<feed xmlns='http://www.w3.org/2005/Atom'>
<title>openembedded-core.git/meta/recipes-core/initscripts, branch dora</title>
<subtitle>Mirror of openembedded-core</subtitle>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/'/>
<entry>
<title>initscripts: add missing dmesg.sh to run-level S</title>
<updated>2013-12-12T16:53:10+00:00</updated>
<author>
<name>Hans Beckerus</name>
<email>hans.beckerus@gmail.com</email>
</author>
<published>2013-11-15T14:25:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=fb7a6e0e3c5790415d56d14ec1a5eda5f9e6d039'/>
<id>fb7a6e0e3c5790415d56d14ec1a5eda5f9e6d039</id>
<content type='text'>
In commit faa8cc6c2a582a32c695f3f2b0d45b6892c769fd dmesg.sh was
added to the set of init.d scripts. But the script was never put
in any run-level. This patch will add dmesg.sh to run-level S.

(From OE-Core master rev: 7d2767d4e27c6d0eaa56f3e126df56e65a5364c9)

Signed-off-by: Hans Beckerus &lt;hans.beckerus AT gmail.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
Signed-off-by: Robert Yang &lt;liezhi.yang@windriver.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In commit faa8cc6c2a582a32c695f3f2b0d45b6892c769fd dmesg.sh was
added to the set of init.d scripts. But the script was never put
in any run-level. This patch will add dmesg.sh to run-level S.

(From OE-Core master rev: 7d2767d4e27c6d0eaa56f3e126df56e65a5364c9)

Signed-off-by: Hans Beckerus &lt;hans.beckerus AT gmail.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
Signed-off-by: Robert Yang &lt;liezhi.yang@windriver.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>initscripts: insure checkroot.sh runs before anything writing to the file</title>
<updated>2013-12-11T02:12:18+00:00</updated>
<author>
<name>Jeffrey C Honig</name>
<email>jeffrey.honig@windriver.com</email>
</author>
<published>2013-10-12T05:30:29+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=e54864e6db9fddb8ec98e323fe6da88e02ff070a'/>
<id>e54864e6db9fddb8ec98e323fe6da88e02ff070a</id>
<content type='text'>
If bootlogd was configured to write to a log file on the root file system,
the checkroot.sh was not able to change the rootfs to read-only because
bootlogd was started earlier and had a file descriptor open.  Lowering
the order of checkroot.sh ensures that the volatile filesystem is set
up before anything writes to it.

(From OE-Core master rev: 13c9bc143f6861517970dafdc7e7a45740d0933d)

Signed-off-by: Jeffrey C Honig &lt;jeffrey.honig@windriver.com&gt;
Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
Signed-off-by: Robert Yang &lt;liezhi.yang@windriver.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If bootlogd was configured to write to a log file on the root file system,
the checkroot.sh was not able to change the rootfs to read-only because
bootlogd was started earlier and had a file descriptor open.  Lowering
the order of checkroot.sh ensures that the volatile filesystem is set
up before anything writes to it.

(From OE-Core master rev: 13c9bc143f6861517970dafdc7e7a45740d0933d)

Signed-off-by: Jeffrey C Honig &lt;jeffrey.honig@windriver.com&gt;
Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
Signed-off-by: Robert Yang &lt;liezhi.yang@windriver.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>initscripts: filter out empty lines from volatiles in check_requirements()</title>
<updated>2013-09-10T22:00:46+00:00</updated>
<author>
<name>Ming Liu</name>
<email>ming.liu@windriver.com</email>
</author>
<published>2013-09-10T07:53:24+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=71ab9ee58b0ba5e3f5cbf403d1b8fb79fc7f5ed1'/>
<id>71ab9ee58b0ba5e3f5cbf403d1b8fb79fc7f5ed1</id>
<content type='text'>
So far the rules of check_requirements() is too strict to lead mismatch
when empty lines exist in volatiles.

Signed-off-by: Ming Liu &lt;ming.liu@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
So far the rules of check_requirements() is too strict to lead mismatch
when empty lines exist in volatiles.

Signed-off-by: Ming Liu &lt;ming.liu@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>read-only-rootfs-hook.sh: check before bind mounting /var/lib</title>
<updated>2013-08-26T12:19:36+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-08-26T06:43:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=86ac10995fd08226f82d87e23fda5d4898c3190f'/>
<id>86ac10995fd08226f82d87e23fda5d4898c3190f</id>
<content type='text'>
It's possible that /var/lib is on a separate writable partition. In such
situation, we should not bind mount /var/lib with tmpfs, becasue it's
already writable.

This patch fixes this problem by checking whether /var/lib is already
on a writable partition.

[YOCTO #4888]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It's possible that /var/lib is on a separate writable partition. In such
situation, we should not bind mount /var/lib with tmpfs, becasue it's
already writable.

This patch fixes this problem by checking whether /var/lib is already
on a writable partition.

[YOCTO #4888]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>checkroot.sh: check for conflicting configurations</title>
<updated>2013-08-26T12:19:24+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-08-26T06:43:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=1565a0c5a3f245703e280ca90cf11d3f9374788a'/>
<id>1565a0c5a3f245703e280ca90cf11d3f9374788a</id>
<content type='text'>
In our current boot scripts, two ways are used to determine whether
the rootfs is read-only or not. One by checking the READ_ONLY_ROOTFS
value in /etc/default/rcS, the other by checking the /etc/fstab entry.

&gt;From a normal Linux user's point of view, the way of checking the
/etc/fstab entry is preferred. However, as there are several boot scripts
that need to know whether rootfs is read-only or not, checking /etc/fstab
in each script brings too much overhead. Thus, these boot scripts
use the READ_ONLY_ROOTFS value in /etc/default/rcS.

In normal use cases, there would be no problem, as both /etc/default/rcS and
the /etc/fstab are modified at rootfs time. However, if the user changes
the mount option for rootfs in /etc/fstab to read-write, and he/she forgets
to change the value in /etc/default/rcS, there would be unexpected results.
For example, the udev-cache would still be disabled.

So at a minimal, a check for conflicting configurations between /etc/fstab
and /etc/default/rcS should be added in checkroot.sh so that there would be
reasonable warnings if users have configured the system in a non-consistent
way.

[YOCTO #4880]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In our current boot scripts, two ways are used to determine whether
the rootfs is read-only or not. One by checking the READ_ONLY_ROOTFS
value in /etc/default/rcS, the other by checking the /etc/fstab entry.

&gt;From a normal Linux user's point of view, the way of checking the
/etc/fstab entry is preferred. However, as there are several boot scripts
that need to know whether rootfs is read-only or not, checking /etc/fstab
in each script brings too much overhead. Thus, these boot scripts
use the READ_ONLY_ROOTFS value in /etc/default/rcS.

In normal use cases, there would be no problem, as both /etc/default/rcS and
the /etc/fstab are modified at rootfs time. However, if the user changes
the mount option for rootfs in /etc/fstab to read-write, and he/she forgets
to change the value in /etc/default/rcS, there would be unexpected results.
For example, the udev-cache would still be disabled.

So at a minimal, a check for conflicting configurations between /etc/fstab
and /etc/default/rcS should be added in checkroot.sh so that there would be
reasonable warnings if users have configured the system in a non-consistent
way.

[YOCTO #4880]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Richard Purdie &lt;richard.purdie@linuxfoundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>initscripts: remove obsolete device_table.txt</title>
<updated>2013-08-20T14:09:07+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-08-16T09:36:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=cc3842255ade0ad6fb312f06f5e5e1edd9767c94'/>
<id>cc3842255ade0ad6fb312f06f5e5e1edd9767c94</id>
<content type='text'>
device_table.txt was once used by an init script named 'devices' to
create the basic files under /dev. However, it's no longer used now.
The devices init script has been removed, and makedevs command has been
disabled by default in busybox.

Besides, considering the device managers and devtmpfs filesystem, this
file is not likely to be used again. So we remove it.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
device_table.txt was once used by an init script named 'devices' to
create the basic files under /dev. However, it's no longer used now.
The devices init script has been removed, and makedevs command has been
disabled by default in busybox.

Besides, considering the device managers and devtmpfs filesystem, this
file is not likely to be used again. So we remove it.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>populate-volatile.sh: use 'cp -a' to avoid potential problem</title>
<updated>2013-08-16T10:14:15+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-08-07T07:17:15+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=742440441222e0627abbdd3eb2ee16401e8f4adf'/>
<id>742440441222e0627abbdd3eb2ee16401e8f4adf</id>
<content type='text'>
Previously, dead links in target directory will not be copied.
This is incorrect as dead links are not uncommon in our rootfs.
So we use '-a' option instead.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Previously, dead links in target directory will not be copied.
This is incorrect as dead links are not uncommon in our rootfs.
So we use '-a' option instead.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>populate-volatile.sh: don't spawn background process at rootfs time</title>
<updated>2013-08-16T10:14:14+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-07-27T14:21:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=4bd419f95868d5b8707a45cee5f6c5c6a840a65b'/>
<id>4bd419f95868d5b8707a45cee5f6c5c6a840a65b</id>
<content type='text'>
If we're building a read-only rootfs, we'll get the following error now
and then.

    tar: .: file changed as we read it

The root cause is that we spawn background process at rootfs time.
When the tar command is running, it's possible that files under rootfs
are changed by background processes, thus this error.

[YOCTO #4937]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If we're building a read-only rootfs, we'll get the following error now
and then.

    tar: .: file changed as we read it

The root cause is that we spawn background process at rootfs time.
When the tar command is running, it's possible that files under rootfs
are changed by background processes, thus this error.

[YOCTO #4937]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>populate-volatile.sh: use $ROOT_DIR/var/volatile/tmp as TMPDIR</title>
<updated>2013-08-16T10:14:14+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-07-23T02:56:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=7c2c36a97bab46c73a3a1fb743ad3ec67a4c072e'/>
<id>7c2c36a97bab46c73a3a1fb743ad3ec67a4c072e</id>
<content type='text'>
It's possible that a 'No such directory' error occurs when doing
check_requirement in populate-volatile.sh at rootfs time. This is
because the $ROOT_DIR/var/tmp might be a dead link.

Use $ROOT_DIR/var/volatile/tmp as the TMPDIR instead to avoid this
error.

[YOCTO #4883]
[YOCTO #4103]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It's possible that a 'No such directory' error occurs when doing
check_requirement in populate-volatile.sh at rootfs time. This is
because the $ROOT_DIR/var/tmp might be a dead link.

Use $ROOT_DIR/var/volatile/tmp as the TMPDIR instead to avoid this
error.

[YOCTO #4883]
[YOCTO #4103]

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>checkroot.sh: respect the value of ENABLE_ROOTFS_FSCK</title>
<updated>2013-08-09T15:54:47+00:00</updated>
<author>
<name>Chen Qi</name>
<email>Qi.Chen@windriver.com</email>
</author>
<published>2013-08-08T10:26:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.multitech.net/cgit/openembedded-core.git/commit/?id=5c11cf4db8f43003fb7132b92a78659f70f07ce0'/>
<id>5c11cf4db8f43003fb7132b92a78659f70f07ce0</id>
<content type='text'>
Previously, fsck is always disabled because the value of rootcheck in
the checkroot.sh script is always set to 'no'.

We should respect the value of ENABLE_ROOTFS_FSCK in /etc/default/rcS
to allow for filesystem check.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Previously, fsck is always disabled because the value of rootcheck in
the checkroot.sh script is always set to 'no'.

We should respect the value of ENABLE_ROOTFS_FSCK in /etc/default/rcS
to allow for filesystem check.

Signed-off-by: Chen Qi &lt;Qi.Chen@windriver.com&gt;
Signed-off-by: Saul Wold &lt;sgw@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
