Avro means > > that I have to be concerned with the format of the messages in order to > run > > the infrastructure (audit, mirroring, etc.). That means that I have to > > handle the schemas, and I have to enforce rules about good formats. This > is > > not something I want to be in the business of, because I should be able > to > > run a service infrastructure without needing to be in the weeds of > dealing > > with customer data formats. > > > > Trust me, a sizable portion of my support time is spent dealing with > schema > > issues. I really would like to get away from that. Maybe I'd have more > time > > for other hobbies. Like writing. ;) > > > > -Todd > > > > On Thu, Dec 1, 2016 at 4:04 PM Gwen Shapira <g...@confluent.io> wrote: > > > >> I'm pretty satisfied with the current workarounds (Avro container > >> format), so I'm not too excited about the extra work required to do > >> headers in Kafka. I absolutely don't mind it if you do it... > >> I think the Apache convention for "good idea, but not willing to put > >> any work toward it" is +0.5? anyway, that's what I was trying to > >> convey :) > >> > >> On Thu, Dec 1, 2016 at 3:05 PM, Todd Palino <tpal...@gmail.com> wrote: > >> > Well I guess my question for you, then, is what is holding you back > from > >> > full support for headers? What’s the bit that you’re missing that has > you > >> > under a full +1? > >> > > >> > -Todd > >> > > >> > > >> > On Thu, Dec 1, 2016 at 1:59 PM, Gwen Shapira <g...@confluent.io> > wrote: > >> > > >> >> I know why people who support headers support them, and I've seen > what > >> >> the discussion is like. > >> >> > >> >> This is why I'm asking people who are against headers (especially > >> >> committers) what will make them change their mind - so we can get > this > >> >> part over one way or another. > >> >> > >> >> If I sound frustrated it is not at Radai, Jun or you (Todd)... I am > >> >> just looking for something concrete we can do to move the discussion > >> >> along to the yummy design details (which is the argument I really am > >> >> looking forward to). > >> >> > >> >> On Thu, Dec 1, 2016 at 1:53 PM, Todd Palino <tpal...@gmail.com> > wrote: > >> >> > So, Gwen, to your question (even though I’m not a committer)... > >> >> > > >> >> > I have always been a strong supporter of introducing the concept > of an > >> >> > envelope to messages, which headers accomplishes. The message key > is > >> >> > already an example of a piece of envelope information. By > providing a > >> >> means > >> >> > to do this within Kafka itself, and not relying on use-case > specific > >> >> > implementations, you make it much easier for components to > >> interoperate. > >> >> It > >> >> > simplifies development of all these things (message routing, > auditing, > >> >> > encryption, etc.) because each one does not have to reinvent the > >> wheel. > >> >> > > >> >> > It also makes it much easier from a client point of view if the > >> headers > >> >> are > >> >> > defined as part of the protocol and/or message format in general > >> because > >> >> > you can easily produce and consume messages without having to take > >> into > >> >> > account specific cases. For example, I want to route messages, but > >> >> client A > >> >> > doesn’t support the way audit implemented headers, and client B > >> doesn’t > >> >> > support the way encryption or routing implemented headers, so now > my > >> >> > application has to create some really fragile (my autocorrect just > >> tried > >> >> to > >> >> > make that “tragic”, which is probably appropriate too) code to > strip > >> >> > everything off, rather than just consuming the messages, picking > out > >> the > >> >> 1 > >> >> > or 2 headers it’s interested in, and performing its function. > >> >> > > >> >> > Honestly, this discussion has been going on for a long time, and > it’s > >> >> > always “Oh, you came up with 2 use cases, and yeah, those use cases > >> are > >> >> > real things that someone would want to do. Here’s an alternate way > to > >> >> > implement them so let’s not do headers.” If we have a few use cases > >> that > >> >> we > >> >> > actually came up with, you can be sure that over the next year > >> there’s a > >> >> > dozen others that we didn’t think of that someone would like to > do. I > >> >> > really think it’s time to stop rehashing this discussion and > instead > >> >> focus > >> >> > on a workable standard that we can adopt. > >> >> > > >> >> > -Todd > >> >> > > >> >> > > >> >> > On Thu, Dec 1, 2016 at 1:39 PM, Todd Palino <tpal...@gmail.com> > >> wrote: > >> >> > > >> >> >> C. per message encryption > >> >> >>> One drawback of this approach is that this significantly reduce > the > >> >> >>> effectiveness of compression, which happens on a set of > serialized > >> >> >>> messages. An alternative is to enable SSL for wire encryption and > >> rely > >> >> on > >> >> >>> the storage system (e.g. LUKS) for at rest encryption. > >> >> >> > >> >> >> > >> >> >> Jun, this is not sufficient. While this does cover the case of > >> removing > >> >> a > >> >> >> drive from the system, it will not satisfy most compliance > >> requirements > >> >> for > >> >> >> encryption of data as whoever has access to the broker itself > still > >> has > >> >> >> access to the unencrypted data. For end-to-end encryption you > need to > >> >> >> encrypt at the producer, before it enters the system, and decrypt > at > >> the > >> >> >> consumer, after it exits the system. > >> >> >> > >> >> >> -Todd > >> >> >> > >> >> >> > >> >> >> On Thu, Dec 1, 2016 at 1:03 PM, radai <radai.rosenbl...@gmail.com > > > >> >> wrote: > >> >> >> > >> >> >>> another big plus of headers in the protocol is that it would > enable > >> >> rapid > >> >> >>> iteration on ideas outside of core kafka and would reduce the > >> number of > >> >> >>> future wire format changes required. > >> >> >>> > >> >> >>> a lot of what is currently a KIP represents use cases that are > not > >> 100% > >> >> >>> relevant to all users, and some of them require rather invasive > wire > >> >> >>> protocol changes. a thing a good recent example of this is > kip-98. > >> >> >>> tx-utilizing traffic is expected to be a very small fraction of > >> total > >> >> >>> traffic and yet the changes are invasive. > >> >> >>> > >> >> >>> every such wire format change translates into painful and slow > >> >> adoption of > >> >> >>> new versions. > >> >> >>> > >> >> >>> i think a lot of functionality currently in KIPs could be "spun > out" > >> >> and > >> >> >>> implemented as opt-in plugins transmitting data over headers. > this > >> >> would > >> >> >>> keep the core wire format stable(r), core codebase smaller, and > >> avoid > >> >> the > >> >> >>> "burden of proof" thats sometimes required to prove a certain > >> feature > >> >> is > >> >> >>> useful enough for a wide-enough audience to warrant a wire format > >> >> change > >> >> >>> and code complexity additions. > >> >> >>> > >> >> >>> (to be clear - kip-98 goes beyond "mere" wire format changes and > im > >> not > >> >> >>> saying it could have been completely done with headers, but > >> >> exactly-once > >> >> >>> delivery certainly could) > >> >> >>> > >> >> >>> On Thu, Dec 1, 2016 at 11:20 AM, Gwen Shapira <g...@confluent.io > > > >> >> wrote: > >> >> >>> > >> >> >>> > On Thu, Dec 1, 2016 at 10:24 AM, radai < > >> radai.rosenbl...@gmail.com> > >> >> >>> wrote: > >> >> >>> > > "For use cases within an organization, one could always use > >> other > >> >> >>> > > approaches such as company-wise containers" > >> >> >>> > > this is what linkedin has traditionally done but there are > now > >> >> cases > >> >> >>> > (read > >> >> >>> > > - topics) where this is not acceptable. this makes headers > >> useful > >> >> even > >> >> >>> > > within single orgs for cases where one-container-fits-all > cannot > >> >> >>> apply. > >> >> >>> > > > >> >> >>> > > as for the particular use cases listed, i dont want this to > >> devolve > >> >> >>> to a > >> >> >>> > > discussion of particular use cases - i think its enough that > >> some > >> >> of > >> >> >>> them > >> >> >>> > > >> >> >>> > I think a main point of contention is that: We identified few > >> >> >>> > use-cases where headers are useful, do we want Kafka to be a > >> system > >> >> >>> > that supports those use-cases? > >> >> >>> > > >> >> >>> > For example, Jun said: > >> >> >>> > "Not sure how widely useful record-level lineage is though > since > >> the > >> >> >>> > overhead could > >> >> >>> > be significant." > >> >> >>> > > >> >> >>> > We know NiFi supports record level lineage. I don't think it > was > >> >> >>> > developed for lols, I think it is safe to assume that the NSA > >> needed > >> >> >>> > that functionality. We also know that certain financial > institutes > >> >> >>> > need to track tampering with records at a record level and > there > >> are > >> >> >>> > federal regulations that absolutely require this. They also > need > >> to > >> >> >>> > prove that routing apps that "touches" the messages and either > >> reads > >> >> >>> > or updates headers couldn't have possibly modified the payload > >> >> itself. > >> >> >>> > They use record level encryption to do that - apps can read and > >> >> >>> > (sometimes) modify headers but can't touch the payload. > >> >> >>> > > >> >> >>> > We can totally say "those are corner cases and not worth adding > >> >> >>> > headers to Kafka for", they should use a different pubsub > message > >> for > >> >> >>> > that (Nifi or one of the other 1000 that cater specifically to > the > >> >> >>> > financial industry). > >> >> >>> > > >> >> >>> > But this gets us into a catch 22: > >> >> >>> > If we discuss a specific use-case, someone can always say it > isn't > >> >> >>> > interesting enough for Kafka. If we discuss more general > trends, > >> >> >>> > others can say "well, we are not sure any of them really needs > >> >> headers > >> >> >>> > specifically. This is just hand waving and not interesting.". > >> >> >>> > > >> >> >>> > I think discussing use-cases in specifics is super important to > >> >> decide > >> >> >>> > implementation details for headers (my use-cases lean toward > >> >> numerical > >> >> >>> > keys with namespaces and object values, others differ), but I > >> think > >> >> we > >> >> >>> > need to answer the general "Are we going to have headers" > question > >> >> >>> > first. > >> >> >>> > > >> >> >>> > I'd love to hear from the other committers in the discussion: > >> >> >>> > What would it take to convince you that headers in Kafka are a > >> good > >> >> >>> > idea in general, so we can move ahead and try to agree on the > >> >> details? > >> >> >>> > > >> >> >>> > I feel like we keep moving the goal posts and this is truly > >> >> exhausting. > >> >> >>> > > >> >> >>> > For the record, I mildly support adding headers to Kafka > (+0.5?). > >> >> >>> > The community can continue to find workarounds to the issue and > >> there > >> >> >>> > are some benefits to keeping the message format and clients > >> simpler. > >> >> >>> > But I see the usefulness of headers to many use-cases and if we > >> can > >> >> >>> > find a good and generally useful way to add it to Kafka, it > will > >> make > >> >> >>> > Kafka easier to use for many - worthy goal in my eyes. > >> >> >>> > > >> >> >>> > > are interesting/feasible, but: > >> >> >>> > > A+B. i think there are use cases for polyglot topics. > >> especially if > >> >> >>> kafka > >> >> >>> > > is being used to "trunk" something else. > >> >> >>> > > D. multiple topics would make it harder to write portable > >> consumer > >> >> >>> code. > >> >> >>> > > partition remapping would mess with locality of consumption > >> >> >>> guarantees. > >> >> >>> > > E+F. a use case I see for lineage/metadata is > >> billing/chargeback. > >> >> for > >> >> >>> > that > >> >> >>> > > use case it is not enough to simply record the point of > origin, > >> but > >> >> >>> every > >> >> >>> > > replication stop (think mirror maker) must also add a record > to > >> >> form a > >> >> >>> > > "transit log". > >> >> >>> > > > >> >> >>> > > as for stream processing on top of kafka - i know samza has a > >> >> metadata > >> >> >>> > map > >> >> >>> > > which they carry around in addition to user values. headers > are > >> the > >> >> >>> > perfect > >> >> >>> > > fit for these things. > >> >> >>> > > > >> >> >>> > > > >> >> >>> > > > >> >> >>> > > On Wed, Nov 30, 2016 at 6:50 PM, Jun Rao <j...@confluent.io> > >> wrote: > >> >> >>> > > > >> >> >>> > >> Hi, Michael, > >> >> >>> > >> > >> >> >>> > >> In order to answer the first two questions, it would be > helpful > >> >> if we > >> >> >>> > could > >> >> >>> > >> identify 1 or 2 strong use cases for headers in the space > for > >> >> >>> > third-party > >> >> >>> > >> vendors. For use cases within an organization, one could > always > >> >> use > >> >> >>> > other > >> >> >>> > >> approaches such as company-wise containers to get around w/o > >> >> >>> headers. I > >> >> >>> > >> went through the use cases in the KIP and in Radai's wiki ( > >> >> >>> > >> https://cwiki.apache.org/confluence/display/KAFKA/A+ > >> >> >>> > Case+for+Kafka+Headers > >> >> >>> > >> ). > >> >> >>> > >> The following are the ones that that I understand and could > be > >> in > >> >> the > >> >> >>> > >> third-party use case category. > >> >> >>> > >> > >> >> >>> > >> A. content-type > >> >> >>> > >> It seems that in general, content-type should be set at the > >> topic > >> >> >>> level. > >> >> >>> > >> Not sure if mixing messages with different content types > >> should be > >> >> >>> > >> encouraged. > >> >> >>> > >> > >> >> >>> > >> B. schema id > >> >> >>> > >> Since the value is mostly useless without schema id, it > seems > >> that > >> >> >>> > storing > >> >> >>> > >> the schema id together with serialized bytes in the value is > >> >> better? > >> >> >>> > >> > >> >> >>> > >> C. per message encryption > >> >> >>> > >> One drawback of this approach is that this significantly > reduce > >> >> the > >> >> >>> > >> effectiveness of compression, which happens on a set of > >> serialized > >> >> >>> > >> messages. An alternative is to enable SSL for wire > encryption > >> and > >> >> >>> rely > >> >> >>> > on > >> >> >>> > >> the storage system (e.g. LUKS) for at rest encryption. > >> >> >>> > >> > >> >> >>> > >> D. cluster ID for mirroring across Kafka clusters > >> >> >>> > >> This is actually interesting. Today, to avoid introducing > >> cycles > >> >> when > >> >> >>> > doing > >> >> >>> > >> mirroring across data centers, one would either have to set > up > >> two > >> >> >>> Kafka > >> >> >>> > >> clusters (a local and an aggregate) per data center or > rename > >> >> topics. > >> >> >>> > >> Neither is ideal. With headers, the producer could tag each > >> >> message > >> >> >>> with > >> >> >>> > >> the producing cluster ID in the header. MirrorMaker could > then > >> >> avoid > >> >> >>> > >> mirroring messages to a cluster if they are tagged with the > >> same > >> >> >>> cluster > >> >> >>> > >> id. > >> >> >>> > >> > >> >> >>> > >> However, an alternative approach is to introduce sth like > >> >> >>> hierarchical > >> >> >>> > >> topic and store messages from different clusters in > different > >> >> >>> partitions > >> >> >>> > >> under the same topic. This approach avoids filtering out > >> unneeded > >> >> >>> data > >> >> >>> > and > >> >> >>> > >> makes offset preserving easier to support. It may make > >> compaction > >> >> >>> > trickier > >> >> >>> > >> though since the same key may show up in different > partitions. > >> >> >>> > >> > >> >> >>> > >> E. record-level lineage > >> >> >>> > >> For example, a source connector could store in the message > the > >> >> >>> metadata > >> >> >>> > >> (e.g. UUID) of the source record. Similarly, if a stream job > >> >> >>> transforms > >> >> >>> > >> messages from topic A to topic B, the library could include > the > >> >> >>> source > >> >> >>> > >> message offset in each of the transformed message in the > >> header. > >> >> Not > >> >> >>> > sure > >> >> >>> > >> how widely useful record-level lineage is though since the > >> >> overhead > >> >> >>> > could > >> >> >>> > >> be significant. > >> >> >>> > >> > >> >> >>> > >> F. auditing metadata > >> >> >>> > >> We could put things like clientId/host/user in the header in > >> each > >> >> >>> > message > >> >> >>> > >> for auditing. These metadata are really at the producer > level > >> >> though. > >> >> >>> > So, a > >> >> >>> > >> more efficient way is to only include a "producerId" per > >> message > >> >> and > >> >> >>> > send > >> >> >>> > >> the producerId -> metadata mapping independently. KIP-98 is > >> >> actually > >> >> >>> > >> proposing including such a producerId natively in the > message. > >> >> >>> > >> > >> >> >>> > >> So, overall, I not sure that I am fully convinced of the > strong > >> >> >>> > third-party > >> >> >>> > >> use cases of headers yet. Perhaps we could discuss a bit > more > >> to > >> >> make > >> >> >>> > one > >> >> >>> > >> or two really convincing use cases. > >> >> >>> > >> > >> >> >>> > >> Another orthogonal question is whether header should be > >> exposed > >> >> in > >> >> >>> > stream > >> >> >>> > >> processing systems such Kafka stream, Samza, and Spark > >> streaming. > >> >> >>> > >> Currently, those systems just deal with key/value pairs. > >> Should we > >> >> >>> > expose a > >> >> >>> > >> third thing header there too or somehow map header to key or > >> >> value? > >> >> >>> > >> > >> >> >>> > >> Thanks, > >> >> >>> > >> > >> >> >>> > >> Jun > >> >> >>> > >> > >> >> >>> > >> > >> >> >>> > >> On Tue, Nov 29, 2016 at 3:35 AM, Michael Pearce < > >> >> >>> michael.pea...@ig.com> > >> >> >>> > >> wrote: > >> >> >>> > >> > >> >> >>> > >> > I assume, that after a period of a week, that there is no > >> >> concerns > >> >> >>> now > >> >> >>> > >> > with points 1, and 2 and now we have agreement that > headers > >> are > >> >> >>> useful > >> >> >>> > >> and > >> >> >>> > >> > needed in Kafka. As such if put to a KIP vote, this > wouldn’t > >> be > >> >> a > >> >> >>> > reason > >> >> >>> > >> to > >> >> >>> > >> > reject. > >> >> >>> > >> > > >> >> >>> > >> > @ > >> >> >>> > >> > Ignacio on point 4). > >> >> >>> > >> > I think for purpose of getting this KIP moving past this, > we > >> can > >> >> >>> state > >> >> >>> > >> the > >> >> >>> > >> > key will be a 4 bytes space that can will be naturally > >> >> interpreted > >> >> >>> as > >> >> >>> > an > >> >> >>> > >> > Int32 (if namespacing is later wanted you can easily split > >> this > >> >> >>> into > >> >> >>> > two > >> >> >>> > >> > int16 spaces), from the wire protocol implementation this > >> makes > >> >> no > >> >> >>> > >> > difference I don’t believe. Is this reasonable to all? > >> >> >>> > >> > > >> >> >>> > >> > On 5) as per point 4 therefor happy we keep with 32 bits. > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > On 18/11/2016, 20:34, "ignacio.so...@gmail.com on behalf > of > >> >> >>> Ignacio > >> >> >>> > >> > Solis" <ignacio.so...@gmail.com on behalf of > iso...@igso.net > >> > > >> >> >>> wrote: > >> >> >>> > >> > > >> >> >>> > >> > Summary: > >> >> >>> > >> > > >> >> >>> > >> > 3) Yes - Header value as byte[] > >> >> >>> > >> > > >> >> >>> > >> > 4a) Int,Int - No > >> >> >>> > >> > 4b) Int - Yes > >> >> >>> > >> > 4c) String - Reluctant maybe > >> >> >>> > >> > > >> >> >>> > >> > 5) I believe the header system should take a single > >> int. I > >> >> >>> think > >> >> >>> > >> > 32bits is > >> >> >>> > >> > a good size, if you want to interpret this as to 16bit > >> >> numbers > >> >> >>> in > >> >> >>> > the > >> >> >>> > >> > layer > >> >> >>> > >> > above go right ahead. If somebody wants to argue for > 16 > >> >> bits > >> >> >>> or > >> >> >>> > 64 > >> >> >>> > >> > bits of > >> >> >>> > >> > header key space I would listen. > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > Discussion: > >> >> >>> > >> > Dividing the key space into sub_key_1 and sub_key_2 > >> makes no > >> >> >>> > sense to > >> >> >>> > >> > me at > >> >> >>> > >> > this layer. Are we going to start providing APIs to > get > >> all > >> >> >>> the > >> >> >>> > >> > sub_key_1s? or all the sub_key_2s? If there is no > >> >> >>> distinguishing > >> >> >>> > >> > functions > >> >> >>> > >> > that are applied to each one then they should be a > single > >> >> >>> value. > >> >> >>> > At > >> >> >>> > >> > this > >> >> >>> > >> > layer all we're doing is equality. > >> >> >>> > >> > If the above layer wants to interpret this as 2, 3 or > >> more > >> >> >>> values > >> >> >>> > >> > that's a > >> >> >>> > >> > different question. I personally think it's all one > >> >> keyspace > >> >> >>> > that is > >> >> >>> > >> > getting assigned using some structure, but if you > want to > >> >> >>> > sub-assign > >> >> >>> > >> > parts > >> >> >>> > >> > of it then that's fine. > >> >> >>> > >> > > >> >> >>> > >> > The same discussion applies to strings. If somebody > >> argued > >> >> for > >> >> >>> > >> > strings, > >> >> >>> > >> > would we be arguing to divide the strings with dots > ('.') > >> >> as a > >> >> >>> > >> > requirement? > >> >> >>> > >> > Would we want them to give us the different name > segments > >> >> >>> > separately? > >> >> >>> > >> > Would we be performing any actions on this key other > than > >> >> >>> > matching? > >> >> >>> > >> > > >> >> >>> > >> > Nacho > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > > >> >> >>> > >> > On Fri, Nov 18, 2016 at 9:30 AM, Michael Pearce < > >> >> >>> > >> michael.pea...@ig.com > >> >> >>> > >> > > > >> >> >>> > >> > wrote: > >> >> >>> > >> > > >> >> >>> > >> > > #jay #jun any concerns on 1 and 2 still? > >> >> >>> > >> > > > >> >> >>> > >> > > @all > >> >> >>> > >> > > To get this moving along a bit more I'd also like to > >> ask > >> >> to > >> >> >>> get > >> >> >>> > >> > clarity on > >> >> >>> > >> > > the below last points: > >> >> >>> > >> > > > >> >> >>> > >> > > 3) I believe we're all roughly happy with the header > >> value > >> >> >>> > being a > >> >> >>> > >> > byte[]? > >> >> >>> > >> > > > >> >> >>> > >> > > 4) I believe consensus has been for an namespace > based > >> int > >> >> >>> > approach > >> >> >>> > >> > > {int,int} for the key. Any objections if this is > what > >> we > >> >> go > >> >> >>> > with? > >> >> >>> > >> > > > >> >> >>> > >> > > 5) as we have if assumption in (4) is correct, > >> {int,int} > >> >> >>> keys. > >> >> >>> > >> > > Should both int's be int16 or int32? > >> >> >>> > >> > > I'm for them being int16(2 bytes) as combined is > space > >> of > >> >> >>> > 4bytes as > >> >> >>> > >> > per > >> >> >>> > >> > > original and gives plenty of combinations for the > >> >> >>> foreseeable, > >> >> >>> > and > >> >> >>> > >> > keeps > >> >> >>> > >> > > the overhead small. > >> >> >>> > >> > > > >> >> >>> > >> > > Do we see any benefit in another kip call to discuss > >> >> these at > >> >> >>> > all? > >> >> >>> > >> > > > >> >> >>> > >> > > Cheers > >> >> >>> > >> > > Mike > >> >> >>> > >> > > ________________________________________ > >> >> >>> > >> > > From: K Burstev <k.burs...@yandex.com> > >> >> >>> > >> > > Sent: Friday, November 18, 2016 7:07:07 AM > >> >> >>> > >> > > To: dev@kafka.apache.org > >> >> >>> > >> > > Subject: Re: [DISCUSS] KIP-82 - Add Record Headers > >> >> >>> > >> > > > >> >> >>> > >> > > For what it is worth also i agree. As a user: > >> >> >>> > >> > > > >> >> >>> > >> > > 1) Yes - Headers are worthwhile > >> >> >>> > >> > > 2) Yes - Headers should be a top level option > >> >> >>> > >> > > > >> >> >>> > >> > > 14.11.2016, 21:15, "Ignacio Solis" <iso...@igso.net > >: > >> >> >>> > >> > > > 1) Yes - Headers are worthwhile > >> >> >>> > >> > > > 2) Yes - Headers should be a top level option > >> >> >>> > >> > > > > >> >> >>> > >> > > > On Mon, Nov 14, 2016 at 9:16 AM, Michael Pearce < > >> >> >>> > >> > michael.pea...@ig.com> > >> >> >>> > >> > > > wrote: > >> >> >>> > >> > > > > >> >> >>> > >> > > >> Hi Roger, > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> The kip details/examples the original proposal > for > >> key > >> >> >>> > spacing > >> >> >>> > >> , > >> >> >>> > >> > not > >> >> >>> > >> > > the > >> >> >>> > >> > > >> new mentioned as per discussion namespace idea. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> We will need to update the kip, when we get > >> agreement > >> >> >>> this > >> >> >>> > is a > >> >> >>> > >> > better > >> >> >>> > >> > > >> approach (which seems to be the case if I have > >> >> understood > >> >> >>> > the > >> >> >>> > >> > general > >> >> >>> > >> > > >> feeling in the conversation) > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Re the variable ints, at very early stage we did > >> think > >> >> >>> about > >> >> >>> > >> > this. I > >> >> >>> > >> > > think > >> >> >>> > >> > > >> the added complexity for the saving isn't worth > it. > >> >> I'd > >> >> >>> > rather > >> >> >>> > >> go > >> >> >>> > >> > > with, if > >> >> >>> > >> > > >> we want to reduce overheads and size int16 > (2bytes) > >> >> keys > >> >> >>> as > >> >> >>> > it > >> >> >>> > >> > keeps it > >> >> >>> > >> > > >> simple. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> On the note of no headers, there is as per the > kip > >> as > >> >> we > >> >> >>> > use an > >> >> >>> > >> > > attribute > >> >> >>> > >> > > >> bit to denote if headers are present or not as > such > >> >> >>> > provides a > >> >> >>> > >> > zero > >> >> >>> > >> > > >> overhead currently if headers are not used. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> I think as radai mentions would be good first > if we > >> >> can > >> >> >>> get > >> >> >>> > >> > clarity if > >> >> >>> > >> > > do > >> >> >>> > >> > > >> we now have general consensus that (1) headers > are > >> >> >>> > worthwhile > >> >> >>> > >> and > >> >> >>> > >> > > useful, > >> >> >>> > >> > > >> and (2) we want it as a top level entity. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Just to state the obvious i believe (1) headers > are > >> >> >>> > worthwhile > >> >> >>> > >> > and (2) > >> >> >>> > >> > > >> agree as a top level entity. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Cheers > >> >> >>> > >> > > >> Mike > >> >> >>> > >> > > >> ________________________________________ > >> >> >>> > >> > > >> From: Roger Hoover <roger.hoo...@gmail.com> > >> >> >>> > >> > > >> Sent: Wednesday, November 9, 2016 9:10:47 PM > >> >> >>> > >> > > >> To: dev@kafka.apache.org > >> >> >>> > >> > > >> Subject: Re: [DISCUSS] KIP-82 - Add Record > Headers > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Sorry for going a little in the weeds but thanks > >> for > >> >> the > >> >> >>> > >> replies > >> >> >>> > >> > > regarding > >> >> >>> > >> > > >> varint. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Agreed that a prefix and {int, int} can be the > >> same. > >> >> It > >> >> >>> > doesn't > >> >> >>> > >> > look > >> >> >>> > >> > > like > >> >> >>> > >> > > >> that's what the KIP is saying the "Open" > section. > >> The > >> >> >>> > example > >> >> >>> > >> > shows > >> >> >>> > >> > > >> 2100001 > >> >> >>> > >> > > >> for New Relic and 210002 for App Dynamics > implying > >> >> that > >> >> >>> the > >> >> >>> > New > >> >> >>> > >> > Relic > >> >> >>> > >> > > >> organization will have only a single header id > to > >> work > >> >> >>> > with. Or > >> >> >>> > >> > is > >> >> >>> > >> > > 2100001 > >> >> >>> > >> > > >> a prefix? The main point of a namespace or > prefix > >> is > >> >> to > >> >> >>> > reduce > >> >> >>> > >> > the > >> >> >>> > >> > > >> overhead of config mapping or registration > >> depending > >> >> on > >> >> >>> how > >> >> >>> > >> > > >> namespaces/prefixes are managed. > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Would love to hear more feedback on the > >> higher-level > >> >> >>> > questions > >> >> >>> > >> > > though... > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Cheers, > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> Roger > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> On Wed, Nov 9, 2016 at 11:38 AM, radai < > >> >> >>> > >> > radai.rosenbl...@gmail.com> > >> >> >>> > >> > > wrote: > >> >> >>> > >> > > >> > >> >> >>> > >> > > >> > I think this discussion is getting a bit into > the > >> >> >>> weeds on > >> >> >>> > >> > technical > >> >> >>> > >> > > >> > implementation details. > >> >> >>> > >> > > >> > I'd liek to step back a minute and try and > >> establish > >> >> >>> > where we > >> >> >>> > >> > are in > >> >> >>> > >> > > the > >> >> >>> > >> > > >> > larger picture: > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > (re-wording nacho's last paragraph) > >> >> >>> > >> > > >> > 1. are we all in agreement that headers are a > >> >> >>> worthwhile > >> >> >>> > and > >> >> >>> > >> > useful > >> >> >>> > >> > > >> > addition to have? this was contested early on > >> >> >>> > >> > > >> > 2. are we all in agreement on headers as top > >> level > >> >> >>> entity > >> >> >>> > vs > >> >> >>> > >> > headers > >> >> >>> > >> > > >> > squirreled-away in V? > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > if there are still concerns around these #2 > >> points > >> >> >>> (#jay? > >> >> >>> > >> > #jun?)? > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > (and now back to our normal programming ...) > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > varints are nice. having said that, its adding > >> >> >>> complexity > >> >> >>> > >> (see > >> >> >>> > >> > > >> > https://github.com/addthis/ > >> >> stream-lib/blob/master/src/ > >> >> >>> > >> > > >> > main/java/com/clearspring/ > >> >> analytics/util/Varint.java > >> >> >>> > >> > > >> > as 1st google result) and would require anyone > >> >> writing > >> >> >>> > other > >> >> >>> > >> > clients > >> >> >>> > >> > > (C? > >> >> >>> > >> > > >> > Python? Go? Bash? ;-) ) to get/implement the > >> same, > >> >> and > >> >> >>> for > >> >> >>> > >> > relatively > >> >> >>> > >> > > >> > little gain (int vs string is order of > magnitude, > >> >> this > >> >> >>> > isnt). > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > int namespacing vs {int, int} namespacing are > >> >> basically > >> >> >>> > the > >> >> >>> > >> > same > >> >> >>> > >> > > thing - > >> >> >>> > >> > > >> > youre just namespacing an int64 and giving > people > >> >> while > >> >> >>> > 2^32 > >> >> >>> > >> > ranges > >> >> >>> > >> > > at a > >> >> >>> > >> > > >> > time. the part i like about this is letting > >> people > >> >> >>> have a > >> >> >>> > >> large > >> >> >>> > >> > > swath of > >> >> >>> > >> > > >> > numbers with one registration so they dont > have > >> to > >> >> come > >> >> >>> > back > >> >> >>> > >> > for > >> >> >>> > >> > > every > >> >> >>> > >> > > >> > single plugin/header they want to "reserve". > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > On Wed, Nov 9, 2016 at 11:01 AM, Roger Hoover > < > >> >> >>> > >> > > roger.hoo...@gmail.com> > >> >> >>> > >> > > >> > wrote: > >> >> >>> > >> > > >> > > >> >> >>> > >> > > >> > > Since some of the debate has been about > >> overhead + > >> >> >>> > >> > performance, I'm > >> >> >>> > >> > > >> > > wondering if we have considered a varint > >> encoding > >> >> ( > >> >> >>> > >> > > >> > > https://developers.google.com/ > >> >> protocol-buffers/docs/ > >> >> >>> > >> > > encoding#varints) > >> >> >>> > >> > > >> > for > >> >> >>> > >> > > >> > > the header length field (int32 in the > proposal) > >> >> and > >> >> >>> for > >> >> >>> > >> > header > >> >> >>> > >> > > ids? If > >> >> >>> > >> > > >> > you > >> >> >>> > >> > > >> > > don't use headers, the overhead would be a > >> single > >> >> >>> byte > >> >> >>> > and > >> >> >>> > >> > for each > >> >> >>> > >> > > >> > header > >> >> >>> > >> > > >> > > id < 128 would also need only a single byte? > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > On Wed, Nov 9, 2016 at 6:43 AM, radai < > >> >> >>> > >> > radai.rosenbl...@gmail.com> > >> >> >>> > >> > > >> > wrote: > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > > @magnus - and very dangerous (youre > >> essentially > >> >> >>> > >> > downloading and > >> >> >>> > >> > > >> > executing > >> >> >>> > >> > > >> > > > arbitrary code off the internet on your > >> servers > >> >> ... > >> >> >>> > bad > >> >> >>> > >> > idea > >> >> >>> > >> > > without > >> >> >>> > >> > > >> a > >> >> >>> > >> > > >> > > > sandbox, even with) > >> >> >>> > >> > > >> > > > > >> >> >>> > >> > > >> > > > as for it being a purely administrative > task > >> - i > >> >> >>> > >> disagree. > >> >> >>> > >> > > >> > > > > >> >> >>> > >> > > >> > > > i wish it would, really, because then my > >> earlier > >> >> >>> > point on > >> >> >>> > >> > the > >> >> >>> > >> > > >> > complexity > >> >> >>> > >> > > >> > > of > >> >> >>> > >> > > >> > > > the remapping process would be invalid, > but > >> at > >> >> >>> > linkedin, > >> >> >>> > >> > for > >> >> >>> > >> > > example, > >> >> >>> > >> > > >> > we > >> >> >>> > >> > > >> > > > (the team im in) run kafka as a service. > we > >> dont > >> >> >>> > really > >> >> >>> > >> > know > >> >> >>> > >> > > what our > >> >> >>> > >> > > >> > > users > >> >> >>> > >> > > >> > > > (developing applications that use kafka) > are > >> up > >> >> to > >> >> >>> at > >> >> >>> > any > >> >> >>> > >> > given > >> >> >>> > >> > > >> moment. > >> >> >>> > >> > > >> > > it > >> >> >>> > >> > > >> > > > is very possible (given the existance of > >> headers > >> >> >>> and a > >> >> >>> > >> > > corresponding > >> >> >>> > >> > > >> > > plugin > >> >> >>> > >> > > >> > > > ecosystem) for some application to "equip" > >> their > >> >> >>> > >> producers > >> >> >>> > >> > and > >> >> >>> > >> > > >> > consumers > >> >> >>> > >> > > >> > > > with the required plugin without us > knowing. > >> i > >> >> dont > >> >> >>> > mean > >> >> >>> > >> > to imply > >> >> >>> > >> > > >> thats > >> >> >>> > >> > > >> > > > bad, i just want to make the point that > its > >> not > >> >> as > >> >> >>> > simple > >> >> >>> > >> > > keeping it > >> >> >>> > >> > > >> in > >> >> >>> > >> > > >> > > > sync across a large-enough organization. > >> >> >>> > >> > > >> > > > > >> >> >>> > >> > > >> > > > > >> >> >>> > >> > > >> > > > On Wed, Nov 9, 2016 at 6:17 AM, Magnus > >> Edenhill > >> >> < > >> >> >>> > >> > > mag...@edenhill.se> > >> >> >>> > >> > > >> > > > wrote: > >> >> >>> > >> > > >> > > > > >> >> >>> > >> > > >> > > > > I think there is a piece missing in the > >> >> Strings > >> >> >>> > >> > discussion, > >> >> >>> > >> > > where > >> >> >>> > >> > > >> > > > > pro-Stringers > >> >> >>> > >> > > >> > > > > reason that by providing unique string > >> >> >>> identifiers > >> >> >>> > for > >> >> >>> > >> > each > >> >> >>> > >> > > header > >> >> >>> > >> > > >> > > > > everything will just > >> >> >>> > >> > > >> > > > > magically work for all parts of the > stream > >> >> >>> pipeline. > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > But the strings dont mean anything by > >> >> themselves, > >> >> >>> > and > >> >> >>> > >> > while we > >> >> >>> > >> > > >> could > >> >> >>> > >> > > >> > > > > probably envision > >> >> >>> > >> > > >> > > > > some auto plugin loader that downloads, > >> >> compiles, > >> >> >>> > links > >> >> >>> > >> > and > >> >> >>> > >> > > runs > >> >> >>> > >> > > >> > > plugins > >> >> >>> > >> > > >> > > > > on-demand > >> >> >>> > >> > > >> > > > > as soon as they're seen by a consumer, I > >> dont > >> >> >>> really > >> >> >>> > >> see > >> >> >>> > >> > a > >> >> >>> > >> > > use-case > >> >> >>> > >> > > >> > for > >> >> >>> > >> > > >> > > > > something > >> >> >>> > >> > > >> > > > > so dynamic (and fragile) in practice. > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > In the real world an application will be > >> >> >>> configured > >> >> >>> > >> with > >> >> >>> > >> > a set > >> >> >>> > >> > > of > >> >> >>> > >> > > >> > > plugins > >> >> >>> > >> > > >> > > > > to either add (producer) > >> >> >>> > >> > > >> > > > > or read (consumer) headers. > >> >> >>> > >> > > >> > > > > This is an administrative task based on > >> what > >> >> >>> > features a > >> >> >>> > >> > client > >> >> >>> > >> > > >> > > > > needs/provides and results in > >> >> >>> > >> > > >> > > > > some sort of configuration to enable and > >> >> >>> configure > >> >> >>> > the > >> >> >>> > >> > desired > >> >> >>> > >> > > >> > plugins. > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > Since this needs to be kept somewhat in > >> sync > >> >> >>> across > >> >> >>> > an > >> >> >>> > >> > > organisation > >> >> >>> > >> > > >> > > > (there > >> >> >>> > >> > > >> > > > > is no point in having producers > >> >> >>> > >> > > >> > > > > add headers no consumers will read, and > >> vice > >> >> >>> versa), > >> >> >>> > >> the > >> >> >>> > >> > added > >> >> >>> > >> > > >> > > complexity > >> >> >>> > >> > > >> > > > > of assigning an id namespace > >> >> >>> > >> > > >> > > > > for each plugin as it is being > configured > >> >> should > >> >> >>> be > >> >> >>> > >> > tolerable. > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > /Magnus > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > 2016-11-09 13:06 GMT+01:00 Michael > Pearce < > >> >> >>> > >> > > michael.pea...@ig.com>: > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > > Just following/catching up on what > seems > >> to > >> >> be > >> >> >>> an > >> >> >>> > >> > active > >> >> >>> > >> > > night :) > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > @Radai sorry if it may seem obvious > but > >> what > >> >> >>> does > >> >> >>> > MD > >> >> >>> > >> > stand > >> >> >>> > >> > > for? > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > My take on String vs Int: > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > I will state first I am pro Int (16 or > >> 32). > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > I do though playing devils advocate > see a > >> >> big > >> >> >>> plus > >> >> >>> > >> > with the > >> >> >>> > >> > > >> > argument > >> >> >>> > >> > > >> > > of > >> >> >>> > >> > > >> > > > > > String keys, this is around > integrating > >> >> into an > >> >> >>> > >> > existing > >> >> >>> > >> > > >> > eco-system. > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > As many other systems use String based > >> >> headers > >> >> >>> > >> (Flume, > >> >> >>> > >> > JMS) > >> >> >>> > >> > > it > >> >> >>> > >> > > >> > makes > >> >> >>> > >> > > >> > > > it > >> >> >>> > >> > > >> > > > > > much easier for these to be > >> >> >>> > incorporated/integrated > >> >> >>> > >> > into. > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > How with Int based headers could we > >> provide > >> >> a > >> >> >>> > >> > way/guidence to > >> >> >>> > >> > > >> make > >> >> >>> > >> > > >> > > this > >> >> >>> > >> > > >> > > > > > integration simple / easy with > transition > >> >> flows > >> >> >>> > over > >> >> >>> > >> to > >> >> >>> > >> > > kafka? > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > * tough luck buddy you're on your own > >> >> >>> > >> > > >> > > > > > * simply hash the string into int code > >> and > >> >> hope > >> >> >>> > for > >> >> >>> > >> no > >> >> >>> > >> > > collisions > >> >> >>> > >> > > >> > > (how > >> >> >>> > >> > > >> > > > to > >> >> >>> > >> > > >> > > > > > convert back though?) > >> >> >>> > >> > > >> > > > > > * http2 style as mentioned by nacho. > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > cheers, > >> >> >>> > >> > > >> > > > > > Mike > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > ______________________________ > __________ > >> >> >>> > >> > > >> > > > > > From: radai < > radai.rosenbl...@gmail.com> > >> >> >>> > >> > > >> > > > > > Sent: Wednesday, November 9, 2016 > 8:12 AM > >> >> >>> > >> > > >> > > > > > To: dev@kafka.apache.org > >> >> >>> > >> > > >> > > > > > Subject: Re: [DISCUSS] KIP-82 - Add > >> Record > >> >> >>> Headers > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > thinking about it some more, the best > >> way to > >> >> >>> > transmit > >> >> >>> > >> > the > >> >> >>> > >> > > header > >> >> >>> > >> > > >> > > > > remapping > >> >> >>> > >> > > >> > > > > > data to consumers would be to put it > in > >> the > >> >> MD > >> >> >>> > >> response > >> >> >>> > >> > > payload, > >> >> >>> > >> > > >> so > >> >> >>> > >> > > >> > > > maybe > >> >> >>> > >> > > >> > > > > > it should be discussed now. > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > On Wed, Nov 9, 2016 at 12:09 AM, > radai < > >> >> >>> > >> > > >> radai.rosenbl...@gmail.com > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > > > wrote: > >> >> >>> > >> > > >> > > > > > > >> >> >>> > >> > > >> > > > > > > im not opposed to the idea of > namespace > >> >> >>> mapping. > >> >> >>> > >> all > >> >> >>> > >> > im > >> >> >>> > >> > > saying > >> >> >>> > >> > > >> is > >> >> >>> > >> > > >> > > > that > >> >> >>> > >> > > >> > > > > > its > >> >> >>> > >> > > >> > > > > > > not part of the "mvp" and, since it > >> >> requires > >> >> >>> no > >> >> >>> > >> wire > >> >> >>> > >> > format > >> >> >>> > >> > > >> > change, > >> >> >>> > >> > > >> > > > can > >> >> >>> > >> > > >> > > > > > > always be added later. > >> >> >>> > >> > > >> > > > > > > also, its not as simple as just > >> >> configuring > >> >> >>> MM > >> >> >>> > to > >> >> >>> > >> do > >> >> >>> > >> > the > >> >> >>> > >> > > >> > transform: > >> >> >>> > >> > > >> > > > > lets > >> >> >>> > >> > > >> > > > > > > say i've implemented large message > >> >> support as > >> >> >>> > >> > {666,1} and > >> >> >>> > >> > > on > >> >> >>> > >> > > >> some > >> >> >>> > >> > > >> > > > > mirror > >> >> >>> > >> > > >> > > > > > > target cluster its been remapped to > >> >> {999,1}. > >> >> >>> the > >> >> >>> > >> > consumer > >> >> >>> > >> > > >> plugin > >> >> >>> > >> > > >> > > code > >> >> >>> > >> > > >> > > > > > would > >> >> >>> > >> > > >> > > > > > > also need to be told to look for the > >> large > >> >> >>> > message > >> >> >>> > >> > "part X > >> >> >>> > >> > > of > >> >> >>> > >> > > >> Y" > >> >> >>> > >> > > >> > > > header > >> >> >>> > >> > > >> > > > > > > under {999,1}. doable, but tricky. > >> >> >>> > >> > > >> > > > > > > > >> >> >>> > >> > > >> > > > > > > On Tue, Nov 8, 2016 at 10:29 PM, > Gwen > >> >> >>> Shapira < > >> >> >>> > >> > > >> g...@confluent.io > >> >> >>> > >> > > >> > > > >> >> >>> > >> > > >> > > > > wrote: > >> >> >>> > >> > > >> > > > > > > > >> >> >>> > >> > > >> > > > > > >> While you can do whatever you want > >> with a > >> >> >>> > >> namespace > >> >> >>> > >> > and > >> >> >>> > >> > > your > >> >> >>> > >> > > >> > code, > >> >> >>> > >> > > >> > > > > > >> what I'd expect is for each app to > >> >> >>> namespaces > >> >> >>> > >> > > configurable... > >> >> >>> > >> > > >> > > > > > >> > >> >> >>> > >> > > >> > > > > > >> So if I accidentally used 666 for > my > >> HR > >> >> >>> > >> department, > >> >> >>> > >> > and > >> >> >>> > >> > > still > >> >> >>> > >> > > >> > want > >> >> >>> > >> > > >> > > > to > >> >> >>> > >> > > >> > > > > > >> run RadaiApp, I can config > >> "namespace=42" > >> >> >>> for > >> >> >>> > >> > RadaiApp and > >> >> >>> > >> > > >> > > > everything > >> >> >>> > >> > > >> > > > > > >> will look normal. > >> >> >>> > >> > > >> > > > > > >> > >> >> >>> > >> > > >> > > > > > >> This means you only need to sync > usage > >> >> >>> inside > >> >> >>> > your > >> >> >>> > >> > own > >> >> >>> > >> > > >> > > organization. > >> >> >>> > >> > > >> > > > > > >> Still hard, but somewhat easier > than > >> >> syncing > >> >> >>> > with > >> >> >>> > >> > the > >> >> >>> > >> > > entire > >> >> >>> > >> > > >> > > world. > >> >> >>> > >> > > >> > > > > > >> > >> >> >>> > >> > > >> > > > > > >> On Tue, Nov 8, 2016 at 10:07 PM, > >> radai < > >> >> >>> > >> > > >> > > radai.rosenbl...@gmail.com> > >> >> >>> > >> > > >> > > > > > >> wrote: > >> >> >>> > >> > > >> > > > > > >> > and we can start with {namespace, > >> id} > >> >> and > >> >> >>> no > >> >> >>> > >> > re-mapping > >> >> >>> > >> > > >> > support > >> >> >>> > >> > > >> > > > and > >> >> >>> > >> > > >> > > > > > >> always > >> >> >>> > >> > > >> > > > > > >> > add it later on if/when > collisions > >> >> >>> actually > >> >> >>> > >> > happen (i > >> >> >>> > >> > > dont > >> >> >>> > >> > > >> > think > >> >> >>> > >> > > >> > > > > > they'd > >> >> >>> > >> > > >> > > > > > >> be > >> >> >>> > >> > > >> > > > > > >> > a problem). > >> >> >>> > >> > > >> > > > > > >> > > >> >> >>> > >> > > >> > > > > > >> > every interested party (so orgs > or > >> >> >>> > individuals) > >> >> >>> > >> > could > >> >> >>> > >> > > then > >> >> >>> > >> > > >> > > > register > >> >> >>> > >> > > >> > > > > a > >> >> >>> > >> > > >> > > > > > >> > prefix (0 = reserved, 1 = > confluent > >> ... > >> >> >>> 666 > >> >> >>> > = me > >> >> >>> > >> > :-) ) > >> >> >>> > >> > > and > >> >> >>> > >> > > >> do > >> >> >>> > >> > > >> > > > > whatever > >> >> >>> > >> > > >> > > > > > >> with > >> >> >>> > >> > > >> > > > > > >> > the 2nd ID - so once linkedin > >> >> registers, > >> >> >>> say > >> >> >>> > 3, > >> >> >>> > >> > then > >> >> >>> > >> > > >> linkedin > >> >> >>> > >> > > >> > > devs > >> >> >>> > >> > > >> > > > > are > >> >> >>> > >> > > >> > > > > > >> free > >> >> >>> > >> > > >> > > > > > >> > to use {3, *} with a reasonable > >> >> >>> expectation > >> >> >>> > to > >> >> >>> > >> to > >> >> >>> > >> > > collide > >> >> >>> > >> > > >> with > >> >> >>> > >> > > >> > > > > > anything > >> >> >>> > >> > > >> > > > > > >> > else. further partitioning of > that * > >> >> >>> becomes > >> >> >>> > >> > linkedin's > >> >> >>> > >> > > >> > problem, > >> >> >>> > >> > > >> > > > but > >> >> >>> > >> > > >> > > > > > the > >> >> >>> > >> > > >> > > > > > >> > "upstream registration" of a > >> namespace > >> >> >>> only > >> >> >>> > has > >> >> >>> > >> to > >> >> >>> > >> > > happen > >> >> >>> > >> > > >> > once. > >> >> >>> > >> > > >> > > > > > >> > > >> >> >>> > >> > > >> > > > > > >> > On Tue, Nov 8, 2016 at 9:03 PM, > >> James > >> >> >>> Cheng < > >> >> >>> > >> > > >> > > wushuja...@gmail.com > >> >> >>> > >> > > >> > > > > > >> >> >>> > >> > > >> > > > > > >> wrote: > >> >> >>> > >> > > >> > > > > > >> > > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> > On Nov 8, 2016, at 5:54 PM, > Gwen > >> >> >>> Shapira < > >> >> >>> > >> > > >> > g...@confluent.io> > >> >> >>> > >> > > >> > > > > > wrote: > >> >> >>> > >> > > >> > > > > > >> >> > > >> >> >>> > >> > > >> > > > > > >> >> > Thank you so much for this > clear > >> and > >> >> >>> fair > >> >> >>> > >> > summary of > >> >> >>> > >> > > the > >> >> >>> > >> > > >> > > > > arguments. > >> >> >>> > >> > > >> > > > > > >> >> > > >> >> >>> > >> > > >> > > > > > >> >> > I'm in favor of ints. Not a > >> >> >>> deal-breaker, > >> >> >>> > but > >> >> >>> > >> > in > >> >> >>> > >> > > favor. > >> >> >>> > >> > > >> > > > > > >> >> > > >> >> >>> > >> > > >> > > > > > >> >> > Even more in favor of Magnus's > >> >> >>> > decentralized > >> >> >>> > >> > > suggestion > >> >> >>> > >> > > >> > with > >> >> >>> > >> > > >> > > > > > Roger's > >> >> >>> > >> > > >> > > > > > >> >> > tweak: add a namespace for > >> headers. > >> >> >>> This > >> >> >>> > will > >> >> >>> > >> > allow > >> >> >>> > >> > > each > >> >> >>> > >> > > >> > app > >> >> >>> > >> > > >> > > to > >> >> >>> > >> > > >> > > > > > just > >> >> >>> > >> > > >> > > > > > >> >> > use whatever IDs it wants > >> >> internally, > >> >> >>> and > >> >> >>> > >> then > >> >> >>> > >> > let > >> >> >>> > >> > > the > >> >> >>> > >> > > >> > admin > >> >> >>> > >> > > >> > > > > > >> deploying > >> >> >>> > >> > > >> > > > > > >> >> > the app figure out an > available > >> >> >>> namespace > >> >> >>> > ID > >> >> >>> > >> > for the > >> >> >>> > >> > > app > >> >> >>> > >> > > >> to > >> >> >>> > >> > > >> > > > live > >> >> >>> > >> > > >> > > > > > in. > >> >> >>> > >> > > >> > > > > > >> >> > So > io.confluent.schema-registry > >> can > >> >> be > >> >> >>> > >> > namespace > >> >> >>> > >> > > 0x01 on > >> >> >>> > >> > > >> my > >> >> >>> > >> > > >> > > > > > >> deployment > >> >> >>> > >> > > >> > > > > > >> >> > and 0x57 on yours, and the > poor > >> guys > >> >> >>> > >> > developing the > >> >> >>> > >> > > app > >> >> >>> > >> > > >> > don't > >> >> >>> > >> > > >> > > > > need > >> >> >>> > >> > > >> > > > > > to > >> >> >>> > >> > > >> > > > > > >> >> > worry about that. > >> >> >>> > >> > > >> > > > > > >> >> > > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> Gwen, if I understand your > example > >> >> >>> right, an > >> >> >>> > >> > > application > >> >> >>> > >> > > >> > > deployer > >> >> >>> > >> > > >> > > > > > might > >> >> >>> > >> > > >> > > > > > >> >> decide to use 0x01 in one > >> deployment, > >> >> and > >> >> >>> > that > >> >> >>> > >> > means > >> >> >>> > >> > > that > >> >> >>> > >> > > >> > once > >> >> >>> > >> > > >> > > > the > >> >> >>> > >> > > >> > > > > > >> message > >> >> >>> > >> > > >> > > > > > >> >> is written into the broker, it > >> will be > >> >> >>> > saved on > >> >> >>> > >> > the > >> >> >>> > >> > > broker > >> >> >>> > >> > > >> > with > >> >> >>> > >> > > >> > > > > that > >> >> >>> > >> > > >> > > > > > >> >> specific namespace (0x01). > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> If you were to mirror that > message > >> >> into > >> >> >>> > another > >> >> >>> > >> > > cluster, > >> >> >>> > >> > > >> the > >> >> >>> > >> > > >> > > 0x01 > >> >> >>> > >> > > >> > > > > > would > >> >> >>> > >> > > >> > > > > > >> >> accompany the message, right? > What > >> if > >> >> the > >> >> >>> > >> > deployers of > >> >> >>> > >> > > the > >> >> >>> > >> > > >> > same > >> >> >>> > >> > > >> > > > app > >> >> >>> > >> > > >> > > > > > in > >> >> >>> > >> > > >> > > > > > >> the > >> >> >>> > >> > > >> > > > > > >> >> other cluster uses 0x57? They > won't > >> >> >>> > understand > >> >> >>> > >> > each > >> >> >>> > >> > > other? > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> I'm not sure that's an avoidable > >> >> >>> problem. I > >> >> >>> > >> > think it > >> >> >>> > >> > > simply > >> >> >>> > >> > > >> > > means > >> >> >>> > >> > > >> > > > > > that > >> >> >>> > >> > > >> > > > > > >> in > >> >> >>> > >> > > >> > > > > > >> >> order to share data, you have to > >> also > >> >> >>> have a > >> >> >>> > >> > shared > >> >> >>> > >> > > (agreed > >> >> >>> > >> > > >> > > upon) > >> >> >>> > >> > > >> > > > > > >> >> understanding of what the > >> namespaces > >> >> >>> mean. > >> >> >>> > >> Which > >> >> >>> > >> > I > >> >> >>> > >> > > think > >> >> >>> > >> > > >> > makes > >> >> >>> > >> > > >> > > > > sense, > >> >> >>> > >> > > >> > > > > > >> >> because the alternate (sharing > >> >> *nothing* > >> >> >>> at > >> >> >>> > >> all) > >> >> >>> > >> > would > >> >> >>> > >> > > mean > >> >> >>> > >> > > >> > > that > >> >> >>> > >> > > >> > > > > > there > >> >> >>> > >> > > >> > > > > > >> >> would be no way to understand > each > >> >> other. > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> -James > >> >> >>> > >> > > >> > > > > > >> >> > >> >> >>> > >> > > >> > > > > > >> >> > Gwen > >> >> >>> > >> > > >> > > > > > >> >> > > >> >> >>> > >> > > >> > > > > > >> >> > On Tue, Nov 8, 2016 at 4:23 > PM, > >> >> radai < > >> >> >>> > >> > > >> > > > > radai.rosenbl...@gmail.com> > >> >> >>> > >> > > >> > > > > > >> >> wrote: > >> >> >>> > >> > > >> > > > > > >> >> >> +1 for sean's document. it > >> covers > >> >> >>> pretty > >> >> >>> > >> much > >> >> >>> > >> > all > >> >> >>> > >> > > the > >> >> >>> > >> > > >> > > > trade-offs > >> >> >>> > >> > > >> > > > > > and > >> >> >>> > >> > > >> > > > > > >> >> >> provides concrete figures to > >> argue > >> >> >>> about > >> >> >>> > :-) > >> >> >>> > >> > > >> > > > > > >> >> >> (nit-picking - used the same > >> xkcd > >> >> >>> twice, > >> >> >>> > >> also > >> >> >>> > >> > trove > >> >> >>> > >> > > has > >> >> >>> > >> > > >> > been > >> >> >>> > >> > > >> > > > > > >> superceded > >> > > > > > -- > Gwen Shapira > Product Manager | Confluent > 650.450.2760 | @gwenshap > Follow us: Twitter | blog > -- *Todd Palino* Staff Site Reliability Engineer Data Infrastructure Streaming linkedin.com/in/toddpalino
