+1 (non-binding) Thanks for improving the CLI tool!
Best regards, Kuan-Po Tseng On Wed, Aug 20, 2025 at 1:23 AM Lianet M. <liane...@gmail.com> wrote: > Thanks for the updates Andrew from the discussion thread, lgtm. Looking > forward to using this! > > +1 (binding) > > Lianet > > On Wed, Aug 13, 2025 at 10:31 AM Andrew Schofield < > andrew_schofield_j...@outlook.com> wrote: > > > Hi Lianet, > > Thanks for the comments. > > > > LM1: I am horrified to be told that I had missed Chia-Ping's comments. > > I'll rectify > > that today. I think that we probably should add `--consumer-property` to > > those > > other tools. > > > > LM2: I did mention that uses of the command-line tools affected by this > > KIP would be changed over, but I was not explicit about system tests. I > > have > > change the KIP to be explicit about this. > > > > > > I would leave the completion of any precedence changes to KIP-1137. This > > KIP > > doesn't change precedence of configs for the existing tools. > > > > Thanks, > > Andrew > > ________________________________________ > > From: Lianet M. <liane...@gmail.com> > > Sent: 13 August 2025 13:39 > > To: dev@kafka.apache.org <dev@kafka.apache.org> > > Subject: Re: [VOTE] KIP-1147: Improve consistency of command-line > arguments > > > > Hi Andrew, just a couple of comments left: > > > > LM1: since we’re changing the option to pass explicit properties to the > > producer perf tool, should we also align the consumer and share perf > tools > > and introduce the option consistently? (--consumer-property). This was > also > > raised by Chia btw, but it seems it hasn't been addressed. > > > > LM2: should the test plan include not only unit tests but also system > > tests? (these tools are heavily used there, so would be a good > validation) > > > > There are also some open comments from Tommi above. Regarding parameters > > precedence, my take is that it makes total sense to review but we could > > address it with the existing KIP-1137 and limit the scope of this one. > > Thoughts? > > > > Thanks! > > > > Lianet > > > > > > > > On Thu, Aug 7, 2025 at 3:17 AM Tommi Vainikainen > <tvain...@aiven.io.invalid > > > > > wrote: > > > > > I'd like to ask for a clarification on the precedence of > configurations. > > > Most concretely this KIP adds --bootstrap-server > > > for kafka-producer-perf-test.sh. Almost all tools have > --bootstrap-server > > > as *required* argument, but kafka-producer-perf-test has gone without > > > --bootstrap-server command line argument and instead reading bootstrap > > > server either from properties file or from producer property command > line > > > argument. > > > > > > How about also defining standard precedence of different ways of > passing > > > e.g. bootstrap server, but also some other properties (in cases like > > > --consumer-property or --producer-property arguments)? And what about > > > making command line --bootstrap-server argument optional in cases where > > the > > > bootstrap server is defined in a property file passed with > > > --command-config? > > > > > > There was a little bit related thread about precedence issue in May > > > https://lists.apache.org/thread/nzov19dm1xckp1l8pl2zkcn6w4zgp8rb > > > > > > On Mon, Jul 21, 2025 at 11:49 AM Andrew Schofield < > > > andrew_schofield_j...@outlook.com> wrote: > > > > > > > Hi, > > > > I would like to start a vote for KIP-1147 which improves the > > consistency > > > of > > > > command-line arguments of the various Kafka CLI tools. > > > > > > > > > > > > > > > > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-1147%3A+Improve+consistency+of+command-line+arguments > > > > > > > > Thanks, > > > > Andrew > > > > > >
