From: Matthias Fischer <matthias.fischer@ipfire.org>
To: development@lists.ipfire.org
Subject: Re: [PATCH] tar: Update to 1.30 (V2)
Date: Wed, 06 Jun 2018 21:49:41 +0200 [thread overview]
Message-ID: <fcf0b540-eb74-6dbb-28df-534e4423cf3b@ipfire.org> (raw)
In-Reply-To: <ad74f9578ba87824bf9fed38ef159e8f77e79fee.camel@ipfire.org>
[-- Attachment #1: Type: text/plain, Size: 4464 bytes --]
On 06.06.2018 14:32, Michael Tremer wrote:
> Hello,
Hi,
> did you test our backup and restore scripts as well? Are they affected by any of
> these changes?
Twice: Yes.
I installed the new 'tar 1.30' on my production machine (Duo Box, Core
120), made a backup and restored it on my testmachine (Core 120, clean
install). Everything worked.
No seen problems so far, except for the fact that right after the
restore I had to watch out hellishly which machine I administered... ;-)
All that is left: I have to install a bunch of addons.
HTH,
Matthias
> -Michael
>
> On Thu, 2018-05-31 at 11:11 +0200, Matthias Fischer wrote:
>> Hi,
>>
>> On 29.05.2018 21:36, Michael Tremer wrote:
>> > Hi,
>> >
>> > On Tue, 2018-05-29 at 20:42 +0200, Matthias Fischer wrote:
>> > > For details see:
>> > > https://www.gnu.org/software/tar/
>> > >
>> > > Second attempt to update 'tar' to 1.30.
>> > > ...
>> > > lfs/Config | 8 +++-----
>> > > lfs/cdrom | 5 ++---
>> > > lfs/core-updates | 7 +++----
>> > > lfs/tar | 6 +++---
>> > > make.sh | 4 ++--
>> > > 5 files changed, 13 insertions(+), 17 deletions(-)
>> > >
>> > > diff --git a/lfs/Config b/lfs/Config
>> > > index cf70f81f9..8e8e5879b 100644
>> > > --- a/lfs/Config
>> > > +++ b/lfs/Config
>> > > @@ -205,13 +205,11 @@ define PAK
>> > > sed -e 's/BUILDTARGET/$(BUILDTARGET)/g' -e 's/KVER/$(KVER)/g' -e
>> > > 's/MACHINE/$(BUILD_ARCH)/g' -i /install/packages/package/ROOTFILES
>> > > sed -e 's/xxxKVERxxx/$(KVER)/g' -i
>> > > /install/packages/package/install.sh
>> > > chmod 755 /install/packages/package/{{,un}install,update}.sh
>> > > - cd / && tar cf /install/packages/package/files.tmp --files-
>> > > from=/install/packages/package/ROOTFILES --exclude='#*' \
>> > > - -p --numeric-owner
>> > > + cd / && tar cfp /install/packages/package/files.tmp --numeric-
>> > > owner
>> > > --exclude='#*' --files-from=/install/packages/package/ROOTFILES
>> > > # Double tar to remove double files
>> > > - tar xf /install/packages/package/files.tmp -C
>> > > /install/packages/package/tmp/ \
>> > > - -p --numeric-owner
>> > > + tar xfp /install/packages/package/files.tmp --numeric-owner -C
>> > > /install/packages/package/tmp/
>> > > rm -f /install/packages/package/files.tmp
>> > > - cd /install/packages/package/tmp/ && tar cf - -p --numeric-owner
>> > > * |
>> > > xz $(XZ_OPT) > /install/packages/package/files.tar.xz
>> > > + cd /install/packages/package/tmp/ && tar cJp --numeric-owner * |
>> > > xz
>> > > $(XZ_OPT) > /install/packages/package/files.tar.xz
>> >
>> > You are introducing another -J parameter here again which will cause the
>> > tarball
>> > to be compressed twice.
>>
>> Yep. Now I saw this too. I should have checked the logs a little more
>> closely. Overlooked this.
>>
>> > > ...
>> > > diff --git a/lfs/cdrom b/lfs/cdrom
>> > > index 00f185d81..6382034b5 100644
>> > > --- a/lfs/cdrom
>> > > +++ b/lfs/cdrom
>> > > ...
>> > > diff --git a/lfs/tar b/lfs/tar
>> > > index 6a949a003..4e914b535 100644
>> > > --- a/lfs/tar
>> > > +++ b/lfs/tar
>>
>>
>> > > ...
>> > > diff --git a/make.sh b/make.sh
>> > > index bfcfbc3e4..1050eb03b 100755
>> > > --- a/make.sh
>> > > +++ b/make.sh
>> > > @@ -1785,8 +1785,8 @@ toolchain)
>> > > buildtoolchain
>> > > echo "`date -u '+%b %e %T'`: Create toolchain image for
>> > > ${BUILD_ARCH}" | tee -a $LOGFILE
>> > > test -d $BASEDIR/cache/toolchains || mkdir -p
>> > > $BASEDIR/cache/toolchains
>> > > - cd $BASEDIR && tar -cf- --exclude='log/_build.*.log'
>> > > build/${TOOLS_DIR} build/bin/sh log | xz ${XZ_OPT} \
>> > > - > cache/toolchains/$SNAME-$VERSION-toolchain-
>> > > $TOOLCHAINVER-
>> > > ${BUILD_ARCH}.tar.xz
>> > > + cd $BASEDIR && XZ_OPT="$(XZ_OPT)" tar -Jc --
>> > > exclude='log/_build.*.log' -f cache/toolchains/$SNAME-$VERSION-toolchain-
>> > > $TOOLCHAINVER-${BUILD_ARCH}.tar.xz --xz \
>> > > + build/${TOOLS_DIR} build/bin/sh log >> $LOGFILE
>> >
>> > This bit also reverts some of the previous changes.
>>
>> Yep. Hm! Too bad. I was a bit too fast in committing. I'm dropping the
>> changes for 'make.sh' and try to make only the changes necessary for
>> 'tar 1.30' in 'Config', 'cdrom' and 'core-updates'.
>> I updated the branch and rebuild.
>>
>> > Are you sure you worked on next?
>>
>> Yes. Definitely.
>>
>> Best,
>> Matthias
>
next prev parent reply other threads:[~2018-06-06 19:49 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <fb10fb9d-c59d-1e47-6287-bb4948c528e3@ipfire.org>
2018-06-06 12:32 ` Michael Tremer
2018-06-06 19:49 ` Matthias Fischer [this message]
2018-06-06 19:55 ` Matthias Fischer
2018-06-07 19:33 ` Michael Tremer
2018-05-29 18:42 Matthias Fischer
2018-05-29 19:36 ` Michael Tremer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=fcf0b540-eb74-6dbb-28df-534e4423cf3b@ipfire.org \
--to=matthias.fischer@ipfire.org \
--cc=development@lists.ipfire.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox