summaryrefslogtreecommitdiff
path: root/src/libsystemd/sd-rtnl
AgeCommit message (Collapse)Author
2015-02-12Revert "tree-wide: Always use recvmsg with MSG_CMSG_CLOEXEC"Lennart Poettering
This reverts commit d6d810fbf8071f8510450dbacd1d083f37603656. It's apparently not OK to pass MSG_CMSG_CLOEXEC to recvmsg() of raw sockets.
2015-02-10tree-wide: Always use recvmsg with MSG_CMSG_CLOEXECCristian Rodríguez
2015-02-09sd-rtnl: fix typesystem for IFLA_AF_SPECTom Gundersen
Got this one wrong, it is not a union, just a nested container.
2015-02-08sd-rtnl: add support for IFLA_INET6_*Tom Gundersen
2015-02-08sd-rtnl: extend type system to allow address-family to decide the union membersTom Gundersen
So far we only supported selecting them by sibling attributes. (This stuff is all a bit crazy, but there seems to be no other way...)
2015-02-08sd-rtnl: don't treat 0-length messages speciallyTom Gundersen
Still parse the CMSG data, and most importantly make sure we drop the message when peeking.
2015-02-06sd-rtnl: don't fall over when receiving 0 bytes from the socketTom Gundersen
causes EOF. Seems like a kernel bug. Ignoring it seems to work be the best we can do for now... See https://bugs.freedesktop.org/show_bug.cgi?id=88397
2015-02-06sd-rtnl: fix creation of synthetic error repliesTom Gundersen
2015-02-03sd-rtnl: allow sd_rtnl_open_fd() to be used on bound socketsTom Gundersen
2015-02-02networkd: support socket activationTom Gundersen
Still keep the non-socket activation code around for starting from the commandline, but will likely drop that too in the future.
2015-02-02sd-rtnl: introduce sd_rtnl_new_from_netlinkTom Gundersen
2015-01-31sd-rtnl: don't fail event handler when callback failsTom Gundersen
As in sd-bus, simply log at debug level when a callback fails, but don't fail the event handler. Otherwise any error returned by any callback will disable the rtnl event handler. We should only do that on serious internal errors in sd-rtnl that we know cannot be recovered from.
2015-01-22networkd: Introduce ip6gre and ip6gretapSusant Sahani
This patch introduces ipv6 gre and gretap. test: ip6gre.netdev: [NetDev] Name=ip6gretap Kind=ip6gretap [Tunnel] Local=2a00:ffde:4567:edde::4987 Remote=2001:473:fece:cafe::5179 ip6gre.network: [Match] Name=eno16777736 [Network] Tunnel=ip6gretap ip link 6: ip6gre@eno16777736: <POINTOPOINT,NOARP> mtu 1448 qdisc noop state DOWN mode DEFAULT group default link/gre6 2a:00:ff:de:45:67:ed:de:00:00:00:00:00:00:49:87 peer 20:01:04:73:fe:ce:ca:fe:00:00:00:00:00:00:51:79
2015-01-22networkd: Introduce IP6 tunnelSusant Sahani
This patch enables networkd to create IP6 tunnels example conf: ipip6.netdev: [NetDev] Name=ipip6-tunnel Kind=ip6tnl [Tunnel] Mode=ip4ipv6 Local=2a00:ffde:4567:edde::4987 Remote=2001:473:fece:cafe::5179 ipip6.network [Match] Name=wlan0 [Network] Tunnel=ipip6-tunnel 23: ipip6-tunnel@wlan0: <POINTOPOINT,NOARP> mtu 1452 qdisc noop state DOWN mode DEFAULT group default link/tunnel6 2a00:ffde:4567:edde::4987 peer 2001:473:fece:cafe::5179
2015-01-22networkd: introduce gretapSusant Sahani
This patch introdeces gretap to networkd
2015-01-22Assorted format fixesZbigniew Jędrzejewski-Szmek
Types used for pids and uids in various interfaces are unpredictable. Too bad.
2015-01-22Fix some format strings for enums, they are signedZbigniew Jędrzejewski-Szmek
2015-01-19networkd: netdev - add ipvlan supportTom Gundersen
2015-01-13refcnt: refcnt is unsigned, fix comparisonsTom Gundersen
This does not make a difference, but the code was confusing.
2015-01-13nspawn: add new option "--port=" for exposing container ports on the local hostLennart Poettering
This exposes an IP port on the container as local port using DNAT.
2014-12-30tree-wide: spelling fixesVeres Lajos
https://github.com/vlajos/misspell_fixer https://github.com/torstehu/systemd/commit/b6fdeb618cf2f3ce1645b3315f15f482710c7ffa Thanks to Torstein Husebo <torstein@huseboe.net>.
2014-12-29rtnl: recv_message - don't enforce sender uidTom Gundersen
All we care about is that the kernel (pid==0) sent the message. Verifying the sender uid seems to break when using userns. Reported by Stéphane Graber.
2014-12-29sd-rtnl: rtnl_call - don't dispatch wqueue after timeout has passedTom Gundersen
Only a minor change as the timeout would be hit soon thereafetr at the next loop.
2014-12-29sd-rtnl: rtnl_poll - fix typoTom Gundersen
This caused rtnl_poll to always return true immediately in sd_rtnl_call().
2014-12-29sd-rtnl: recv_message - drop message when peeking failsTom Gundersen
Read the message form the socket or we will loop trying to read the same message repeatedly.
2014-12-29sd-rtnl: recv_message - don't fail on interruptTom Gundersen
We should just try again instead.
2014-12-29sd-rtnl: recv_message - log when dropping messageTom Gundersen
We drop messages received from the wrong uid/pid, log this at debug level.
2014-12-18networkd: add FDB supportAlin Rauta
2014-12-09sd-rtnl: fix size check in sd_rtnl_message_append_string()Lennart Poettering
2014-12-08networkd: manager - enumerate addresses globally, rather than per-linkTom Gundersen
The kernel always returns all addresses, rather than only for the given link, so let's only enumerate once.
2014-12-08sd-rtnl: fix bogus warning about dropping 20 bytes from multi-part messagesTom Gundersen
Nothing was being dropped, we just failed to account for the NLMSG_DONE.
2014-12-04sd-rtnl: rename sd_rtnl_message_route_get_{dst,src}_len to *_prefixlenTom Gundersen
This makes the API more consistent.
2014-12-04networkd: add support for source routingTom Gundersen
2014-12-04rtnl: order local addresses/gateways by the address itself as last resortLennart Poettering
This way we can be sure that the returned list is stable regarding modifications in the kernel.
2014-12-04rtnl: when querying local addresses and gateways, take address family into ↵Lennart Poettering
account
2014-12-03rtnl: make checks for default routes more strictLennart Poettering
Also check that the source netmask is 0, not only the destination netmask.
2014-12-03nss-myhostname: always resolve the host name "gateway" to the local default ↵Lennart Poettering
gateway This is useful inside of containers or local networks to intrdouce a stable name of the default gateway host (in case of containers usually the host, in case of LANs usually local router).
2014-12-02sd-rtnl: route - allow GETROUTE with AF_UNSPECTom Gundersen
2014-12-02sd-rtnl: add sd_rtnl_message_{new_neigh,neigh_get_{family,ifindex}}Tom Gundersen
2014-12-02sd-rtnl: add typesystem for RTM_*NEIGHTom Gundersen
2014-12-02sd-rtnl: add a bit more debugging in case a message is droppedTom Gundersen
2014-12-02sd-rtnl: message - allow checking for attributes without reading out their ↵Tom Gundersen
contents
2014-12-02sd-rtnl: add sd_rtnl_message_route_get_familyTom Gundersen
2014-11-28treewide: a few more log_*_errno + return simplificationsMichal Schmidt
The one in tmpfiles.c:create_item() even looks like it fixes a bug.
2014-11-28treewide: no need to negate errno for log_*_errno()Michal Schmidt
It corrrectly handles both positive and negative errno values.
2014-11-28treewide: auto-convert the simple cases to log_*_errno()Michal Schmidt
As a followup to 086891e5c1 "log: add an "error" parameter to all low-level logging calls and intrdouce log_error_errno() as log calls that take error numbers", use sed to convert the simple cases to use the new macros: find . -name '*.[ch]' | xargs sed -r -i -e \ 's/log_(debug|info|notice|warning|error|emergency)\("(.*)%s"(.*), strerror\(-([a-zA-Z_]+)\)\);/log_\1_errno(-\4, "\2%m"\3);/' Multi-line log_*() invocations are not covered. And we also should add log_unit_*_errno().
2014-11-27networkd: fix kernel rtnl receive buffer overrun errorAlin Rauta
We got the following error when running systemd on a device with many ports: "rtnl: kernel receive buffer overrun Event source 'rtnl-receive-message' returned error, disabling: No buffer space available" I think the kernel socket receive buffer queue should be increased. The default value is taken from: "/proc/sys/net/core/rmem_default", but we can overwrite it using SO_RCVBUF socket option. This is already done in networkd for other sockets. For example, the bus socket (sd-bus/bus-socket.c) has a receive queue of 8MB. In our case, the default is 208KB. Increasing the buffer receive queue for manager socket to 512KB should be enough to get rid of the above error. [tomegun: bump the limit even higher to 8M]
2014-11-04sd-event: rename sd_event_source_set_name() to sd_event_source_get_name()Lennart Poettering
To mirror the recent name change of the concept for sd_bus objects, follow the same logic for sd_event_source objects, too.
2014-10-29sd-rtnl: add bridge port rtnl attributes.Susant Sahani
Add bridge port attributes to sd-rtnl to configure via networkd.
2014-10-29sd-rtnl: add support to set packet family typeSusant Sahani
This patch adds functionality to set family type in the rtnl message for example PF_BRIDGE.