diff for duplicates of <da8dbb76f0cb3762227d8d0a99eaff21@localhost> diff --git a/a/1.txt b/N1/1.txt index 6ffbde3..9eb5533 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -1,31 +1,60 @@ On Tue, 31 May 2011 07:48:09 -0700, tsuna <tsunanet@gmail.com> wrote: + + > I talked to Jerry and he's agreed to share some patches that Google + > has been using internally for years. + + Great! + + > Personally what I think would be ideal would be: + > 1. A sysctl knob for initRTO, to allow people to adjust this + > appropriately for their environment. + > 2. Apply the srtt / rttvar seen on previous connections to new + > connections. + > + > Does that sound reasonable? + > + > For 2), I'm not sure how the details would work yet, I believe the + > kernel already has what's necessary to remember these things on a per + > peer basis, but it would be nice if I could specify things like "for + > 10.x.0.0/16 (local datacenter) use this aggressive setting, for + > 10.0.0.0/8 (my internal backend network) use that, for everything else + > (Internets etc.) use the default". + + Skip sysctl, it is deprecated. The initRTO is the ideal candidate for a + per route knob. And happily you will solve 2) with the per route thing too! + ;-) + + Search the web, you will find some patches where you can see how to extend + the per route system - including iproute2. + + Hagen diff --git a/a/content_digest b/N1/content_digest index 27d7cee..7c2fbc9 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -25,33 +25,62 @@ "\n" "On Tue, 31 May 2011 07:48:09 -0700, tsuna <tsunanet@gmail.com> wrote:\n" "\n" + "\n" + "\n" "> I talked to Jerry and he's agreed to share some patches that Google\n" + "\n" "> has been using internally for years.\n" "\n" + "\n" + "\n" "Great!\n" "\n" + "\n" + "\n" "> Personally what I think would be ideal would be:\n" + "\n" "> 1. A sysctl knob for initRTO, to allow people to adjust this\n" + "\n" "> appropriately for their environment.\n" + "\n" "> 2. Apply the srtt / rttvar seen on previous connections to new\n" + "\n" "> connections.\n" + "\n" "> \n" + "\n" "> Does that sound reasonable?\n" + "\n" "> \n" + "\n" "> For 2), I'm not sure how the details would work yet, I believe the\n" + "\n" "> kernel already has what's necessary to remember these things on a per\n" + "\n" "> peer basis, but it would be nice if I could specify things like \"for\n" + "\n" "> 10.x.0.0/16 (local datacenter) use this aggressive setting, for\n" + "\n" "> 10.0.0.0/8 (my internal backend network) use that, for everything else\n" + "\n" "> (Internets etc.) use the default\".\n" "\n" + "\n" + "\n" "Skip sysctl, it is deprecated. The initRTO is the ideal candidate for a\n" + "\n" "per route knob. And happily you will solve 2) with the per route thing too!\n" + "\n" ";-)\n" "\n" + "\n" + "\n" "Search the web, you will find some patches where you can see how to extend\n" + "\n" "the per route system - including iproute2.\n" "\n" + "\n" + "\n" Hagen -f9676f26626ded7f5838dba06d855c029bb782757a52efaa7531e9355feb44ed +bba9bbdcd2924d8b39d7889ab4a30a014599599e87efddd5e8ca7c8f229bdb75
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.