Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

or instance in version 0 > > > > >>> >>> > > >> >> >> sending an update> >> >> >> for > producer_byte_rate > > > > for userA would result in removing all > > > > >>> >>> > > >> >> >> previous data> >> >> >> and sett

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

;> >>> > > >> >> > replacement of the existing configuration, we will > > > break all > > > >>> >>> > > >> >> > of t

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

n to AlterConfigsOptions. > Callers > > can set > > >>> >>> > > >> >> > this boolean to true when they want the update to be > > >>> >>> > > >> >> > incremental. It should default to false so that old

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

st. > >>> >>> > > >> >> > > >>> >>> > > >> >> > Hmm. I don't think AlterOperation is necessary. If > the user > >>> >>> > > >>

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

t;>> >>> > > >> >> >> yes, I have/had a hard time finding a place for this >>> >>> > > >> >> >> operation. I think ADD> >> >> >> and DELETE should be on >>> >>> > >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

alue = null.> >> >> >> >>> > > >> >> > > >> >>> > > >> >> >> But also at the same time, SET is as you're suggesting >> >>> > > >> >> >> really >> >>> > >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

t;> >> > > >>> > > >> >> >> > >>> > > >> >> >> Keeping the --zookeeper option working is not infeasible > of > >>> > > >> >> >> course - and as> >> >&g

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

t; > > >> >> >> AlterConfigOptions (AdminClient sets the flag on the > >>> > > >> >> >> protocol, which will> >> >> >> be reflected after parsing in > >>> > > >> >> >> AdminManager). Also i

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

t; >> >> >> Keeping the --zookeeper option working is not infeasible of >>> > > >> >> >> course - and as> >> >> >> per the community's feedback it may

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

27;d need to call into the >> > > >> >> >> Scala code with >> > > >> >> >> reflection, which> >> >> >> could be quite complicated. >> > > >> >> > >> > > >&

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

tead of -- >> > >> >> > zookeeper).> >> >> > >> > >> >> > best, >> > >> >> > Colin >> > >> >> > >> > >> >> >> >> > >> >> >> Also rewrote the request semantics, hopefully

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

; >> >> > best, > > > >> >> > Colin > > > >> >> > > > > >> >> >> > > > >> >> >> Also rewrote the request semantics, hopefully it's more clear > > > >> >> >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

gt; >> >> >> > > > >> >> >> > If I'm reading the KIP right, it looks like the new > > >> >> >> > proposed verison of> >> >> >> > AlterConfigs sets an > > >> >> >> > Opera

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

onfig_name config_value config_operation> >> >> > >> >> >> > >> > > config_name => STRING > >> >> >> > > config_value => STRING > >> >> >> > > config_operation =>

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

;> >> > >> >> >> > Another consideration here is that we should do this in a compatible >> >> >> > fashion in AdminClient. Existing code that relies on the "set >> >> >> > e

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

rk Unless we believe that the existing AlterConfigs > >> >> > behavior is so broken that it should be changed, even in a > >> >> > compatibility-breaking way. > >> >> > > >> >> > Similarly, for other tools, we managed to support

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

gt; >> >> > >> >> > On Thu, Apr 26, 2018, at 01:47, Magnus Edenhill wrote: >> >> > > Hi Viktor, >> >> > > >> >> > > after speaking to Rajini it seems like this KIP will allow clients to >> >> > > perform incremental configur

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

;> > > I.e., that using Version 1 of AlterConfigs has the incremental > >> > > behaviour, > >> > > while > >> > > version 0 is atomic. > >> > > > >> > > Thanks, > >> > > Magnus > >> > > > >> > > &g

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

doing >> > > > it through kafka-run-class.sh. >> > > > In an ideal world I'd keep it in the current ConfigCommand command so >> > we >> > > > wouldn't mix the zookeeper and admin client configs. Perhaps I could >> > create >&g

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

The KIP proposes to remove the ability to configure using ZooKeeper. > > This > > > > > means we will no longer have the ability to start up a cluster with > > SCRAM > > > > > credentials since we first need to create SCRAM credentials befo

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

bility to start up a cluster with > SCRAM > > > > credentials since we first need to create SCRAM credentials before > > > brokers > > > > can start if the broker uses SCRAM for inter-broker communication > and we > > > > need SCRAM credentials fo

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

configure using ZooKeeper for these initialization scenarios and > > support > > > only AdminClient for dynamic updates? > > > > > > What do others think? > > > > > > Regards, > > > > > > Rajini > > > > > >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

> On Sun, Apr 15, 2018 at 10:28 AM, Ted Yu wrote: > > > > > +1 > > > Original message From: zhenya Sun > > > Date: 4/15/18 12:42 AM (GMT-08:00) To: dev > Cc: > > > dev Subject: Re: [VOTE] #2 KIP-248: Create New > > > C

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

d support > only AdminClient for dynamic updates? > > What do others think? > > Regards, > > Rajini > > On Sun, Apr 15, 2018 at 10:28 AM, Ted Yu wrote: > > > +1 > > Original message From: zhenya Sun > > Date: 4/15/18 12:42 AM (GMT-08:00)

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

and support only AdminClient for dynamic updates? What do others think? Regards, Rajini On Sun, Apr 15, 2018 at 10:28 AM, Ted Yu wrote: > +1 > Original message From: zhenya Sun > Date: 4/15/18 12:42 AM (GMT-08:00) To: dev Cc: > dev Subject: Re: [VOTE] #2 KIP-248

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

+1 Original message From: zhenya Sun Date: 4/15/18 12:42 AM (GMT-08:00) To: dev Cc: dev Subject: Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient non-binding +1 from my iphone！ On 04/15/2018 15:41, Attila Sasvári wrote: Thanks for

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

non-binding +1 from my iphone！ On 04/15/2018 15:41, Attila Sasvári wrote: Thanks for updating the KIP. +1 (non-binding) Viktor Somogyi ezt írta (időpont: 2018. ápr. 9., H 16:49): > Hi Magnus, > > Thanks for the heads up, added the endianness to the KIP. Here is the > current text: > > "Dou

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Thanks for updating the KIP. +1 (non-binding) Viktor Somogyi ezt írta (időpont: 2018. ápr. 9., H 16:49): > Hi Magnus, > > Thanks for the heads up, added the endianness to the KIP. Here is the > current text: > > "Double > A new type needs to be added to transfer quota values. Since the protocol

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Hi Magnus, Thanks for the heads up, added the endianness to the KIP. Here is the current text: "Double A new type needs to be added to transfer quota values. Since the protocol classes in Kafka already uses ByteBuffers it is logical to use their functionality for serializing doubles. The serializ

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Hi Viktor, since serialization of floats isn't as straight forward as integers, please specify the exact serialization format of DOUBLE in the protocol docs (e.g., IEEE 754), including endianness (big-endian please). This will help the non-java client ecosystem. Thanks, Magnus 2018-04-09 15:16

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Hi Attila, 1. It uses ByteBuffers, which in turn will use Double.doubleToLongBits to convert the double value to a long and that long will be written in the buffer. I'v updated the KIP with this. 2. Good idea, modified it. 3. During the discussion I remember we didn't really decide which one would

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Thanks for working on it Viktor. It looks good to me, but I have some questions: - I see a new type DOUBLE is used for quota_value , and it is not listed among the primitive types on the Kafka protocol guide. Can you add some more details? - I am not sure that using an environment (i.e. USE_OLD_CO

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

+1 (binding) Thanks for the KIP, Viktor! On Tue, Apr 3, 2018 at 8:25 AM, Peter Toth wrote: > +1 (non binding) > > Thanks Viktor. > Please remove "Unknown" AlterOperation type from Wire Format Types section > as you did from the AdminClient APIs section. > > > On Wed, Mar 21, 2018 at 5:41 PM, Vi

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

+1 (non binding) Thanks Viktor. Please remove "Unknown" AlterOperation type from Wire Format Types section as you did from the AdminClient APIs section. On Wed, Mar 21, 2018 at 5:41 PM, Viktor Somogyi wrote: > Hi Everyone, > > I've started a vote on KIP-248 >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

+1 （non-binding） | | zhenya Sun 邮箱：toke...@126.com | 签名由 网易邮箱大师 定制 On 03/29/2018 19:40, Sandor Murakozi wrote: +1 (non-binding) Thanks for the KIP, Viktor On Wed, Mar 21, 2018 at 5:41 PM, Viktor Somogyi wrote: > Hi Everyone, > > I've started a vote on KIP-248 >

Re: [VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

+1 (non-binding) Thanks for the KIP, Viktor On Wed, Mar 21, 2018 at 5:41 PM, Viktor Somogyi wrote: > Hi Everyone, > > I've started a vote on KIP-248 > ConfigCommand+That+Uses+The+New+AdminClient#KIP-248- > CreateNewConfi

[VOTE] #2 KIP-248: Create New ConfigCommand That Uses The New AdminClient

Hi Everyone, I've started a vote on KIP-248 a few weeks ago but at the time I got a couple more comments and i