Coverity complains rightly about "strncpy (dst, ifname, IFNAMSIZ)"
because it might leave @dst non-NULL-terminated, in case @ifname
is too long (which already would be a bug in the first place).
Replace the strcpy() uses by a new helper nm_utils_ifname_cpy()
that asserts against valid arguments.
IFNAMSIZ is 16 but strncpy doesn't ensure a NUL terminated string in
this case. Coverity rightly complains about it. Use g_strlcpy() instead.
Note that no valid ifname should be 16 characters long in the first
place.
Since 87a3df2e57 ("device: remove default-unmanaged and refactor
unmanaged flags") the device must properly report the IS_SOFTWARE
capability in order to be potentially activated.
When an ifcfg file doesn't specify the TYPE, ifup will
look for a script "ifup-${DEVICETYPE}", where DEVICETYPE
is determined as
[ -z "$DEVICETYPE" ] && DEVICETYPE=$(echo ${DEVICE} | sed "s/[0-9]*$//")
Avoid handling such files by checking that no such ifup script exists.
If a ifcfg file has no TYPE=sit, we would detect it as ethernet,
although the presence of IPV6TUNNELIPV4 indicates that it of type
"sit". Ignore such connections.
In commit 6dc35e66d4 ("settings: add hostnamed support") we started
to use systemd-hostnamed for setting the system static hostname
(i.e. the one written to /etc/hostname), but nm-policy.c still called
sethostname() to set the transient (dynamic) hostname when this needs
to be changed, for example after a reverse lookup of our dynamic IP
address.
Thus, when using systemd the hostname change failed because process'
capabilities are restricted and sethostname() requires CAP_SYS_ADMIN.
We should set also the transient hostname through hostnamed when this
is available.
https://bugzilla.redhat.com/show_bug.cgi?id=1308974
Introduce logging helpers where possible, to uniform the format of
messages and prepend a tag specifying the module that generated the
message, along with other useful information (interface name, ...).
https://bugzilla.gnome.org/show_bug.cgi?id=763040
Originally m4/introspection.m4 was copied from gobject-introspection
repository. We however modified it in commit f6272144e9.
Reimport the file with latest changes, but still preserving our
workaround.
GError codes are only unique per domain, so logging the code without
also indicating the domain is not helpful. And anyway, if the error
messages are not distinctive enough to tell the whole story then we
should fix the error messages.
Based-on-patch-by: Dan Winship <danw@gnome.org>
Use g_error_matches() where we're testing error codes. In particular,
use it rather than looking at only ->code and not also ->domain, which
is just wrong.
[thaller@redhat.com: rebase and modify original patch]
Functions that take a GError** MUST fill it in on error. There is no
need to check whether error is NULL if the function it was passed to
had a failing return value.
Likewise, a proper GError must have a non-NULL message, so there's no
need to double-check that either.
Based-on-patch-by: Dan Winship <danw@gnome.org>
We construct new IP6Config on each rdisc_config_changed(). That's not a smart
thing to do, since that makes us throw away the previous configuration.
In case the two routers on the same network, the first RA triggers
rdisc_config_changed() for changed gateway and addresses. On handling the
second RA rdisc_config_changed() doesn't add the address, resulting in the
address being removed on ip6 config sync.
A side effect of this is that the address is still tentative, resulting in DAD
retry and an new address being added. So the networking still works, but at the
expense of a single DAD failure and an address that's perhaps different from what
was expected.
Until now the internal DHCP client could start a DHCPv6 transaction
but was not able to parse the lease and pass the information back to
the core. Add the missing glue code to make this work.
https://bugzilla.gnome.org/show_bug.cgi?id=762432
