summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* dhcp/systemd: disable callback before stopping the clientbg/internal-dhcp-fixesBeniamino Galvani2016-01-251-2/+5
| | | | | Otherwise the call to stop() would generate a new unwanted state transition.
* dhcp/systemd: honor timeoutBeniamino Galvani2016-01-253-7/+19
|
* dhcp/systemd: gracefully fail when the interface has unknown ARP typeBeniamino Galvani2016-01-251-2/+9
|
* device: merge branch 'lr/default-unmanaged-bgo746566' (part 5)Thomas Haller2016-01-242-58/+78
|\ | | | | | | | | | | Merge another early part of 'lr/default-unmanaged-bgo746566' https://bugzilla.gnome.org/show_bug.cgi?id=746566
| * device: add _NM_DEVICE_CHECK_CON_AVAILABLE_FOR_USER_REQUEST enumThomas Haller2016-01-241-4/+6
| | | | | | | | | | | | NM_DEVICE_CHECK_CON_AVAILABLE_FOR_USER_REQUEST is a multi-flag value combining all the hooks that compose a user-request. Add a special value that has no esplicit meaning except that it ~is~ a user-request.
| * device: cleanup handling available-connectionsThomas Haller2016-01-241-46/+64
| | | | | | | | | | | | | | | | | | For update, don't delete first and add it again. Just do it in one step. For recheck, don't delete all connections first to add them all anew. Instead, check what changes and only emit the changed signal if there are any actual changes.
| * device/trivial: rename functions for available-connectionsThomas Haller2016-01-241-19/+19
|/
* core: refactor functions to use NM_UTILS_LOOKUP*()Thomas Haller2016-01-243-118/+64
| | | | | | Also use NM_UTILS_LOOKUP_STR() macro, which can transform unknown values to their numerical representation using alloca().
* core/trivial: rename NM_UTILS_STRING_LOOKUP_TABLE* to NM_UTILS_LOOKUP*Thomas Haller2016-01-243-81/+81
| | | | | Due to using a switch stetement, it is no longer only useful for looking up strings, and it certainly doesn't use a table anymore.
* utils: implement NM_UTILS_STRING_LOOKUP_TABLE_DEFINE() as switch() statementThomas Haller2016-01-241-6/+7
| | | | | | The compiler will likely optimize a switch() statement also to a table lookup, but it also works for negative values and values with large gaps.
* utils: use NM_UTILS_STRING_LOOKUP_TABLE_ITEM() macroThomas Haller2016-01-242-68/+68
|
* utils: add NM_UTILS_STRING_LOOKUP_TABLE_ITEM() macroThomas Haller2016-01-241-0/+3
|
* utils: fix out-of-bound access for negative indexes in ↵Thomas Haller2016-01-241-1/+1
| | | | _NM_UTILS_STRING_LOOKUP_TABLE_DEFINE()
* systemd: order NetworkManager.serivce after network-pre.targetThomas Haller2016-01-231-1/+1
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=761001
* build: install nm-settings-ifcfg-rh.5 man page conditionallyMichael Biebl2016-01-231-2/+7
| | | | | | | | | | Only install nm-settings-ifcfg-rh.5 man page if the ifcfg-rh configuration plugin has been enabled. But still add them to EXTRA_DIST so that the source tarball contains the generated man-file, but it isn't installed. https://bugzilla.gnome.org/show_bug.cgi?id=759823
* travis: fix bug in .travis.ymlThomas Haller2016-01-231-1/+1
| | | | Fixes: 1408b8c0a21105f3ea6d2e58d0fc03835f255d34
* core: merge branch 'bg/iptables-sharing-rules-bgo692673'Beniamino Galvani2016-01-233-10/+12
|\ | | | | | | | | | | | | Remove iptables rules for connection sharing upon termination to avoid the presence of duplicate rules when NM is restarted. https://bugzilla.gnome.org/show_bug.cgi?id=692673
| * manager: cleanup active connections upon exitBeniamino Galvani2016-01-231-0/+2
| | | | | | | | | | | | | | | | | | | | | | When connection sharing is enabled, the removal of iptables rules is delegated to the NMActRequest destructor; but for this to work it is required that the object is properly dereferenced upon NM termination. Clean up the active connections which are in DEACTIVATED state when quitting, so that they are unexported and destroyed. https://bugzilla.gnome.org/show_bug.cgi?id=692673
| * core: list iptables sharing rules in the right orderBeniamino Galvani2016-01-232-10/+10
|/ | | | | | | | | The rules were added to the list using g_slist_append() and then applied one at time using "iptables --insert" which puts them at the beginning of the chain, reversing the initial order. Instead, list them in the desired order and use g_slist_prepend() to achieve the same result. This has no functional changes.
* systemd: add chroot capabilityLubomir Rintel2016-01-221-1/+1
| | | | CAP_SYS_CHROOT is needed for openvpn hardening.
* travis: add coverity submissionLubomir Rintel2016-01-222-8/+22
| | | | | Also, decouple the build from the test run; it looks better in the .yml file as well as in the travis UI.
* build: merge branch 'th/tests-dbus-run-session'Thomas Haller2016-01-227-23/+21
|\ | | | | | | Update build script to prefer dbus-run-session over dbus-launch.
| * tests: use dbus-run-session for tools/run-test-valgrind.shThomas Haller2016-01-221-3/+8
| |
| * tests: allow fallback to dbus-launch in case dbus-run-session is not availableThomas Haller2016-01-221-0/+7
| | | | | | | | Like on Ubuntu 12.04 which is currently used by our Travic CI setup.
| * tests: move common dbus test-runners to tools/ directoryThomas Haller2016-01-225-9/+3
| |
| * tests: use dbus-run-session instead of dbus-launchMichael Biebl2016-01-222-10/+2
|/ | | | | | | | | The dbus-run-session utility was designed to run a process within a D-Bus session, specifically for running regressions tests and is much better suited then dbus-launch. As an additional benefit, this avoids any X dependencies. https://mail.gnome.org/archives/networkmanager-list/2016-January/msg00023.html
* ifcfg-rh: convert more testcases to g_assert()Dan Williams2016-01-221-304/+74
|
* libnm-core: fail verify() for NMSettingVlan for invalid vlan idThomas Haller2016-01-224-6/+24
| | | | | | | | | | | | | | | Point in case: # ip link add link dummy0 name dummy0.vlan type vlan id 4095 RTNETLINK answers: Numerical result out of range This potentially causes existing (invalid) connections to disappear as they now fail verification. Instead of adjusting the range of the GObject property NM_SETTING_VLAN_ID, reject it during vlan. This is a bit more forgiving to an older client that isn't aware of this new restriction, so he can first set the value without raising a critical warning.
* tui: disallow reserved vlan id 4095Thomas Haller2016-01-221-1/+1
|
* cli: disallow reserved vlan id 4095Thomas Haller2016-01-221-3/+3
|
* core: allow VLAN ids between 1 and 4095Beniamino Galvani2016-01-222-2/+3
| | | | | | | nm_utils_new_vlan_name() should reject the reserved VLAN id 4095, and the test case should not test reserved values 0 and 4095. Fixes: 778207f23eb2ac022b2ee47968a484fdb94626fe
* travis: move build script to a separate script and improve debugging for ↵Thomas Haller2016-01-222-1/+30
| | | | failed tests
* libnm: assert against valid dest in NMIPRouteThomas Haller2016-01-221-1/+7
|
* platform: fix leak when netlink dump is too largeDan Williams2016-01-211-0/+2
|
* core: truncate parent ifname when constructing virtual ifname for VLANThomas Haller2016-01-212-2/+69
| | | | https://bugzilla.redhat.com/show_bug.cgi?id=1300755
* core: check generated virtual interfaceplatform nameThomas Haller2016-01-213-22/+48
| | | | https://bugzilla.redhat.com/show_bug.cgi?id=1300755
* libnm-glib: fix leak in get_permissions_reply()Dan Williams2016-01-211-0/+2
|
* libnm,tests: fix error leakDan Williams2016-01-211-0/+1
|
* manager: fix a couple of memory leaksBeniamino Galvani2016-01-211-1/+4
| | | | | Fixes: 4dbaac4ba24ebc8b257fffe5197cc8e362804a58 Fixes: d8e1590c5058de91126bfa790c27aea944a2b596
* wifi/supplicant: refactor logging to use _LOG() macrosThomas Haller2016-01-211-54/+61
|
* wifi/supplicant: implement _LOG() macrosThomas Haller2016-01-211-1/+21
|
* utils: fix error open() handlingLubomir Rintel2016-01-211-1/+1
| | | | | CID 59376 (#2-1 of 2): Argument cannot be negative (NEGATIVE_RETURNS) negative_returns: urandom is passed to a parameter that cannot be negative.
* linux-platform: fix accidental PythonLubomir Rintel2016-01-211-1/+2
| | | | | | Coverity: CID 59378 (#1 of 1): Nesting level does not match indentation (NESTING_INDENT_MISMATCH) This statement is indented to column 41, as if it were nested within the preceding parent statement, but it is not.
* linux-platform: fix a file descriptor leakLubomir Rintel2016-01-211-0/+1
| | | | | Coverity: CID 59384 (#3-1 of 3): Resource leak (RESOURCE_LEAK)26. leaked_handle: Handle variable fd going out of scope leaks the handle
* wifi: merge branch 'th/supplicant-manager-fix-ref-count-rh1298007'Thomas Haller2016-01-216-199/+230
|\ | | | | | | https://bugzilla.redhat.com/show_bug.cgi?id=1298007
| * supplicant: cleanup GObject property implementation in NMSupplicantInterfaceThomas Haller2016-01-212-35/+83
| | | | | | | | | | nm_supplicant_interface_new() should be merely a convenient interface for creating the object directly.
| * wifi: fix crash due to wrong ownership handling in ↵Thomas Haller2016-01-214-87/+113
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | nm_supplicant_manager_iface_release() nm_supplicant_manager_iface_get() would cache and reuse the supplicant interface. But no ref-counting was in place so that the first user returning the interface via nm_supplicant_manager_iface_release() would destroy the instance for others. This is broken for a very long time. Which shows that we hardly ever have a cache-hit and usually create a new instance. So, instead of letting nm_supplicant_manager_create_interface() check for existing supplicant interface, always create a new instance. This also makes sense, because we would expect that per ifname only one instance is requested at a time. Also add an assertion that we don't return multiple supplicant interface instances for the same ifname. Drop nm_supplicant_manager_iface_release() in favor of requiring users to unref the returned instance. Also, use a GSList instead of a GHashTable for the cache. Also, previously callers would pass @is_wireless to nm_supplicant_manager_iface_get(), but the cache lookup did not consider that value. That doesn't matter now as we always create a new instance. https://bugzilla.redhat.com/show_bug.cgi?id=1298007
| * supplicant: don't pass start_now argument to nm_supplicant_interface_new()Thomas Haller2016-01-213-27/+22
| | | | | | | | | | | | Also, don't only consider the @die_count whether to start the supplicant, but check for is_available() -- which already considers @die_count and @running.
| * supplicant: drop is_wireless argument from NMSupplicantInterface's ↵Thomas Haller2016-01-211-8/+10
| | | | | | | | interface_add()
| * supplicant: remove check for iface instance after nm_supplicant_interface_new()Thomas Haller2016-01-211-5/+3
| | | | | | | | A plain constructor for a GObject cannot fail.
View Lorry Controller Status