netdev - RE: team driver MAC addresses questions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <65502A17874025489B7D8479C16CFA634E91D236@nasanexd02f.na.qualcomm.com>
Date:	Wed, 12 Feb 2014 14:17:10 +0000
From:	qca_dlansky <qca_dlansky@...lcomm.com>
To:	Jiri Pirko <jiri@...nulli.us>
CC:	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	qca_vkondrat <qca_vkondrat@....qualcomm.com>
Subject: RE: team driver MAC addresses questions

>Q2:
>We've noticed the following behavior:
>- wlan1 has some MAC address - m1
>- team0 created with random MAC address - m2
>
>root@...foo:~# ip link
>1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT 
>    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 1000
>    link/ether 00:21:70:be:4f:68 brd ff:ff:ff:ff:ff:ff
>4: wlan1: <BROADCAST,MULTICAST> mtu 1500 qdisc mq state DOWN mode DORMANT qlen 1000
>    link/ether 44:39:c4:18:b5:7a brd ff:ff:ff:ff:ff:ff
>12: team0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT 
>    link/ether 9a:ac:d5:91:df:aa brd ff:ff:ff:ff:ff:ff
>
>- When adding port wlan1 into team0, m1 gets assigned to team0 and m2 
>gets assigned to wlan1
>
>root@...foo:~# LD_LIBRARY_PATH=/usr/local/lib/ teamdctl team0 port add 
>wlan1 root@...foo:~# ip link
>1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT 
>    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 1000
>    link/ether 00:21:70:be:4f:68 brd ff:ff:ff:ff:ff:ff
>4: wlan1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master team0 state UP mode DORMANT qlen 1000
>    link/ether 9a:ac:d5:91:df:aa brd ff:ff:ff:ff:ff:ff
>12: team0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT 
>    link/ether 44:39:c4:18:b5:7a brd ff:ff:ff:ff:ff:ff
>
>This is a bit strange for us. We were expecting both of them to have the same MAC address.
>
>Further investigation showed that when adding the port, the MAC address of the added port is set twice:
>once by team_add_slave(), when wlan1 is still down. The MAC address is set to m2.
>second time by team_set_mac_address(), after wlan1 is up. The MAC address is set to m1.

Are you using teamd? What config?

[DL]  just 'teamd -d', no config at all

>
>However, the 2nd call fails which leaves wlan1 with m2. The reason for the failure is because ath9k doesn't support IFF_LIVE_ADDR_CHANGE.
>
>Our question: in team_set_mac_address(), in case there is a failure setting MAC address for any port inside port_list, doesn't it make sense to undo MAC address changes and report the failure up the stack?
>

Yes. That would be probably the proper thing to do. WOuld you send the patch or I should do it? Would be needed to do something similar what bonding does.

[DL] Let us think about it. We want to prepare patch, but we need time to do so.

>Thanks,
>  Vladimir & Dedy.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html