[SOLVED] SSH won't start/accept connection for a long time

[RobMeades]

Hi there.

A few days ago I somewhat speculatively attempted to install dkms-mt7601 on my Hummingboard II to get an RALink-chipset wifi adapter working. This failed, with some armv7 headers not found I think, but that's not what I want to ask about. I did pacman -Syu, just in case my system was out of date (I had not done so for a few weeks), and that seemed to work fine. However, now I have the problem that I can only log-in via SSH over a wired Ethernet connection (my usual mode of working with this device) if I either:

(a) log-in with screen/keyboard first, or
(b) wait for ~93 seconds after boot.

SSHing into the Hummingboard II from a PC gives me the immedate error message "connection refused" (note: this is not a case of slow SSH log-in, the connection simply fails). If I then just watch a screen connected to the Hummingboard II, approximately 93 seconds after boot it spits out the following:

$this->bbcode_second_pass_code('', '
[ 93.997254] audit: type=1006 audit(93.996:2): pid=497 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=1 res=1
[ 94.157007] audit: type=1006 audit(94.155:3): pid=501 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=2 res=1
[ 94.543658] audit: type=1006 audit(94.542:4): pid=513 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=3 res=1
')
...and after that I can SSH-in from a PC OK. So the tail of dmesg looks like this:

$this->bbcode_second_pass_code('', '
[ 6.120275] fec 2188000.ethernet eth0: Freescale FEC PHY driver [Atheros 8035 ethernet] (mii_bus:phy_addr=2188000.ethernet:00, irq=-1)
[ 6.168562] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 10.127959] libphy: 2188000.ethernet:00 - Link is Up - 1000/Full
[ 10.140254] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 93.997254] audit: type=1006 audit(93.996:2): pid=497 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=1 res=1
[ 94.157007] audit: type=1006 audit(94.155:3): pid=501 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=2 res=1
[ 94.543658] audit: type=1006 audit(94.542:4): pid=513 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=3 res=1]
')
Nothing happens for ~84 seconds, during which time all remote SSH log-in attempts fail. I've looked up the audit message type 1006 and it is to do with log in attempts, however they still appear if I do absolutely no login attempts during this time. journalctl doesn't show any errors during boot, systemctl shows no failed items (though of course I can only check this once I've logged in and by that time everything is working anyway).

Are there any other log files I should be looking at?

Rob

[RobMeades]

SOLVED - in my messing around with the wifi adapter I had ended up with a:

netctl enable wlan0-blah

...service still running, even without a wlan0. All my "Wants" multi-user.target services were still hung off this and hence didn't run until it timed out. I disabled this netctl and all is now running normally once more.

Rob