**** BEGIN LOGGING AT Fri Oct 10 03:00:06 2008
Oct 10 04:27:40 <Bartman007> russell_: tested with the CONFIG_HZ tweak I assume?
Oct 10 04:44:08 <russell_> Bartman007: i used the newer version as linked (the .tar containing a new Makefile and patch).
Oct 10 04:44:50 <Bartman007> russell_: without the kernel patch? or have you only compile tested it so far?
Oct 10 04:45:42 <russell_> i just compile tested it.
Oct 10 04:45:51 <russell_> so it might be broken, usage wise
Oct 10 04:46:24 <russell_> but at least it compiles to the point someone can test it now ;-)
Oct 10 04:46:33 <Bartman007> do you have any zaptel hardware?
Oct 10 04:46:36 <russell_> no
Oct 10 04:48:12 <russell_> it just gives me an irrationally warm feeling to have it compile without vomitting on my shoe, otherwise i have no stake in the problem.
Oct 10 04:48:47 <Bartman007> nevermind then. I've been reading up on it, and zaptel needs a 1000hz timer of some form, be it rtc, uhci controller, digium pci card or 1000hz kernel timer, so basically it looks like zaptel won't work it it's current form on the majority of platforms
Oct 10 04:51:21 <russell_> i like it when i can start a "select all packages" build and not find it has errored out part way through, so either getting it to build or marking it broken or something would be nice
Oct 10 04:52:53 <russell_> this is one of 4 or 5 things that "make world" died on as of r12893
Oct 10 04:53:05 <russell_> (for wgt634u and atheros)
Oct 10 04:56:39 <Bartman007> russell_: make IGNORE_ERRORS=m  <--- will ignore errors in any packages marked 'm' in menuconfig
Oct 10 04:57:58 * russell_ makes a note of that
Oct 10 05:07:44 <Bartman007> the kernel config won't be committed, so the rest of the fix/update will have to stew while we decide what to do
Oct 10 10:59:42 <CIA-12> hcg * r12941 /trunk/ (Config.in include/kernel-defaults.mk): Adds support for using a local clone of a git tree
Oct 10 11:25:51 <blogic> Weedy: ?
Oct 10 11:25:55 <blogic> there ?
Oct 10 11:26:08 <blogic> can you test a patch for me ? to see if it solves your firewall issues ?
Oct 10 11:26:49 <blogic> Weedy: http://pastebin.ca/1224467
Oct 10 12:00:48 <larsc> ich kann die zur not mitnehmen
Oct 10 12:01:01 <larsc> wrong window, wrong channel
Oct 10 14:24:34 <Weedy> blogic: im starting a clean build of 12942 with your patch now
Oct 10 14:58:30 <blogic> Weedy: ok
Oct 10 17:23:45 <Weedy> ]/win 30
Oct 10 19:06:53 <aroscha> hi
Oct 10 19:07:00 <wrt> hi
Oct 10 19:07:22 <aroscha> blogic, [florian: are you going to be at the CCC?
Oct 10 20:20:37 <johnrw> does anyone know is any of the packages include the python-sqlite3 bindings?
Oct 10 20:26:37 <larsc> johnrw: the python-sqlite package contains the bindings for sqlite3
Oct 10 20:27:05 <johnrw> larsc: Ok... great!!!!!!
Oct 10 20:27:50 <johnrw> I wasn't sure... as it's a separate package in debian...
Oct 10 20:29:41 <larsc> johnrw: ah, oh, hm
Oct 10 20:30:01 <larsc> johnrw: i think i was wrong
Oct 10 20:30:28 <larsc> johnrw: it requires sqlite3 but builds the sqlite2 bindings
Oct 10 20:33:27 <johnrw> so now I am more confused...
Oct 10 20:33:58 <johnrw> does that mean it needs the makefile updated?
Oct 10 20:34:32 <larsc> johnrw: i went to pysqlite website. pysqlite2 are the bindings for sqlite3
Oct 10 20:34:36 <larsc> quite confusing
Oct 10 20:37:07 <johnrw> well I guess that means I need to get a makefile in for pysqlite2 then.
Oct 10 20:39:40 <larsc> its there
Oct 10 20:40:22 <larsc> python-sqlite builds bindings for pysqlite3. sorry if i confused you
Oct 10 20:41:15 <johnrw> ok... so I can go ahead and do a build for it then.
Oct 10 20:47:25 <johnrw> looking at the snapshots... I don't see a python-sqlite package... is that because they aren't built by default... but I can enable it?
Oct 10 20:49:17 <johnrw> ok, nevermind... I will just get a svn pull ready.
Oct 10 22:04:03 <noz> MateIn4: ping
Oct 10 22:05:49 <MateIn4> noz: pong
Oct 10 22:05:56 <candrews> I'm hoping to get some help with /etc/config/radvd - I can't find any docs, and what I think makes sense isn't working. I also am finding it really hard to debug the script.
Oct 10 22:06:31 <noz> MateIn4: I have a new version of jungo-image.py for you. It reads /proc/mtd to work out how many and how big the partitions are
Oct 10 22:07:01 <noz> Only tested on 1 partition so far (extent of available test platforms)
Oct 10 22:07:24 <MateIn4> noz: I also made the same change. I sent you a message yesterday but maybe you didn't get it.
Oct 10 22:07:41 <noz> Oh, sorry.
Oct 10 22:07:54 <MateIn4> This is why I think this script should be in the openwrt repository.
Oct 10 22:08:32 <noz> Let's compare changes ;)
Oct 10 22:08:39 <MateIn4> BTW, have you figured out the GPIO assignments for your device?
Oct 10 22:09:12 <MateIn4> Poking around /proc is always useful (e.g. /proc/interrupts)
Oct 10 22:09:21 <xMff> candrews: well if it's similar to the olsrd init in it's structure then I assume that radvd's config options are directly mapped to uci
Oct 10 22:09:25 <noz> Not yet - not planning to do that yet, unless I need it for redboot - 1 thing at a time
Oct 10 22:10:11 <MateIn4> You may need it for redboot; hard to say.
Oct 10 22:11:02 <MateIn4> BTW, I modified my redboot so that I can bypass the redboot boot script using the "reset" button. Very useful if not using UART.
Oct 10 22:14:29 <noz> but I have a UART, so I'm not that bothered, thanks anyway
Oct 10 22:16:33 <MateIn4> I based my redboot on the intel sources because they support the npe which allows telnet access to redboot.
Oct 10 22:17:41 <noz> I didn't realise there was more than one source.
Oct 10 22:19:11 <MateIn4> Unfortunately, the ecos source repository doesn't include support for the npe. Take a look here: http://www.intel.com/design/network/products/npfamily/download_ixp400.htm it was done by redhat.
Oct 10 22:21:00 <noz> Presumably the pre-built binary won't work as-is?
Oct 10 22:22:35 <MateIn4> you have to check the phy assignments and see if they match.
Oct 10 22:23:44 <MateIn4> but from your notes I don't think they do. also you need to see if the switch needs any kind of initialization. some do some don't. in my case, the switch requres init before use.
Oct 10 23:05:59 <Philip5> thepeople: btw, if you remember my problem yesterday with bridging or firewallrules... i solved it in a rather drastic way by rebuilding everything from scratch... something didn't match i guess :)
Oct 10 23:11:37 <thepeople> Philip5: glad to see it is working now
Oct 10 23:12:19 <Philip5> yepp, i guess something didn't like to be updated just by being built in the old stagingdir even though it was updated
Oct 10 23:12:49 <thepeople> I have seen that in hte past
Oct 10 23:13:47 <Philip5> thepeople: an other question... when i install webif with opkg the install script returns a error message like this: "Collected errors: * Could not obtain administrative lock"
Oct 10 23:13:52 <Philip5> what does that mean?
Oct 10 23:14:06 <Philip5> it looks like it doesn't matter anyway but i'm curious
Oct 10 23:14:20 <thepeople> Philip5: what does your opkg.conf look like?
Oct 10 23:15:14 <Philip5> like this: http://pastebin.com/m6d5503c3
Oct 10 23:15:44 <Philip5> but i have installed it from a local repo that is my own build from latest svn
Oct 10 23:16:37 <thepeople> hmm, that is correct
Oct 10 23:16:39 <Philip5> so i don't use that repo and edit it in webif to add my local repo
Oct 10 23:16:58 <thepeople> did it get installed?
Oct 10 23:17:11 <Philip5> i install haskel and webif with a direct link to my local repo
Oct 10 23:17:14 <Philip5> yes
Oct 10 23:17:32 <thepeople> I won't worry about it then
Oct 10 23:17:33 <Philip5> everything seam to work but i'm just curious about that error message
Oct 10 23:17:46 <thepeople> not sure what would cause it off hand
Oct 10 23:17:58 <Philip5> ok, but what is the administrative lock?
Oct 10 23:18:10 <Philip5> something about being root?
Oct 10 23:19:22 <thepeople> nope, something that it is having a issue with its lock file
Oct 10 23:21:14 <Philip5> it all looks like this and is the first thing i have done after a brand new flashed firmware from rescue mode: http://pastebin.com/m5610543c
Oct 10 23:21:40 <Philip5> but it works so i might give it a rest :)
Oct 11 00:03:10 <scanlime> johnrw: hiya. This is Micah.
Oct 11 00:50:02 <johnrw> scanlime: Hi guy
Oct 11 00:50:33 <johnrw> I was just trying to remove a 2 port usb connector...
Oct 11 00:50:49 <johnrw> didn't see you were here...
Oct 11 01:35:08 <CIA-12> nbd * r12948 /trunk/package/mac80211/patches/ (22 files): mac80211: add rate control rewrite and enhance the performance of the minstrel algorithm for non-mrr configurations
Oct 11 01:35:26 <CIA-12> nbd * r12949 /trunk/target/linux/generic-2.6/config-2.6.27: enable debugfs by default
Oct 11 01:58:24 <CIA-12> nbd * r12950 /trunk/package/madwifi/patches/372-queue_vif.patch: madwifi: fix a node refcounting imbalance with a low (but nonzero) crash probability
Oct 11 02:04:55 <CIA-12> nbd * r12951 /branches/8.09/package/madwifi/patches/ (4 files): merge madwifi changes to 8.09
Oct 11 02:18:38 <sn9> nbd: are you awake?
**** ENDING LOGGING AT Sat Oct 11 02:59:57 2008