**** BEGIN LOGGING AT Mon Feb 20 03:00:00 2012
Feb 20 06:45:36 <CIA-56> obsy * r30656 /packages/net/transmission/Makefile: [packages] transmission: use external upnpc library
Feb 20 09:33:28 <CIA-56> nbd * r30657 /trunk/target/linux/generic/ (4 files in 2 dirs): kernel: strip unnecessary symbol table information from kernel modules, reduces default squashfs size on ar71xx by about 4k
Feb 20 09:43:27 <dape> neat
Feb 20 09:57:43 <CIA-56> jow * r30658 /trunk/package/iwinfo/ (Makefile src/iwinfo_lib.c): [package] iwinfo: add more device entries
Feb 20 11:21:54 <CIA-56> nbd * r30659 /trunk/package/gpio-button-hotplug/ (. Makefile src/ src/Makefile src/gpio-button-hotplug.c): add gpio-button-hotplug, like button-hotplug + gpio-keys-polled, fully compatible, but without the nasty input-core dependency chain
Feb 20 11:21:59 <CIA-56> nbd * r30660 /trunk/target/linux/ar71xx/ (2 files in 2 dirs): ar71xx: switch to kmod-gpio-button-hotplug
Feb 20 11:59:31 <CIA-56> nbd * r30661 /packages/lang/python/Makefile: python: disable the readline dynload module even if the build system builds it
Feb 20 15:57:48 <builder_> build #129 of ps3 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/129
Feb 20 18:08:38 <CIA-56> nbd * r30662 /trunk/ (rules.mk scripts/rstrip.sh scripts/strip-kmod.sh): strip kernel modules more aggressively: make all global symbols local, rename all symbols in the symbol table to make them compress better
Feb 20 18:58:43 <btsimonh> anyone got experience of ettacap n
Feb 20 18:58:52 <btsimonh> & https?
Feb 20 20:00:43 <crow> nbd ping
Feb 20 20:01:08 <crow> you asked me yesterday if i can provide some serial console output of why nothing happens then "Booting the kernel" on ALIX.3D3.
Feb 20 20:01:29 <crow> it actually boot but there is just nothing after that msg on VGA... http://paste.pocoo.org/show/UF754qKaQpStEqBO3eeE/
Feb 20 20:02:13 <jow_laptop> same on virtualbox
Feb 20 20:02:23 <jow_laptop> grub ist the last thing printing to vga
Feb 20 20:02:27 <crow> yes
Feb 20 20:02:27 <jow_laptop> afterwards only serial
Feb 20 20:02:33 <jow_laptop> that why people assume it hangs
Feb 20 20:02:41 <nbd> ok, so the kernel doesn't even seem to have vga drivers
Feb 20 20:02:43 <jow_laptop> you could try to add console=tty0
Feb 20 20:02:45 <nbd> or it doesn't activate them
Feb 20 20:02:48 <nbd> console=tty0 is already in there
Feb 20 20:02:55 <jow_laptop> ok
Feb 20 20:03:07 <crow> http://www.abload.de/img/openwrt01rnux5.jpg
Feb 20 20:03:16 <crow> this is/was what i get
Feb 20 20:03:30 <crow> also i needed to press enter to get login (over serial)
Feb 20 20:04:47 <crow> but i am not able to log in over ssh or telnet
Feb 20 20:17:24 <crow> nbd on boot without interupting egg not pressing anything it stay on this: http://paste.pocoo.org/show/bzk69BEnAmZnRWvNPkQm/
Feb 20 20:17:32 <crow> if i press enter then it goes to login
Feb 20 20:23:00 <EqUaTe> having to press enter to get the local console is normal on openwrt
Feb 20 20:23:26 <EqUaTe> note line 119 of your paste crow - it says please press enter to activate this console
Feb 20 20:25:30 <crow> EqUaTe ok you are right about that, but still on VGA same as in picture. and somehow i am not able to ssh/telnet to it, and yes alix.3d3 have internet i can ping outside/internal just fine.. and netstat -atp show me telnet and ssh.
Feb 20 20:26:05 <EqUaTe> ok, so they seem to be running. you can only have one console= line in the kernel boot options
Feb 20 20:26:11 <EqUaTe> the last one will always take precedence
Feb 20 20:26:32 <EqUaTe> and if you don't have inittab set to start a tty on tty0, it won't
Feb 20 20:26:37 <EqUaTe> iirc, openwrt doesn't by default
Feb 20 20:26:50 <EqUaTe> as most devices don't /have/ video out, and thus tty0-6 are rather pointless
Feb 20 20:27:13 <EqUaTe> so the image of vga is also perfectly normal imo
Feb 20 20:27:22 <nbd> well, in older versions we did have both console=tty0 and console=ttyS0
Feb 20 20:27:28 <nbd> and it was working on both
Feb 20 20:27:31 <crow> if you check that picture above there is console=tty0 and serial
Feb 20 20:28:43 <crow> in inittab I see: tty1::askfirst:/bin/ash --login
Feb 20 20:29:06 <crow> so seems kernel line is then wrong as it point to tty0 ?
Feb 20 20:41:30 <nbd> i think tty0 is correct
Feb 20 20:44:14 <EqUaTe> hm, ok. in my experience the 2.6 kernels don't like that. but it can't hurt to try
Feb 20 20:44:20 <EqUaTe> and tty0 should be correct
Feb 20 20:44:37 <EqUaTe> crow: if you drop the serial one, do you get output on boot?
Feb 20 20:50:38 <crow> i am trying with tty0 in inittab
Feb 20 20:52:04 <crow> tty0::askfirst:/bin/ash --login is in inittab but still on VGA i see same as on picture
Feb 20 20:53:05 <nbd> inittab has nothing to do with it
Feb 20 20:53:13 <nbd> don't bother changing things in there
Feb 20 20:53:20 <crow> ok
Feb 20 20:53:46 <EqUaTe> nbd: openwrt doesn't use inittab to spawn tty's then?
Feb 20 20:56:22 <nbd> EqUaTe: it does, but it is much later than the kernel console assignment
Feb 20 20:56:32 <crow> nbd with what should i bother ?
Feb 20 20:56:54 <EqUaTe> nbd: so it's still a worthy mod, if only to get the login thing on the tty..
Feb 20 20:57:12 <nbd> EqUaTe: the primary goal is to get it to show something on the monitor
Feb 20 20:57:33 <nbd> EqUaTe: which it would do if it had actually brought up a framebuffer device
Feb 20 20:57:34 <EqUaTe> crow: i'd still try taking the console=ttyS0 line out of the boot options
Feb 20 20:57:40 <EqUaTe> and see what happens after
Feb 20 20:57:43 <EqUaTe> fair enough.
Feb 20 20:57:58 <nbd> so step 1 is figure out why no framebuffer driver is loading
Feb 20 20:58:01 <nbd> suspect number 1: kernel config
Feb 20 20:58:08 <nbd> check that
Feb 20 20:58:16 <crow> EqUaTe if i find where is that written i would try it.. there is no /grub/ .. and sorry first time using x86 from openwrt
Feb 20 20:58:27 <EqUaTe> np. i don't remember myself
Feb 20 20:58:31 <EqUaTe> it may actually be hardcoded.
Feb 20 20:58:35 <crow> nbd ok let me check the .config
Feb 20 21:03:39 <jow_laptop> crow: there is a boot partition (/dev/sda1 usually) containing the grub files
Feb 20 21:03:45 <jow_laptop> you must mount it to access them
Feb 20 21:07:58 <crow> jow_laptop ok thanks
Feb 20 21:48:11 <CIA-56> jow * r30663 /trunk/package/iwinfo/ (Makefile src/iwinfo_utils.c): [package] iwinfo: recognize AR5416 EEPROMs as seen on the Ubiquiti Picostation M2
Feb 20 21:57:52 <CIA-56> jow * r30664 /trunk/package/iwinfo/src/iwinfo_utils.c: [package] iwinfo: iwinfo_hardware_id_from_mtd() never writes, so do not request write access in mmap()
Feb 20 22:20:01 <crow> is there any default setting on x86 to disable telnet/dropbear? ssh: connect to host 192.168.1.10 port 22: Connection refused
Feb 20 22:21:53 <jow_laptop> no
Feb 20 22:22:02 <soma> is that your wan ip?
Feb 20 22:22:45 <crow> erm this device have only one eth0, and is connected to switch->router, i see it get statis lease from router (MAC rule)..
Feb 20 22:23:04 <jow_laptop> if it gets a lease its acting as wan
Feb 20 22:23:20 <jow_laptop> since lan on openwrt usually is statically configured
Feb 20 22:23:49 <crow> jow_laptop yea its wan then, asn eth0 is in wan part and its dhcp
Feb 20 22:23:59 <jow_laptop> and whats in lan?
Feb 20 22:24:24 <crow> jow_laptop eth1, bridge,statis,192.168.1.1
Feb 20 22:24:36 <jow_laptop> but your alix variant only has one eth?
Feb 20 22:24:39 <crow> yes
Feb 20 22:24:42 <jow_laptop> I see
Feb 20 22:24:48 <jow_laptop> uci del network.wan.ifname
Feb 20 22:24:57 <jow_laptop> uci set network.lan.ifname=eth0
Feb 20 22:25:00 <jow_laptop> uci commit network
Feb 20 22:25:04 <jow_laptop> reboot
Feb 20 22:25:07 <crow> ifconfig show me correct ip.. and i am trying to connect to that one
Feb 20 22:25:08 <crow> ok
Feb 20 22:25:35 <jow_laptop> it will be at 192.168.1.1 then
Feb 20 22:25:42 <jow_laptop> might conflict with existing gear
Feb 20 22:26:04 <jow_laptop> in this case;  uci set network.lan.ipaddr=192.168.1.whatever; uci commit network; ifup lan
Feb 20 22:26:23 <jow_laptop> or use your favorite editor to modify /etc/config/network
Feb 20 22:26:33 <crow> ok
Feb 20 22:26:36 <jow_laptop> which could be tedious over serial so I give the uci commands here
Feb 20 22:27:46 <crow> cool thanks
Feb 20 22:27:47 <crow> it worked
Feb 20 22:29:38 <crow> also there are some warning after ifup lan : http://dpaste.com/705954/ its just fyi..
Feb 20 22:29:59 <jow_laptop> yeah its harmless
Feb 20 22:30:48 <crow> let me now mount sda1 and try without two console=
Feb 20 22:33:13 * russell-- has had some trouble with recent openvpn on openwrt 
Feb 20 22:33:49 <jow_laptop> revent as in openvpn-devel or a recent as in latest openvpn package version?
Feb 20 22:34:04 <jow_laptop> -v +c
Feb 20 22:34:24 <russell--> it seems that the ifconfig <iface> 0.0.0.0 occurs after the interface has disappeared, as a result openvpn exits and then fails to recreate the tunnel
Feb 20 22:34:47 <russell--> regular openvpn i think
Feb 20 22:34:55 <jow_laptop> you mean hotplug is interfering?
Feb 20 22:35:13 <russell--> when our server restarts or disappears, all the clients just die
Feb 20 22:35:40 <russell--> the logs seem to imply an external command exited with non-zero
Feb 20 22:35:50 <crow> EqUaTe just fyi i tried with only console=tty0 (removed that serial part in grub) and still same, nothing after Booting the kernel..
Feb 20 22:36:02 <russell--> Closing TUN/TAP interface
Feb 20 22:36:13 <russell--> /sbin/ifconfig ptp 0.0.0.0
Feb 20 22:36:24 <russell--> Linux ip addr del failed: external program exited with error status: 1
Feb 20 22:36:39 <jow_laptop> hm
Feb 20 22:36:40 <russell--> Note: Cannot open TUN/TAP dev /dev/net/tun: Permission denied (errno=13)
Feb 20 22:36:55 <russell--> etc
Feb 20 22:37:16 <jow_laptop> openvpn itself didn't change I believe
Feb 20 22:37:20 <russell--> our older clients seem to recover okay
Feb 20 22:37:37 <russell--> where older == ancient r18xxx something
Feb 20 22:38:07 <russell--> we just has a weekend of server failures so we've had some practice
Feb 20 22:38:18 <russell--> had*
Feb 20 22:38:38 <jow_laptop> well I believe it is a hotplug issue
Feb 20 22:38:57 <jow_laptop> are the tun/tap interfaces uci managed? do you have a config interface section for it?
Feb 20 22:39:08 <russell--> no
Feb 20 22:39:52 <russell--> the client gets a PUSH or something from the server with the network config
Feb 20 22:40:02 <jow_laptop> did you try to trace yet whether anything in /etc/hotplug.d/net/ fires for tun/tap ?
Feb 20 22:40:15 <russell--> no, just starting to look at the problem
Feb 20 22:40:18 <jow_laptop> ok
Feb 20 22:40:33 <jow_laptop> though this ifconfig ptp command looks odd
Feb 20 22:40:42 <russell--> ptp is our interface name
Feb 20 22:40:45 <jow_laptop> ah
Feb 20 22:40:54 <danmackay> hmm
Feb 20 22:40:56 <russell--> we == personaltelco project
Feb 20 22:41:03 * danmackay should setup extroot again and fix my ipv6
Feb 20 22:41:18 <danmackay> or a usb overlay
Feb 20 23:16:40 <CIA-56> hauke * r30665 /trunk/package/mac80211/patches/805-fix_gpio_bcm5354.patch:
Feb 20 23:16:40 <CIA-56> mac80211: fix memory leak on bcm5354
Feb 20 23:16:40 <CIA-56> The ucode triggers the gpio pins used for the buttons and causes diag to handle them and send them to userspace.
Feb 20 23:16:40 <CIA-56> This only happened with a more recent firmware.
Feb 20 23:16:40 <CIA-56> This patch was tested on an Asus wl-520GU and should also work on the other bcm5354 based asus devices. I do not know if it works on the dir320.
Feb 20 23:18:30 <danmackay> yuck dlink routers :p
Feb 20 23:36:07 <builder_> build #132 of cobalt is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/132
Feb 21 00:12:03 <CIA-56> nbd * r30666 /trunk/package/hostapd/ (Config.in Makefile patches/770-limit_debug_messages.patch): hostapd: add configurable debug message minimum priority to cut down on bloat generated by excessive debug messages
Feb 21 00:40:50 <Epsylon3> any idea why my leds are no more exported to sysfs in 3.0.18 ?
Feb 21 00:40:57 <Epsylon3> config is there, its weird
Feb 21 00:57:42 <CIA-56> nbd * r30667 /trunk/package/hostapd/files/ (hostapd-full.config hostapd-mini.config): hostapd: disable the state dump callback to save a few more kb
Feb 21 01:22:54 <CIA-56> nbd * r30668 /trunk/package/dropbear/patches/140-disable_assert.patch: dropbear: disable assertions to reduce code size
**** ENDING LOGGING AT Tue Feb 21 02:59:58 2012