Port GlusterFS as a GSoC 2013 project
Hi there, I'm now a student and trying to get involved in GSoC this year. I found the proposal of about GlusterFS in the idea list wiki page very interesting to me, possibly it will be porting from NetBSD implementation. As I'm quite distant from idea owner so there's a big time difference, he also suggested me to try to find some folks that are physically closer to me to help mentoring. So I'm writing to ask if there's any folk in Europe is interested helping me with the project in any way, it could be easier for communication and discussion. Thanks a lot. Regards, Mike ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Port GlusterFS as a GSoC project
Hi there, I'm now a student and trying to get involved in GSoC this year. I found the proposal of about GlusterFS in the idea list wiki page very interesting to me, possibly it will be porting from NetBSD implementation. As I'm quite distant from idea owner, he also suggested me to try to find some folks that are physically closer to me to help mentoring. So I'm writing to ask if there's any folk in Europe is interested helping me with the project in any way, it could be easier for communication and discussion. Thanks a lot. Regards, Mike ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Re: Some improvements to rm(1)
On Thu, Apr 25, 2013 at 10:56:10PM -0400, Eitan Adler wrote: > On 25 April 2013 22:50, Brooks Davis wrote: > > On Thu, Apr 25, 2013 at 10:16:32PM -0400, Eitan Adler wrote: > >> Anyone have thoughts on the following? > >> commit 82c78ba923d8ce4a1bfbb309658c49021c8bb384 > >> Author: Eitan Adler > >> Date: Thu Apr 25 22:14:49 2013 -0400 > >> Take some improvements from DragonFlyBSD: > >> - add const where appropriate > >> - add static where appropriate > >> - fix a whitespace issues > > > > The no-op changes look more correct to me. > > I think the -x option seems a bit odd. What is the use case? At a > > first thought, it seems to raise more questions than it resolves. > It goes along with cp -x, find -x, and others. > Quick example #1: You have /usr/ports /usr/ports/distfiles as > different mount points it lets you wipe /usr/ports without wiping your > distfile cache. > Quick example #2: You have /usr/src/ null mounted in every user's > /home/ and you want to wipe one home directory. Hmm, isn't this already possible using find -x DIR -delete ? There will be an error message 'Device busy' about attempting to delete the mount point but this does not even affect the exit status and all files not under the mount point are removed. -- Jilles Tjoelker ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Re: Some improvements to rm(1)
On 26 April 2013 08:22, Jilles Tjoelker wrote: > On Thu, Apr 25, 2013 at 10:56:10PM -0400, Eitan Adler wrote: >> On 25 April 2013 22:50, Brooks Davis wrote: >> > On Thu, Apr 25, 2013 at 10:16:32PM -0400, Eitan Adler wrote: >> >> Anyone have thoughts on the following? > >> >> commit 82c78ba923d8ce4a1bfbb309658c49021c8bb384 >> >> Author: Eitan Adler >> >> Date: Thu Apr 25 22:14:49 2013 -0400 > >> >> Take some improvements from DragonFlyBSD: >> >> - add const where appropriate >> >> - add static where appropriate >> >> - fix a whitespace issues >> > >> > The no-op changes look more correct to me. > >> > I think the -x option seems a bit odd. What is the use case? At a >> > first thought, it seems to raise more questions than it resolves. > >> It goes along with cp -x, find -x, and others. > >> Quick example #1: You have /usr/ports /usr/ports/distfiles as >> different mount points it lets you wipe /usr/ports without wiping your >> distfile cache. > >> Quick example #2: You have /usr/src/ null mounted in every user's >> /home/ and you want to wipe one home directory. > > Hmm, isn't this already possible using find -x DIR -delete ? Yes, rm's functionality can be fully replicated by find. -- Eitan Adler ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Re: Some improvements to rm(1)
On 4/26/2013 7:23 AM, Eitan Adler wrote: Yes, rm's functionality can be fully replicated by find. As well as anything using -R. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Re: Some improvements to rm(1)
On Thu, Apr 25, 2013 at 9:50 PM, Brooks Davis wrote: > I think the -x option seems a bit odd. What is the use case? At a > first thought, it seems to raise more questions than it resolves. > I was cleaning up a system a year ago and I had an "rm -rf" traverse into a production NFS mountpoint.. oops. I only realized it when it was taking longer than I expected so I stopped it to investigate. Had to restore a bunch of data from backups. Thank you for proposing the patch, I hope it gets committed. Dan ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
Re: Some improvements to rm(1)
On Fri, Apr 26, 2013 at 08:33:54AM -0500, Joshua Isom wrote:
> On 4/26/2013 7:23 AM, Eitan Adler wrote:
> > Yes, rm's functionality can be fully replicated by find.
> As well as anything using -R.
Emulating other -R things using find becomes quite slow when you don't
want to impose {PATH_MAX} limits or open up symlink-based race windows
because the only safe option is -execdir UTILITY {} \;. Any find command
based on -exec, -print or -print0 passes pathnames which are subject to
{PATH_MAX} limits and directories concurrently replaced with symlinks.
The construct -execdir ... {} + is unusably broken in older FreeBSD
versions and gives no advantage compared to -execdir ... {} \; in recent
-CURRENT.
With -L, this is not a new problem because symlinks are followed anyway
and the underlying code (fts(3)) always imposes the {PATH_MAX} limit in
that case.
The -delete primary is safe like -execdir.
I'm not entirely sure about this because the rm(1) patch is simple and
the new syntax is fairly clear.
--
Jilles Tjoelker
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "freebsd-hackers-unsubscr...@freebsd.org"
makeman issues
I'm trying to regenerate src.conf.5 after my recent WITH*_IDEA removal but the diff shows: Does anyone know why WITHOUT_BIND_UTILS is showing up a bazillion times? diff --git a/share/man/man5/src.conf.5 b/share/man/man5/src.conf.5 index 17afca2..54b63db 100644 --- a/share/man/man5/src.conf.5 +++ b/share/man/man5/src.conf.5 @@ -1,7 +1,7 @@ .\" DO NOT EDIT-- this file is automatically generated. -.\" from FreeBSD: head/tools/build/options/makeman 236279 2012-05-30 02:37:20Z gjb +.\" from FreeBSD$ .\" $FreeBSD$ -.Dd February 15, 2013 +.Dd April 26, 2013 .Dt SRC.CONF 5 .Os .Sh NAME @@ -78,57 +78,117 @@ The following list provides a name and short description for variables that can be used for source builds. .Bl -tag -width indent .It Va WITHOUT_ACCT -.\" from FreeBSD: head/tools/build/options/WITHOUT_ACCT 223201 2011-06-17 20:47:44Z ed +.\" $FreeBSD$ Set to not build process accounting tools such as .Xr accton 8 and .Xr sa 8 . +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_ACPI -.\" from FreeBSD: head/tools/build/options/WITHOUT_ACPI 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to not build .Xr acpiconf 8 , .Xr acpidump 8 and related programs. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_AMD -.\" from FreeBSD: head/tools/build/options/WITHOUT_AMD 183242 2008-09-21 22:02:26Z sam +.\" $FreeBSD$ Set to not build .Xr amd 8 , and related programs. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_APM -.\" from FreeBSD: head/tools/build/options/WITHOUT_APM 183242 2008-09-21 22:02:26Z sam +.\" $FreeBSD$ Set to not build .Xr apm 8 , .Xr apmd 8 and related programs. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITH_ARM_EABI -.\" from FreeBSD: head/tools/build/options/WITH_ARM_EABI 245539 2013-01-17 05:56:28Z andrew +.\" $FreeBSD$ Set the ARM ABI to EABI. .It Va WITHOUT_ASSERT_DEBUG -.\" from FreeBSD: head/tools/build/options/WITHOUT_ASSERT_DEBUG 162215 2006-09-11 13:55:27Z ru +.\" $FreeBSD$ Set to compile programs and libraries without the .Xr assert 3 checks. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_AT -.\" from FreeBSD: head/tools/build/options/WITHOUT_AT 183242 2008-09-21 22:02:26Z sam +.\" $FreeBSD$ Set to not build .Xr at 1 and related utilities. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_ATF -.\" from FreeBSD: head/tools/build/options/WITHOUT_ATF 241823 2012-10-22 01:18:41Z marcel +.\" $FreeBSD$ Set to not build programs and libraries related to the ATF testing framework. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_ATM -.\" from FreeBSD: head/tools/build/options/WITHOUT_ATM 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to not build programs and libraries related to ATM networking. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_AUDIT -.\" from FreeBSD: head/tools/build/options/WITHOUT_AUDIT 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to not build audit support into system programs. +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_AUTHPF -.\" from FreeBSD: head/tools/build/options/WITHOUT_AUTHPF 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to not build .Xr authpf 8 . +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_BIND -.\" from FreeBSD: head/tools/build/options/WITHOUT_BIND 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Setting this variable will prevent any part of BIND from being built. When set, it also enforces the following options: .Pp @@ -149,31 +209,49 @@ When set, it also enforces the following options: .Va WITHOUT_BIND_UTILS .El .It Va WITHOUT_BIND_DNSSEC -.\" from FreeBSD: head/tools/build/options/WITHOUT_BIND_DNSSEC 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to avoid building or installing the DNSSEC related binaries, .Xr dnssec-keygen 8 and .Xr dnssec-signzone 8 . +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITHOUT_BIND_ETC -.\" from FreeBSD: head/tools/build/options/WITHOUT_BIND_ETC 156932 2006-03-21 07:50:50Z ru +.\" $FreeBSD$ Set to avoid installing the default files to .Pa /var/named/etc/namedb . +When set, it also enforces the following options: +.Pp +.Bl -item -compact +.It +.Va WITHOUT_BIND_UTILS +.El .It Va WITH_BIND_IDN -.\" from FreeBSD: head/tools/build/options/WI
Re: makeman issues
On Fri, Apr 26, 2013 at 08:05:50PM -0400, Eitan Adler wrote: > I'm trying to regenerate src.conf.5 after my recent WITH*_IDEA removal > but the diff shows: > > Does anyone know why WITHOUT_BIND_UTILS is showing up a bazillion times? > Looks like DES already spotted this too. r246838 | des | 2013-02-15 10:09:24 -0500 (Fri, 15 Feb 2013) | 4 lines Manually add the correct text for the LDNS and LDNS_UTILS options. There seems to be a bug in makeman that causes it to misunderstand the relationship between LDNS_UTILS and BIND_UTILS. Glen pgpsfZakpdpGx.pgp Description: PGP signature
