On Tuesday, January 24, 2017 at 10:21:20 AM UTC-8, Mark D. Roth wrote: > > On Mon, Jan 23, 2017 at 2:54 PM, 'Carl Mastrangelo' via grpc.io < > [email protected] <javascript:>> wrote: > >> >> >> On Monday, January 23, 2017 at 7:31:51 AM UTC-8, Mark D. Roth wrote: >>> >>> On Fri, Jan 20, 2017 at 9:47 AM, 'Carl Mastrangelo' via grpc.io < >>> [email protected]> wrote: >>> >>>> Initial thoughts: >>>> >>>> * percentage needs to be declared to be an integer (as opposed to >>>> number). This will make it consistent internally and externally. >>>> >>> >>> Done. (There's no reason to require that these fields be consistent >>> internally and externally, since they're each going to be read by >>> independent resolver implementations. But I agree that integer makes more >>> sense.) >>> >>> >>>> * TXT records are limitted to ASCII chars. What will happen if the >>>> method name, programming language, or load balancing policy is not pure >>>> ascii? >>>> >>> >>> Good question. As a data point, do we know if protobuf allows non-ASCII >>> chars in service or method names? >>> >> >> I don't think protobuf allows it for method or service names that it >> generates, but protobuf may not be used as the IDL. Also, the other fields >> wouldn't be affected by protobuf restrictions. >> > > Right. I was just asking about this as one data point. > > The simplest solution would be to impose a restriction that these fields > all have to be ASCII. Are there any problems with that approach? >
I don't think there are any problems, just would like it called out in the spec. The restriction isn't obvious from the spec. > > >> >> >>> >>> >>>> * Are TXT records for a superdomain applicable? For example, if there >>>> was a SC for foo.bar.com, but not sub.foo.bar.com, does it apply? >>>> >>> >>> No, the name has to exactly match the server name given to the client. >>> Otherwise, we'd need to make a bunch of additional DNS lookups for each >>> server name. >>> >> >> What about FQDNs? Does "foo.bar.com" match "foo.bar.com." ? >> > > The presence or absence of the trailing dot affects how DNS is searched > for a match, but the same search rules apply for the addresses as for the > TXT records -- if you find one, you find them both. > > >> >>> >>>> >>>> >>>> On Thursday, January 19, 2017 at 11:42:29 AM UTC-8, Mark D. Roth wrote: >>>>> >>>>> It's obviously going to have to be a heuristic, since we don't have >>>>> any way of knowing the full set of clients a priori. I was thinking that >>>>> we would take a hash of the client's hostname and pid, which >>>>> unfortunately >>>>> wouldn't really be that deterministic. But I'd welcome suggestions for a >>>>> more deterministic algorithm. >>>>> >>>>> On Thu, Jan 19, 2017 at 9:18 AM, 'Craig Tiller' via grpc.io < >>>>> [email protected]> wrote: >>>>> >>>>>> How does the percentage field work? >>>>>> >>>>>> Do clients roll a die to determine if they're in the canary subset? >>>>>> Or is there a deterministic way of determining this? >>>>>> >>>>>> On Thu, Jan 19, 2017 at 8:57 AM 'Mark D. Roth' via grpc.io < >>>>>> [email protected]> wrote: >>>>>> >>>>>>> I've created a gRFC describing how service configs will be encoded >>>>>>> in DNS: >>>>>>> >>>>>>> https://github.com/grpc/proposal/pull/5 >>>>>>> >>>>>>> I'd welcome feedback, especially on the proposed use of TXT records. >>>>>>> >>>>>>> Please keep discussion in this thread. Thanks! >>>>>>> >>>>>>> -- >>>>>>> Mark D. Roth <[email protected]> >>>>>>> Software Engineer >>>>>>> Google, Inc. >>>>>>> >>>>>>> -- >>>>>>> You received this message because you are subscribed to the Google >>>>>>> Groups "grpc.io" group. >>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>> send an email to [email protected]. >>>>>>> To post to this group, send email to [email protected]. >>>>>>> Visit this group at https://groups.google.com/group/grpc-io. >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/d/msgid/grpc-io/CAJgPXp5VBE%3DBVJq8JKXAdKV%3D3-nrjFDHFd0sTwUW%3DGOr%2B3q6Tw%40mail.gmail.com >>>>>>> >>>>>>> <https://groups.google.com/d/msgid/grpc-io/CAJgPXp5VBE%3DBVJq8JKXAdKV%3D3-nrjFDHFd0sTwUW%3DGOr%2B3q6Tw%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>>>>> . >>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "grpc.io" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>> send an email to [email protected]. >>>>>> To post to this group, send email to [email protected]. >>>>>> Visit this group at https://groups.google.com/group/grpc-io. >>>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/grpc-io/CAAvp3oM7boP7X4GeGXmXf0C5pHCBSXZ8_vCF7jFGwceHyjDbDg%40mail.gmail.com >>>>>> >>>>>> <https://groups.google.com/d/msgid/grpc-io/CAAvp3oM7boP7X4GeGXmXf0C5pHCBSXZ8_vCF7jFGwceHyjDbDg%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> Mark D. Roth <[email protected]> >>>>> Software Engineer >>>>> Google, Inc. >>>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "grpc.io" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to [email protected]. >>>> To post to this group, send email to [email protected]. >>>> Visit this group at https://groups.google.com/group/grpc-io. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/grpc-io/ccc7f85d-3f0d-4f59-a438-cedd6c8dd074%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/grpc-io/ccc7f85d-3f0d-4f59-a438-cedd6c8dd074%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> >>> >>> -- >>> Mark D. Roth <[email protected]> >>> Software Engineer >>> Google, Inc. >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "grpc.io" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To post to this group, send email to [email protected] >> <javascript:>. >> Visit this group at https://groups.google.com/group/grpc-io. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/grpc-io/d88e0282-26f6-45d9-afa2-b560c16bbd24%40googlegroups.com >> >> <https://groups.google.com/d/msgid/grpc-io/d88e0282-26f6-45d9-afa2-b560c16bbd24%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Mark D. Roth <[email protected] <javascript:>> > Software Engineer > Google, Inc. > -- You received this message because you are subscribed to the Google Groups "grpc.io" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/grpc-io. To view this discussion on the web visit https://groups.google.com/d/msgid/grpc-io/e29aed3d-c539-4efd-bbff-d15f21429899%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
