Re: [OpenWrt-Devel] ttyUSB identification
I started adding USB ids to a shell script which echos the USB product / ID into /sys/bus/usb-serial/drivers/option1/new_id if appropriate. Unfortunately, you can receive many hotplug events for a single device insertion so I wrote a program which will "compress" all of the matching events into a single event, and call /sbin/usb-added with the USB product / id which was added. It's fairly clean, but it is by no means complete. -M On Mon, 2010-10-25 at 18:20 +0200, Daniel Golle wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > hi benjamin! > afaik the creation of /dev/serial, /dev/disk, ... symlinks is done by > udev and therefore requires udev installed. > while this could solve the problem, it occupies 166kB (which is more > than all needed USB-serial kernel modules together...) > i tried that anyway. however, after installing the udev package, udevd > doesn't get started during boot and this seems to be an architectural > problem already discussed a couple of years ago (see @9412). starting > udevd manually and then re-plugging the usb wwan does create the > symlink, but this doesn't solve the problem. (i.e. hotplug after udevd > is started works, but coldpluged devices are not getting processed by > udevd as it is started afterward) > according to it's author, running hotplug2 and udevd together is not > really a good idea anyway, see > http://svn.nomi.cz/svn/isteve/hotplug2/README. > > i guess i should get hotplug2 to create those symlinks then. in order to > be able to do this, hotplug2 has to be extended to provide the info from > sysfs (path inside /sys/devices, idVendor, idProduct) to the > rule-scripts which could then create meaningful persistent symlinks. > everybody agree so far? any better way to do the same? > > cheers > > daniel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] [PATCH] Set auth_algs in hostapd.conf, make it configurable for WEP
Comment...
On 10/22/10 7:28 PM, Mark Mentovai wrote:
Always set auth_algs in hostapd.conf. For WEP, auth_algs is configurable
using a new UCI parameter, auth_algs, which may contain "open",
"shared", or both. For other encryption types, auth_algs is always set
to 1 (open).
https://dev.openwrt.org/ticket/8120
Signed-off-by: Mark Mentovai
---
Index: package/hostapd/files/hostapd.sh
===
--- package/hostapd/files/hostapd.sh(revision 23587)
+++ package/hostapd/files/hostapd.sh(working copy)
@@ -77,7 +77,6 @@
append "$var" "eapol_key_index_workaround=1" "$N"
append "$var" "radius_acct_interim_interval=300" "$N"
append "$var" "ieee8021x=1" "$N"
- append "$var" "auth_algs=1" "$N"
append "$var" "wpa_key_mgmt=WPA-EAP" "$N"
append "$var" "wpa_group_rekey=300" "$N"
append "$var" "wpa_gmk_rekey=640" "$N"
@@ -110,6 +109,34 @@
;;
esac
append "$var" "wpa=$wpa" "$N"
+
+ case "$enc" in
+ *wep*)
+ config_get auth_algs "$vif" auth_algs
+ auth_alg_open=
+ auth_alg_shared=
+ for auth_alg in ${auth_algs:-open}; do
+ case "$auth_alg" in
+ open)
+ auth_alg_open=1
+ ;;
+ shared)
+ auth_alg_shared=1
+ ;;
+ esac
+ done
+ auth_alg_val=0
+ [ -n "$auth_alg_open" ]&& \
+ auth_alg_val=1
+ [ -n "$auth_alg_shared" ]&& \
+ auth_alg_val=$(expr "$auth_alg_val" + 2)
Just curious: why not use:
let -i auth_alg_val+=2
here instead?
+ ;;
+ *)
+ auth_alg_val=1
+ ;;
+ esac
+ append "$var" "auth_algs=$auth_alg_val" "$N"
+
[ -n "$crypto" ]&& append "$var" "wpa_pairwise=$crypto" "$N"
[ -n "$wpa_group_rekey" ]&& append "$var" "wpa_group_rekey=$wpa_group_rekey"
"$N"
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] [PATCH] w_scan package
This patch defines a new package for w_scan: "w_scan is a small command line utility used to perform frequency scans for DVB and ATSC transmissions." See: http://wirbel.htpc-forum.de/w_scan/index_en.html The package includes w_scan but not the w_scan_start.sh referred to in the documentation. Signed-off-by: Mike Brady --- I have included the patch as an attachment -- I hope that's OK. >From the "packages" directory the patch adds one directory which contains one Makefile: + utils/w_scan/ + utils/W_scan/Makefile Index: utils/w_scan/Makefile === --- utils/w_scan/Makefile (revision 0) +++ utils/w_scan/Makefile (revision 0) @@ -0,0 +1,46 @@ +# +# Copyright (C) 2006-2010 OpenWrt.org +# +# This is free software, licensed under the GNU General Public License v2. +# See /LICENSE for more information. +# + +include $(TOPDIR)/rules.mk + +PKG_NAME:=w-scan +PKG_VERSION:=20101001 +PKG_RELEASE:=1 + +PKG_SOURCE:=w_scan-$(PKG_VERSION).tar.bz2 +PKG_SOURCE_URL:=http://wirbel.htpc-forum.de/w_scan +PKG_BUILD_DIR:=$(BUILD_DIR)/w_scan-$(PKG_VERSION) + +include $(INCLUDE_DIR)/package.mk + +define Package/w_scan + SECTION:=utils + CATEGORY:=Utilities + DEPENDS:=...@linux_2_6 + DEFAULT:=n + TITLE:=w_scan performs frequency scans for DVB and ATSC. + URL:=http://wirbel.htpc-forum.de/w_scan/index2.html + MAINTAINER:=Mike Brady +endef + +define Package/w_scan/description + w_scan is a small command line utility used to perform frequency scans for DVB and ATSC transmissions. It is capable of creating channels.conf files (in different output formats !) as well as initial tuning data for scan. It's based on the utility scan from linuxtv-dvb-apps, but meanwhile it was heavily changed and has different features: + 1. no need for "initial transponders", and such no need for a config file + 2. w_scan finds automatically the best matching dvb card + 3. complete different command line options + 4. a different default output format as well as bunch of output formats. +This package does not include the script file "w_scan_start.sh". +endef + +define Package/w_scan/install + $(INSTALL_DIR) $(1)/usr/bin + $(CP) \ + $(PKG_BUILD_DIR)/w_scan \ + $(1)/usr/bin/ +endef + +$(eval $(call BuildPackage,w_scan)) ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] ttyUSB identification
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 hi benjamin! afaik the creation of /dev/serial, /dev/disk, ... symlinks is done by udev and therefore requires udev installed. while this could solve the problem, it occupies 166kB (which is more than all needed USB-serial kernel modules together...) i tried that anyway. however, after installing the udev package, udevd doesn't get started during boot and this seems to be an architectural problem already discussed a couple of years ago (see @9412). starting udevd manually and then re-plugging the usb wwan does create the symlink, but this doesn't solve the problem. (i.e. hotplug after udevd is started works, but coldpluged devices are not getting processed by udevd as it is started afterward) according to it's author, running hotplug2 and udevd together is not really a good idea anyway, see http://svn.nomi.cz/svn/isteve/hotplug2/README. i guess i should get hotplug2 to create those symlinks then. in order to be able to do this, hotplug2 has to be extended to provide the info from sysfs (path inside /sys/devices, idVendor, idProduct) to the rule-scripts which could then create meaningful persistent symlinks. everybody agree so far? any better way to do the same? cheers daniel On 10/25/2010 02:21 PM, Benjamin Henrion wrote: > On Mon, Oct 25, 2010 at 1:58 PM, Daniel Golle wrote: >> -BEGIN PGP SIGNED MESSAGE- >> Hash: SHA1 >> >> hi everybody! >> for my current project i need to reliably identify/distinguish several >> USB-to-serial devices connected to a 4-port hub. in a desktop >> environment, i would use hal over dbus (or in future probably device-kit >> over dbus) to select the device by >> vendor/product/serialno/connected-port. i was thinking of using udev to >> create symbolic links, but even udev seems like taking a sledgehammer to >> crack a nut to me... >> my problem is that, depending on whether i cold-start or reboot the >> system (a fon2202 btw), the different USB-to-serial gadgets end up to be >> different ttyUSBx devices. especially the huawei 3g/umts wwan seems to >> take a greatly varying amount of time before it becomes ready, so on a >> reboot it becomes visible to the system as ttyUSB0 & ttyUSB1, in a >> cold-start it's ttyUSB2 & ttyUSB3... >> what way do you suggest to solve the problem? > > Address the devices by going into the /dev/serial/by-id : > > zoo...@buzek /dev/serial/by-id [18]$ l > total 0 > lrwxrwxrwx 1 root root 13 2010-10-25 09:17 > usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if01 > -> ../../ttyACM0 > lrwxrwxrwx 1 root root 13 2010-10-25 09:17 > usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if03 > -> ../../ttyACM1 > lrwxrwxrwx 1 root root 13 2010-10-25 09:17 > usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if09 > -> ../../ttyACM2 > > The path there should be unique. > > Try to insert all the adaptors, and send me the output of "ls -l > /dev/serial/by-id". > > -- > Benjamin Henrion > FFII Brussels - +32-484-566109 - +32-2-4148403 > "In July 2005, after several failed attempts to legalise software > patents in Europe, the patent establishment changed its strategy. > Instead of explicitly seeking to sanction the patentability of > software, they are now seeking to create a central European patent > court, which would establish and enforce patentability rules in their > favor, without any possibility of correction by competing courts or > democratically elected legislators." > ___ > openwrt-devel mailing list > openwrt-devel@lists.openwrt.org > https://lists.openwrt.org/mailman/listinfo/openwrt-devel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJMxa48AAoJELd9RAts2fh5sZ0H/RRoSlNR7n2b7CHGVs06fZFS nw49Ze5YSqARr+6SmuhmOtin+CA/vdZotR0oExEeMo4JL6FoKwKo3n0fiOlhPHxp bmDJqt9GFadHG9NnwdpDTXE9D0Z3C7AFdC7av2MmV7sqpQNqqR3Of+emEq0AcuOI OT9XW7BO7niCdM3FfSubHAMpUFim23bT3qTE7+iVKa4yIFDrO2oJ1j9ToeIynNkv PvgCa5AZCIWpNJRow2jgBfQQmWSxtoQObEHHRhVCyL1bz6k3a5cnFI28Nw4Z6OnS /mr28kTy2P9Q4x58wPGtS06SJArsX7hEqQnEDvgd4QKOWFysCdUKwNpd8QqXQVw= =/OjV -END PGP SIGNATURE- ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] [PATCH 2/3] ZNC: Package modules.
whoa, my bad, that was the wrong patch, dislexia Committed revision 23632. dislexia set in from lack of sleep On Mon, Oct 25, 2010 at 10:37 AM, Jonas Gorski < jonas.gorski+open...@gmail.com > wrote: > The patches aren't fully independent, patch 3 depends on patch 2 being > applied (which looks to be missing here). > > On 25 October 2010 16:25, Outback Dingo wrote: > > Okay, heres the output of patching patch 1 files against trunk packages > > patch -p1 < patch1 > > patching file net/znc/Makefile > > Hunk #1 FAILED at 29. > > Hunk #2 FAILED at 39. > > Hunk #3 succeeded at 57 with fuzz 1 (offset -331 lines). > > 2 out of 3 hunks FAILED -- saving rejects to file net/znc/Makefile.rej > > patching file net/znc/files/znc.conf > > patching file net/znc/files/znc.init > > di...@dingo-laptop:/usr/builder/devel/packages$ cat net/znc/Makefile.rej > > --- net/znc/Makefile > > +++ net/znc/Makefile > > @@ -29,7 +29,7 @@ > > > > define Package/znc > >$(Package/znc/default) > > - DEPENDS:=+libopenssl +uclibcxx > > + DEPENDS:=+libopenssl +uclibcxx +coreutils-su > >MENU:=1 > > endef > > > > @@ -39,6 +39,10 @@ > > bouncing, and c++ module support to name a few. > > endef > > > > +define Package/znc/conffiles > > +/etc/config/znc > > +endef > > + > > define Package/znc-mod-admin > >$(Package/znc/default) > >TITLE+= (admin plugin) > > > > > > > > On Sat, Oct 23, 2010 at 6:32 PM, Jonas Gorski > > > > wrote: > >> > >> Create packages for all modules except webadmin, since it doesn't > compile. > >> The extra modules are ignored for now. > >> > >> Signed-off-by: Jonas Gorski > >> > > > >> --- > >> net/znc/Makefile | 526 > >> +- > >> 1 files changed, 524 insertions(+), 2 deletions(-) > >> > >> diff --git a/net/znc/Makefile b/net/znc/Makefile > >> index e7b7243..08364d4 100644 > >> --- a/net/znc/Makefile > >> +++ b/net/znc/Makefile > >> @@ -19,21 +19,351 @@ PKG_BUILD_PARALLEL:=1 > >> > >> include $(INCLUDE_DIR)/package.mk > >> > >> -define Package/znc > >> +define Package/znc/default > >> SUBMENU:=Instant Messaging > >> SECTION:=net > >> CATEGORY:=Network > >> - DEPENDS:=+libopenssl +uclibcxx > >> TITLE:=ZNC > >> URL:=http://en.znc.in/ > >> endef > >> > >> +define Package/znc > >> + $(Package/znc/default) > >> + DEPENDS:=+libopenssl +uclibcxx > >> + MENU:=1 > >> +endef > >> + > >> define Package/znc/description > >>ZNC is an IRC bouncer with many advanced features like detaching, > >>multiple users, per channel playback buffer, SSL, IPv6, > transparent > >> DCC > >>bouncing, and c++ module support to name a few. > >> endef > >> > >> +define Package/znc-mod-admin > >> + $(Package/znc/default) > >> + TITLE+= (admin plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-admin/description > >> + Allows you to add/remove/edit users and settings on the fly via IRC > >> messages. > >> +endef > >> + > >> +define Package/znc-mod-adminlog > >> + $(Package/znc/default) > >> + TITLE+= (adminlog plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-adminlog/description > >> + Log user connects and disconnects and failed logins to file or syslog. > >> +endef > >> + > >> +define Package/znc-mod-autoattach > >> + $(Package/znc/default) > >> + TITLE+= (autoattach plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-autoattach/description > >> + Reattaches you to channels on activity. > >> +endef > >> + > >> + > >> +define Package/znc-mod-autocycle > >> + $(Package/znc/default) > >> + TITLE+= (autocycle plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-autocycle/description > >> + Cycles a channel when you are the only one in there and you don't have > >> op. > >> +endef > >> + > >> +define Package/znc-mod-autoop > >> + $(Package/znc/default) > >> + TITLE+= (autoop plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-autoop/description > >> + Auto op the good guys. > >> +endef > >> + > >> +define Package/znc-mod-autoreply > >> + $(Package/znc/default) > >> + TITLE+= (autoreply plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-autoreply/description > >> + Gives a automatic reply if someone messages you if you are away. > >> +endef > >> + > >> +define Package/znc-mod-away > >> + $(Package/znc/default) > >> + TITLE+= (away plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-away/description > >> + Stores messages while away, also auto away. > >> +endef > >> + > >> +define Package/znc-mod-awaynick > >> + $(Package/znc/default) > >> + TITLE+= (awaynick plugin) > >> + DEPENDS:=znc > >> +endef > >> + > >> +define Package/znc-mod-awaynick/description > >> + Change your nick while you are away. > >> +endef > >> + > >> +define Package/znc-mod-blockuser > >> + $(Package/znc/default) > >> + TITLE+= (blockuser plugin) > >> + DEPENDS:=znc > >>
Re: [OpenWrt-Devel] [PATCH] Split wing compilation into host and target
Committed revision 23631.
On Thu, Oct 21, Committed revision 23631.
2010 at 2:13 PM, Roberto Riggio wrote:
>
> This patch:
>
> * splits the wing package compilation into host and target
> * upgrades the git revision to a more recent version of the protocol
> which solves a segfault on arm boards and adds support for OLSR-like
> Host-Network-Association messages.
>
> Signed-off-by: Roberto Riggio
>
> --
> Index: net/wing/files/lib/network/wing.sh
> ===
> --- net/wing/files/lib/network/wing.sh(revision 23569)
> +++ net/wing/files/lib/network/wing.sh(working copy)
> @@ -36,7 +36,7 @@
> exit 1
> fi
>
> -local profile rc ls prefix debug
> +local profile rc ls metric prefix period tau debug
>
> config_get profile $config profile "bulk"
> config_get rc $config rc "static"
> @@ -131,8 +131,12 @@
> config_get ifname $1 ifname
> config_get device $1 device
> config_get hwmode $device hwmode "11bg"
> -freq=$(iwlist $ifname freq | grep "Current Frequency" | sed -n
> "s/^.*Current Frequency:\([0-9.]*\) GHz (Channel \([0-9]*\))/\1/p" | awk
> '{print $1*1000}')
> -freq=${freq:-"0"}
> +config_get channel $device channel "0"
> +[ "$channel" = "0" -o "$channel" = "auto" ] && {
> +logger -t "$device" "Channel not specified. Ignoring."
> +return 0
> +}
> +freq=$(iwlist $ifname freq | sed -n "s/^.*Channel 0*$channel :
> \([0-9.]*\).*/\1/p" | awk '{print $1*1000}')
> hwaddr=$(/sbin/ifconfig $ifname 2>&1 | sed -n 's/^.*HWaddr
> \([0-9A-Za-z\-]*\).*/\1/p' | sed -e 's/\-/:/g' | cut -c1-17)
> freqs=${freqs:+"$freqs "}$freq
> hwmodes=${hwmodes:+"$hwmodes "}$hwmode
> Index: net/wing/patches/100-cross-compile.patch
> ===
> --- net/wing/patches/100-cross-compile.patch(revision 23569)
> +++ net/wing/patches/100-cross-compile.patch(working copy)
> @@ -1,21 +0,0 @@
>
> - m4/click.m4 |8
> - 1 file changed, 8 deletions(-)
> -
> wing-20100903.orig/m4/click.m4
> -+++ wing-20100903/m4/click.m4
> -@@ -235,14 +235,6 @@ AC_DEFUN([CLICK_CHECK_BUILD_DYNAMIC_LINK
> - if test "x$ac_build_have_dlopen" = xyes -a "x$ac_build_have_dlfcn_h"
> = xyes; then
> - ac_build_have_dynamic_linking=yes
> - fi
> --if test "x$ac_build_have_dynamic_linking" !=
> "x$ac_have_dynamic_linking"; then
> --AC_MSG_ERROR([
> --=
> --
> --Build system and host system don't have the same dynamic linking state!
> --
> --=])
> --fi
> - AC_SUBST(BUILD_DL_LIBS)
> - eval "$saver"
> - ])
> Index: net/wing/Config.in
> ===
> --- net/wing/Config.in(revision 23569)
> +++ net/wing/Config.in(working copy)
> @@ -1,17 +0,0 @@
> -# Wing configuration
> -
> -choice
> -prompt "Version"
> -depends on PACKAGE_wing
> -default WING_STABLE
> -help
> -This option allows you to select the version of Wing to be built.
> -
> -config WING_STABLE
> -bool "Use the stable version (multi-interface)"
> -
> -config WING_TESTING
> -bool "Use the unstable version (multi-interfaces w/ ITACA)"
> -
> -endchoice
> -
> Index: net/wing/Makefile
> ===
> --- net/wing/Makefile(revision 23569)
> +++ net/wing/Makefile(working copy)
> @@ -8,17 +8,10 @@
> include $(TOPDIR)/rules.mk
>
> PKG_NAME:=wing
> +PKG_VERSION:=20101021
> +PKG_RELEASE:=1
> +PKG_REV:=402b10c1e914157ee1a105e11390ff013aa00748
>
> -ifneq ($(CONFIG_WING_STABLE),)
> -PKG_VERSION:=20100903
> -PKG_RELEASE:=4
> -PKG_REV:=520e3168a5af4d26bec3779cfa4901d128a4bc6f
> -else
> -PKG_VERSION:=$(shell date +%Y%m%d)
> -PKG_RELEASE:=1
> -PKG_REV:=HEAD
> -endif
> -
> PKG_SOURCE:=$(PKG_NAME)-$(PKG_VERSION).tar.bz2
> PKG_SOURCE_URL:=http://github.com/create-net/click-wing.git
> PKG_SOURCE_SUBDIR:=$(PKG_NAME)-$(PKG_VERSION)
> @@ -26,72 +19,58 @@
> PKG_SOURCE_PROTO:=git
>
> PKG_BUILD_DIR:=$(BUILD_DIR)/$(PKG_NAME)-$(PKG_VERSION)
> -PKG_BUILD_PARALLEL:=1
> -PKG_INSTALL:=1
>
> +PKG_BUILD_DEPENDS:=wing/host
> +
> include $(INCLUDE_DIR)/package.mk
> +include $(INCLUDE_DIR)/host-build.mk
>
> define Package/wing
> TITLE:=Wireless mesh networking toolkit
> SECTION:=net
> CATEGORY:=Network
> MAINTAINER:=Roberto Riggio
> -DEPENDS:=+kmod-tun
> +DEPENDS:=+kmod-tun +libpcap +libstdcpp
> URL:=http://www.wing-project.org/
> endef
>
> define Package/wing/Description
> Wing is a wireless mesh routing software. The routing protocol
> -is derived from Roofnet. It supports multiple radio interfaces and
> -link quality routing using the ETX, ETT, and WCETT metrics.
> +is derived from Roofnet. It supports multiple radio interfaces
> +and link quality routing using the ETX, ETT,
Re: [OpenWrt-Devel] [PATCH 2/3] ZNC: Package modules.
Yeah im looking into it, for some reason the patches didnt apply cleanly I manually edited the rejects, though now its not showing in the make menuconfig quite odd... so im working it out from a clean tree to be sure. On Mon, Oct 25, 2010 at 6:19 AM, Jonas Gorski < jonas.gorski+open...@gmail.com > wrote: > On 25 October 2010 04:04, Outback Dingo wrote: > > > > Committed revision 23627 - Reverts Commit 23626 - Needs fixing > > I'd gladly fix it, but I need something more specific than "needs > fixing" ;-). It builds and runs fine for me on ar71xx, current trunk. > > Regards, > Jonas > ___ > openwrt-devel mailing list > openwrt-devel@lists.openwrt.org > https://lists.openwrt.org/mailman/listinfo/openwrt-devel > ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] ttyUSB identification
On Mon, Oct 25, 2010 at 1:58 PM, Daniel Golle wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > hi everybody! > for my current project i need to reliably identify/distinguish several > USB-to-serial devices connected to a 4-port hub. in a desktop > environment, i would use hal over dbus (or in future probably device-kit > over dbus) to select the device by > vendor/product/serialno/connected-port. i was thinking of using udev to > create symbolic links, but even udev seems like taking a sledgehammer to > crack a nut to me... > my problem is that, depending on whether i cold-start or reboot the > system (a fon2202 btw), the different USB-to-serial gadgets end up to be > different ttyUSBx devices. especially the huawei 3g/umts wwan seems to > take a greatly varying amount of time before it becomes ready, so on a > reboot it becomes visible to the system as ttyUSB0 & ttyUSB1, in a > cold-start it's ttyUSB2 & ttyUSB3... > what way do you suggest to solve the problem? Address the devices by going into the /dev/serial/by-id : zoo...@buzek /dev/serial/by-id [18]$ l total 0 lrwxrwxrwx 1 root root 13 2010-10-25 09:17 usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if01 -> ../../ttyACM0 lrwxrwxrwx 1 root root 13 2010-10-25 09:17 usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if03 -> ../../ttyACM1 lrwxrwxrwx 1 root root 13 2010-10-25 09:17 usb-Dell_Dell_Wireless_5530_HSPA_Mobile_Broadband_Minicard_Device_3558620231349960-if09 -> ../../ttyACM2 The path there should be unique. Try to insert all the adaptors, and send me the output of "ls -l /dev/serial/by-id". -- Benjamin Henrion FFII Brussels - +32-484-566109 - +32-2-4148403 "In July 2005, after several failed attempts to legalise software patents in Europe, the patent establishment changed its strategy. Instead of explicitly seeking to sanction the patentability of software, they are now seeking to create a central European patent court, which would establish and enforce patentability rules in their favor, without any possibility of correction by competing courts or democratically elected legislators." ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Is the x86 VirtualBOX VDI target broken ATM? - [SOLVED]
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 24.10.2010 18:50, Alexey Loukianov wrote: > Today I've been trying to build OpenWRT x86 VDI target to do some experiments > with it and ended up with unbootable image. Build process completed fine, > VirtualBOX recognizes the VDI image, GRUB gets started as it should, loads up > the kernel and then I got the panic "Attempted to kill init!". Am I the only > one > who had ran into this problem? > Thanks to the Andrew Sapronov's offlist advice it had turned out the the hang was caused by the COM ports turned off in the VirtualBOX VM configuration. Simply turning the first serial port on and leaving it in "disabled" state in the VM configuration allowed my OpenWRT VirtualBOX VM to boot up successfully. Andrew, thanks a lot for helping! - -- Best regards, Alexey Loukianov mailto:mooro...@mail.ru System Engineer,Mob.:+7(926)218-1320 *nix Specialist -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJMxXYiAAoJEPB9BOdTkBULwKcH/0wz3ZhX7I1rIukPJlYY0RRc pwyhbhVvTnmxI2lOcQaXIgHlX9UhnClOyHw+PQ+4tgiwon/QQZeApe39usRFDseC +xZ/1ib+15sCAcfadzeu2U6pG8RCs4BIoac8m/oVKNKER+lpYjEmBYLtxvwwN9MW ELeJegKn16fzjxWJEJaoXUasDHue3EWQFaiLhO36GHav6mSN0HbT7NwOJeIPiDjc jIeKzAM0t1u8h4NmnFb4LnpYyzJ9dIKj2hEzaF19hM2OwYBN/bzHoES+Dgf9Ihb4 gqQwyS89OSVjRrSOK6xSvJQw4rct4TKCzCzRZ60IaM8pW+lRgfcYT6hVGKyTobk= =68eX -END PGP SIGNATURE- ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] ttyUSB identification
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 hi everybody! for my current project i need to reliably identify/distinguish several USB-to-serial devices connected to a 4-port hub. in a desktop environment, i would use hal over dbus (or in future probably device-kit over dbus) to select the device by vendor/product/serialno/connected-port. i was thinking of using udev to create symbolic links, but even udev seems like taking a sledgehammer to crack a nut to me... my problem is that, depending on whether i cold-start or reboot the system (a fon2202 btw), the different USB-to-serial gadgets end up to be different ttyUSBx devices. especially the huawei 3g/umts wwan seems to take a greatly varying amount of time before it becomes ready, so on a reboot it becomes visible to the system as ttyUSB0 & ttyUSB1, in a cold-start it's ttyUSB2 & ttyUSB3... what way do you suggest to solve the problem? cheers daniel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJMxXDJAAoJELd9RAts2fh5XNEIAKGiFesP67x4wBmOP6EoZKlU bQpj/Tge6i7KKizMaP38ux856nHTLdliRhrf9UxGX02YtlTbKJJz+yDxvnzJJ8Uq nQOd1z6oD56HfTgP5klGKR99N7jmkSEUull2D8MS8hykQIzftsVhcXjBe0Bk8kJe c88p5JuDanBYBu15DFBhF2bry8xza33Ig2oKMPKmQGcXizRYhHh3Pe8RIhpfLazs dRkMdBXFn1p+S/7NO47wFpwWiuG8zxMcyaidMloljRSM8PAxhhA/JnKAv2EKyBu8 eV3YmYxSxSFHY11A6MViOx03YUa0OKHsqgzxl6XG1eyhazw6ACZkcxXXElqGwVk= =f1du -END PGP SIGNATURE- ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] [PATCH 2/3] ZNC: Package modules.
On 25 October 2010 04:04, Outback Dingo wrote: > > Committed revision 23627 - Reverts Commit 23626 - Needs fixing I'd gladly fix it, but I need something more specific than "needs fixing" ;-). It builds and runs fine for me on ar71xx, current trunk. Regards, Jonas ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
