From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Philipp Reisner To: drbd-dev@lists.linbit.com Subject: Re: [Drbd-dev] 0.7.11 soon Date: Mon, 9 May 2005 11:37:01 +0200 References: <200505091030.00455.philipp.reisner@linbit.com> <200505091058.42759.hug@abanet.ch> In-Reply-To: <200505091058.42759.hug@abanet.ch> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <200505091137.01385.philipp.reisner@linbit.com> Cc: Philipp Hug List-Id: Coordination of development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Am Montag, 9. Mai 2005 10:58 schrieb Philipp Hug: > philipp, > > if you release a new version it won't make it into sarge... > 0.7.10 will be in sarge and has to be supported for about 2 years ;-) > > philipp > > On Monday 09 May 2005 10.30, Philipp Reisner wrote: > > Hi, > > > > The last DRBD release happened on the 31st Jan of 2005, this > > is more than 3 month by now. I guess we need to give our > > users the feeling that DRBD is still alive :) > > > > 0.7.11 (api:77/proto:74) > > ----- > > * The upper limit of the runtime tuning parameter max_buffers > > was ways too low for today's IBM storage controllers. > > Was 32 - 10000 ; Now it is 32 - 131072 ( Now one could > > allow DRBD to allocate up to half a gigabyte of memory as > > receive buffers) > > * A fix to the Makefiles to make building of RPMs work on RAHS4. > > * A workaround to have DBRD running on PPC64 with Linux-2.4.x kernels, > > on Linux-2.6.x everything was fine already. > > * Removal of dead code (busy_blocks). > > > > I will tag, create the tar-ball and announce it on wednesday, > > as long as nobody objects. Yes, I expected this.=20 The only relevant patch for Debian is the first item of the changelog: =2D-- user/drbd_limits.h (.../https://svn.drbd.org/drbd/tags/drbd-0.7.10) = (Revision 1789) +++ user/drbd_limits.h (.../http://svn.drbd.org/drbd/branches/drbd-0.7) = (Revision 1789) @@ -49,8 +49,8 @@ /* I don't think that a tcp send buffer of more than 10M is usefull */ RANGE(SNDBUF_SIZE, 1, 10000000); =2D /* arbitrary. */ =2D RANGE(MAX_BUFFERS, 32, 10000); + /* @4k PageSize -> 128kB - 512MB */ + RANGE(MAX_BUFFERS, 32, 131072); /* 0 is disabled. * 200 should be more than enough even for very short timeouts */ Maybe you can slip it in by creating a "drbd-0.7.10-4" or something. But it is nothing critical at all ... as we know drbd-0.7.10 is really fine for most people. =2DPhil =2D-=20 : Dipl-Ing Philipp Reisner Tel +43-1-8178292-50 : : LINBIT Information Technologies GmbH Fax +43-1-8178292-82 : : Sch=F6nbrunnerstr 244, 1120 Vienna, Austria http://www.linbit.com :