Let's sprinkle some snake ointment.
This is questionable, because we copy secrets all over the place where
we their deallocation (and clearing) is not in our control. For example,
the GValue setter/getter copies the string (but does not clean the
secret). Also, when converting the property to a GVariant, we won't
clear it. So this does not catch a lot of cases.
Still, if we can with relative ease avoid leaking the string at some
places, do it.
libnm's data structures are commonly not thread safe (like
NMConnection). However, it must be possible that all operations can
operate on *different* data in a thread safe manner. That means, we need
to take care about our global variables.
nm_utils_ssid_to_utf8() uses a list of encodings, which gets cached.
- replace the GHashTables with a static list. Since it doesn't cost
anything, make the list sorted and look it up via binary search.
nm_hash_str() has the proper name and signature for that it does.
That is, it has a "nm_hash_*" prefix and the parameter is of type
"const char *".
nm_hash_str() has this name because it is primarily about hashing.
For hash tables, glib has g_str_hash() and g_str_equal(). We want
to replace g_str_hash() with our implementation (nm_hash_str()) because
that uses siphash24 with a random seed.
But in those cases we want to use the more familiar name "nm_str_hash()",
which reminds of g_str_hash() and follows the pattern of g_str_equal().
Thus:
g_hash_table_new(nm_str_hash, g_str_equal);
is preferable over
g_hash_table_new(nm_hash_str, g_str_equal);
Hence, we have (and had) nm_str_hash() effectively an alias to
nm_hash_str.
The question is: which name is preferable? Or should they be both present
for their slightly distinct uses? The approach taken here is to have
both names, to reflect their purpose better.
But as the usage of nm_str_hash is as function pointer for GHashTable, it was
not an inline function and we'd pay a small overhead with this approach of
aliasing. Avoid that overhead by defining nm_str_hash with the C
preprocessor.
For similar reasons, do that for nm_direct_hash.
The force-commit flag is used to force the commit of an address or a
route from DHCP/RA even when it was removed from platform externally
(for example because it expired). Routes generated from the l3cd
should also have the flag set.
Without this, NM properly re-adds the DHCP address after the lease is
lost and obtained again, but fails to add the prefix-route.
Fixes: 2838b1c5e8 ('core: track force-commit flag for l3cd and platform objects')
https://bugzilla.redhat.com/show_bug.cgi?id=2033991https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1049
Consider externally removed IPv4LL bad, proceed as if a collision was
detected. Otherwise we trip an assert:
<trace> [1641816260.3963] l3cfg[b8bf8cd16ec4732e,ifindex=47]: emit signal (platform-change-on-idle, obj-type-flags=0x14)
**
nm:ERROR:src/core/nm-l3-ipv4ll.c:888:_ipv4ll_state_change: code should not be reached
Aborted (core dumped)
#3 0x00007f41621d020e in g_assertion_message_expr (domain=domain@entry=0x5559cd829140 "nm",
file=file@entry=0x5559cd823e51 "src/core/nm-l3-ipv4ll.c",
line=line@entry=888, func=func@entry=0x5559cd824d30 <__func__.38810> "_ipv4ll_state_change",
expr=expr@entry=0x0) at gtestutils.c:2556
#4 0x00005559cd719686 in _ipv4ll_state_change (self=0x5559cef886c0,
is_on_idle_handler=0) at src/core/nm-l3-ipv4ll.c:888
#8 0x00007f41626a5093 in <emit signal ??? on instance 0x5559ceffaa30 [NML3Cfg]>
(instance=instance@entry=0x5559ceffaa30, signal_id=<optimized out>,
detail=detail@entry=0) at gsignal.c:3448
#9 0x00005559cd511a03 in _nm_l3cfg_emit_signal_notify
(self=self@entry=0x5559ceffaa30 [NML3Cfg], notify_data=notify_data@entry=0x7ffd1caa8640)
at src/core/nm-l3cfg.c:576
#10 0x00005559cd5122dc in _nm_l3cfg_emit_signal_notify_acd_event (self=0x5559ceffaa30 [NML3Cfg],
acd_data=<optimized out>) at src/core/nm-l3cfg.c:2008
#11 0x00005559cd512463 in _nm_l3cfg_emit_signal_notify_acd_event_all
(self=0x5559ceffaa30 [NML3Cfg]) at src/core/nm-l3cfg.c:2041
#12 0x00005559cd5194ef in _l3_acd_nacd_event (fd=<optimized out>, condition=<optimized out>,
user_data=<optimized out>) at src/core/nm-l3cfg.c:1536
#13 0x00007f41621a895d in g_main_dispatch (context=0x5559ceec8bc0) at gmain.c:3193
#14 0x00007f41621a895d in g_main_context_dispatch (context=context@entry=0x5559ceec8bc0)
at gmain.c:3873
#15 0x00007f41621a8d18 in g_main_context_iterate (context=0x5559ceec8bc0, block=block@entry=1,
dispatch=dispatch@entry=1, self=<optimized out>) at gmain.c:3946
#16 0x00007f41621a9042 in g_main_loop_run (loop=0x5559ceea40f0) at gmain.c:4142
#17 0x00005559cd47c7a4 in main (argc=<optimized out>, argv=<optimized out>)
at src/core/main.c:511
https://bugzilla.redhat.com/show_bug.cgi?id=2028404https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1059
Try to debug a hang in platform code, presumably during poll().
This logging seems useful for debugging this particular issue,
but it might be useful in general.
If the ovs interface goes away, the ifindex gets zeroed out and l3cfg is
cleaned. We can't follow up with IP configuration. Bad things happen if
we try to:
#0 0x00007f769734c895 in _g_log_abort (breakpoint=1) at gmessages.c:580
#1 0x00007f769734db98 in g_logv (log_domain=0x55b2472d8840 "nm",
log_level=G_LOG_LEVEL_CRITICAL, format=<optimized out>,
args=args@entry=0x7fff4041b9d0) at gmessages.c:1391
#2 0x00007f769734dd63 in g_log (log_domain=log_domain@entry=0x55b2472d8840 "nm",
log_level=log_level@entry=G_LOG_LEVEL_CRITICAL,
format=format@entry=0x7f769739a620 "%s: assertion '%s' failed") at gmessages.c:1432
#3 0x00007f769734e59d in g_return_if_fail_warning
(log_domain=log_domain@entry=0x55b2472d8840 "nm",
pretty_function=pretty_function@entry=0x55b2472d5fe0 <__func__.39677> "nm_lndp_ndisc_new",
expression=expression@entry=0x55b2472d5fa3 "NM_IS_L3CFG(config->l3cfg)")
at gmessages.c:2809
#4 0x000055b2471ce3fa in nm_lndp_ndisc_new (config=config@entry=0x7fff4041bb30)
at src/core/ndisc/nm-lndp-ndisc.c:680
#5 0x000055b247123b32 in _dev_ipac6_start (self=self@entry=0x55b248078360 [NMDeviceOvsInterface])
at src/core/devices/nm-device.c:11287
#6 0x000055b2471232f8 in _dev_ipac6_start_continue (self=0x55b248078360 [NMDeviceOvsInterface])
at src/core/devices/nm-device.c:11338
#7 0x000055b2471232f8 in _dev_ipll6_set_llstate (self=0x55b248078360 [NMDeviceOvsInterface],
llstate=<optimized out>, lladdr=<optimized out>) at src/core/devices/nm-device.c:10541
#8 0x000055b2471c9e8b in _emit_changed_on_idle_cb (user_data=user_data@entry=0x55b24807bdd0)
at src/core/nm-l3-ipv6ll.c:221
#9 0x00007f769734327b in g_idle_dispatch (source=0x55b248119200,
callback=0x55b2471c9ce0 <_emit_changed_on_idle_cb>,
user_data=0x55b24807bdd0) at gmain.c:5579
#10 0x00007f769734695d in g_main_dispatch (context=0x55b247f56bc0) at gmain.c:3193
#11 0x00007f769734695d in g_main_context_dispatch (context=context@entry=0x55b247f56bc0)
at gmain.c:3873
#12 0x00007f7697346d18 in g_main_context_iterate (context=0x55b247f56bc0,
block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at gmain.c:3946
#13 0x00007f7697347042 in g_main_loop_run (loop=0x55b247f320f0) at gmain.c:4142
#14 0x000055b246f26b64 in main (argc=<optimized out>,
argv=<optimized out>) at src/core/main.c:511
https://bugzilla.redhat.com/show_bug.cgi?id=2012934https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1044
Fixes-test: @ovs_cloned_mac_set_on_iface
If the ovs interface goes away, the ifindex gets zeroed out and l3cfg is
cleaned. Avoid starting ac6 in that case -- add checks similar to what
we do for ll6.
Bad things happen otherwise:
#0 0x00007f769734c895 in _g_log_abort (breakpoint=1) at gmessages.c:580
#1 0x00007f769734db98 in g_logv (log_domain=0x55b2472d8840 "nm",
log_level=G_LOG_LEVEL_CRITICAL, format=<optimized out>,
args=args@entry=0x7fff4041b9d0) at gmessages.c:1391
#2 0x00007f769734dd63 in g_log (log_domain=log_domain@entry=0x55b2472d8840 "nm",
log_level=log_level@entry=G_LOG_LEVEL_CRITICAL,
format=format@entry=0x7f769739a620 "%s: assertion '%s' failed") at gmessages.c:1432
#3 0x00007f769734e59d in g_return_if_fail_warning
(log_domain=log_domain@entry=0x55b2472d8840 "nm",
pretty_function=pretty_function@entry=0x55b2472d5fe0 <__func__.39677> "nm_lndp_ndisc_new",
expression=expression@entry=0x55b2472d5fa3 "NM_IS_L3CFG(config->l3cfg)")
at gmessages.c:2809
#4 0x000055b2471ce3fa in nm_lndp_ndisc_new (config=config@entry=0x7fff4041bb30)
at src/core/ndisc/nm-lndp-ndisc.c:680
#5 0x000055b247123b32 in _dev_ipac6_start (self=self@entry=0x55b248078360 [NMDeviceOvsInterface])
at src/core/devices/nm-device.c:11287
#6 0x000055b2471232f8 in _dev_ipac6_start_continue (self=0x55b248078360 [NMDeviceOvsInterface])
at src/core/devices/nm-device.c:11338
#7 0x000055b2471232f8 in _dev_ipll6_set_llstate (self=0x55b248078360 [NMDeviceOvsInterface],
llstate=<optimized out>, lladdr=<optimized out>) at src/core/devices/nm-device.c:10541
#8 0x000055b2471c9e8b in _emit_changed_on_idle_cb (user_data=user_data@entry=0x55b24807bdd0)
at src/core/nm-l3-ipv6ll.c:221
#9 0x00007f769734327b in g_idle_dispatch (source=0x55b248119200,
callback=0x55b2471c9ce0 <_emit_changed_on_idle_cb>,
user_data=0x55b24807bdd0) at gmain.c:5579
#10 0x00007f769734695d in g_main_dispatch (context=0x55b247f56bc0) at gmain.c:3193
#11 0x00007f769734695d in g_main_context_dispatch (context=context@entry=0x55b247f56bc0)
at gmain.c:3873
#12 0x00007f7697346d18 in g_main_context_iterate (context=0x55b247f56bc0,
block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at gmain.c:3946
#13 0x00007f7697347042 in g_main_loop_run (loop=0x55b247f320f0) at gmain.c:4142
#14 0x000055b246f26b64 in main (argc=<optimized out>,
argv=<optimized out>) at src/core/main.c:511
We need to first free "priv->bzobjs", which then will unlink all bzobjs
from the lists. The assert needs to go after.
https://bugzilla.redhat.com/show_bug.cgi?id=2028427
Fixes: 4154d9618c ('bluetooth: refactor BlueZ handling and let NMBluezManager cache ObjectManager data')
NMSettingOvsDpdk does not have a verify() implementation that would prevent
the devargs property from being NULL. We must thus anticipate and handle
a NULL value.
Fixes: ae4152120a ('ovs/ovsdb: add support for setting dpdk devargs option')
We get the hostname via D-Bus (from hostnamed) or read it from file.
In the latter case, it is not ensured that it's valid UTF-8.
Non-UTF-8 "strings" are bad, because we might try to expose them
on D-Bus, log them or other bad things.
Sanitize the string by using backslash escaping. Maybe we should
outright reject such binary nonsense, but it's not done here,
for no strong reasons.
We have at least static and transient hostnames. Let's be clear which
one we are talking about.
Note that also NM_SETTINGS_HOSTNAME gets renamed to
NM_SETTINGS_STATIC_HOSTNAME, because it seems clearer.
The only purpose of NM_SETTINGS_STATIC_HOSTNAME is to be the backing
property for the "Hostname" D-Bus property for the NMDBusObject glue.
So, while the new name makes more sense to me, it's now also
inconsistent with it's primary use (the D-Bus property). Still...
We avoid printing pointer values directly, instead we usually call
NM_HASH_OBFUSCATE_PTR(). This hashes the pointers with a random seed
so they are not directly visible.
That obviously makes it harder to debug. Add an environment variable
to disable that.
$ NM_OBFUSCATE_PTR=0 LIBNM_CLIENT_DEBUG=trace,stdout nmcli
Note that this flag is only honored in debug builds (WITH_MORE_ASSERTS>0).
$ nmcli connection modify dummy1 ethtool.feature-rx a
(process:3077356): GLib-WARNING **: GError set over the top of a previous GError or uninitialized memory.
This indicates a bug in someone's code. You must ensure an error is NULL before it's set.
The overwriting error message was: 'a' is not valid; use 'on', 'off', or 'ignore'
Error: failed to modify ethtool.feature-rx: 'a' is not valid; use [true, yes, on], [false, no, off] or [unknown].
Fixes: e5b46aa38a ('cli: use nmc_string_to_ternary() to parse ternary in _set_fcn_ethtool()')
When the device gets realized, similar to the situation that the device
is unmanaged by platform-init, if the device is still unmanaged by
parent and we clear the assume state. Then, when the device becomes
managed, NM is not able to properly assume the device using the UUID.
Therefore, we should not clear the assume state if the device has only
the NM_UNMANAGED_PLATFORM_INIT or the NM_UNMANAGED_PARENT flag set
in the unmanaged flags.
The previous commit 3c4450aa4d ('core: don't reset assume state too
early') did something similar for NM_UNMANAGED_PLATFORM_INIT flag only.
Routing daemons can add a large amount of routes to the
system. Currently NM receives netlink notifications for all those
routes and exposes them on D-Bus. With many routes, the daemon becomes
increasingly slow and uses a lot of memory.
The rtm_protocol field of the route indicates the source of the
route. From /usr/include/linux/rtnetlink.h, the allowed values are:
#define RTPROT_UNSPEC 0
#define RTPROT_REDIRECT 1 /* Route installed by ICMP redirects;
not used by current IPv4 */
#define RTPROT_KERNEL 2 /* Route installed by kernel */
#define RTPROT_BOOT 3 /* Route installed during boot */
#define RTPROT_STATIC 4 /* Route installed by administrator */
/* Values of protocol >= RTPROT_STATIC are not interpreted by kernel;
they are just passed from user and back as is.
It will be used by hypothetical multiple routing daemons.
Note that protocol values should be standardized in order to
avoid conflicts.
*/
#define RTPROT_GATED 8 /* Apparently, GateD */
#define RTPROT_RA 9 /* RDISC/ND router advertisements */
#define RTPROT_MRT 10 /* Merit MRT */
#define RTPROT_ZEBRA 11 /* Zebra */
#define RTPROT_BIRD 12 /* BIRD */
#define RTPROT_DNROUTED 13 /* DECnet routing daemon */
#define RTPROT_XORP 14 /* XORP */
#define RTPROT_NTK 15 /* Netsukuku */
#define RTPROT_DHCP 16 /* DHCP client */
#define RTPROT_MROUTED 17 /* Multicast daemon */
#define RTPROT_KEEPALIVED 18 /* Keepalived daemon */
#define RTPROT_BABEL 42 /* Babel daemon */
#define RTPROT_OPENR 99 /* Open Routing (Open/R) Routes */
#define RTPROT_BGP 186 /* BGP Routes */
#define RTPROT_ISIS 187 /* ISIS Routes */
#define RTPROT_OSPF 188 /* OSPF Routes */
#define RTPROT_RIP 189 /* RIP Routes */
#define RTPROT_EIGRP 192 /* EIGRP Routes */
Since NM uses only values <= RTPROT_STATIC, plus RTPROT_RA and
RTPROT_DHCP, add a BPF filter to the netlink socket to discard
notifications for other route types.
https://bugzilla.redhat.com/show_bug.cgi?id=1861527https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1038
This bumps L3_CONFIG_DATA_TYPE_MANUALIP to be the most important address
source; which is what had been the case before NetworkManager/next and
is presumably what the user expects.
It also comes into play for iBFT-booted machines, where iBFT contains a
permanent address (no lifetime data), while DHCP might lease out the
same one. In that case, expiry of the latter could potentially disrupt
connectivity to a vital storage volume.
Fixes: 14962cb414 ('merge: branch 'next''):
https://bugzilla.redhat.com/show_bug.cgi?id=2013921https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1011
When using the "remove" command on nmcli edit mode it will reset the
value to the default when no property value is specified. If the
property value is specified it will remove that specific property.
Example:
```
nmcli> set ethernet.wake-on-lan phy
nmcli> print ethernet.wake-on-lan
802-3-ethernet.wake-on-lan: phy, default
nmcli> remove ethernet.wake-on-lan default
nmcli> print ethernet.wake-on-lan
802-3-ethernet.wake-on-lan: phy
nmcli> remove ethernet.wake-on-lan
nmcli> print ethernet.wake-on-lan
802-3-ethernet.wake-on-lan: default
```
