**** BEGIN LOGGING AT Fri Feb 26 02:59:57 2010 Feb 26 03:16:52 netprince: ping Feb 26 03:17:39 netprince: does https://dev.openwrt.org/ticket/1618 work still? Feb 26 03:42:00 thepeople * r19865 /packages/net/updatedd/files/updatedd.hotplug: do not lock this to wan, allows it to be used on wan interfaces not named wan. Feb 26 03:44:50 netprince: one more https://dev.openwrt.org/ticket/2342 Feb 26 04:16:20 thepeople: wow, hmm, I might have to get back to you on that Feb 26 04:16:38 I still use them both, but the patches I use might be newer Feb 26 04:18:02 netprince__: ok, the patch team is in the process of cleaning trac so it is useable again without so much cruft in it, so I may have more for you :-) Feb 26 04:18:32 yeah, thats cool, I wont have to maintain so many of my own patches :) Feb 26 04:19:06 Can I get back to you about them tomorrow? Feb 26 04:20:02 netprince__: np, whenever Feb 26 04:54:12 thepeople * r19866 /packages/net/ez-ipupdate/files/ez-ipupdate.hotplug: [patchteam] ez-ipupdate: [PATCH] unnecessarily runs as a daemon (also allow multiple instances), closes #4874 Feb 26 05:01:21 thepeople * r19867 /packages/net/ez-ipupdate/files/ez-ipupdate.hotplug: add server option, closes #4816 Feb 26 05:16:58 thepeople * r19868 /packages/net/olsrd/files/olsrd.init: fixes OLSR 6and4 mode does not work if HNA entries are configured, closes #4803 Feb 26 10:31:07 cshore: you here? Feb 26 10:55:02 lars * r19869 /packages/admin/zabbix/Makefile: Feb 26 10:55:02 [packages] zabbix: Fix build. Feb 26 10:55:02 zabbix build expects $ARCH to be the target OS not the target architecture. Feb 26 11:01:30 kaloz * r19870 /trunk/toolchain/gcc/ (Config.in Makefile): [toolchain/gcc]: make TLS support configurable Feb 26 11:01:30 <_trine> {Nico}, good morning Feb 26 11:01:59 <_trine> can I make a suggestion re FS Feb 26 11:04:00 kaloz * r19871 /trunk/toolchain/uClibc/ (4 files in 3 dirs): [toolchain/uclibc]: upgrade and fixup support for the uclibc nptl branch Feb 26 13:01:37 cshore: ping Feb 26 14:31:01 cshore: you here? Feb 26 15:00:32 ping thepeople: https://dev.openwrt.org/ticket/2342 should be good still Feb 26 15:01:10 the last patch is the one I use, its just a one liner Feb 26 15:01:27 well, one liner for ip-down and one for ip-up Feb 26 15:03:21 is this for .up=[01] ? Feb 26 15:05:12 I'm trying to remember exactly how its used, its been a long time Feb 26 15:08:14 the uci_set_state cant take sections with periods, so having PPP_IPPARAM as an IP address was invalid Feb 26 15:09:02 I see Feb 26 15:16:07 thepeople: the bursttime patch should still be good, but the connbytes patch was already committed IIRC: https://dev.openwrt.org/ticket/1618 Feb 26 15:16:33 thepeople: I have several other useful patches for qos-scripts if you are interested Feb 26 15:19:47 thepeople: hmm, looks like the connbytes patch was only partially applied... Feb 26 15:27:54 thepeople: I believe the second part of the connbytes patch was not necessary, so dont worry about that patch Feb 26 16:38:50 thepeople * r19872 /trunk/target/linux/avr32/config-default: add missing symbol Feb 26 16:57:08 cshore ping Feb 26 17:08:06 netprince: https://dev.openwrt.org/ticket/1893 Feb 26 17:10:52 I still use that one too, its still good Feb 26 17:16:23 thepeople * r19873 /trunk/package/ppp/files/etc/ppp/ (ip-down ip-up): [patchteam] fix issues with pptpd server and firewall, closes #2342 Feb 26 17:22:55 thepeople * r19874 /trunk/package/qos-scripts/files/usr/lib/qos/ (generate.sh tcrules.awk): This patch allows the user to specify esfq as the leaf qdisc, as well as perturb and hash parameters, closes #1893 Feb 26 17:23:23 ^^ its really windy out Feb 26 17:23:37 O.O Feb 26 17:30:34 cshore i've the /overlay directory now. what's next? Feb 26 18:09:17 lars * r19875 /trunk/target/linux/xburst/ (2 files in 2 dirs): Feb 26 18:09:18 [xburst] jz4740_nand: Feb 26 18:09:18 * Use NAND_ECC_HW_OOB_FIRST mode instead of a patched NAND_ECC_HW Feb 26 18:09:18 * Fix subpage writing Feb 26 18:09:18 * speed up reading Feb 26 18:18:44 cshore ping Feb 26 18:35:26 cshore seems i can't get block-extroot to work Feb 26 19:09:33 build #17 of ixp4xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/17 Feb 26 20:02:00 cshore: you forgot to tag kmod-nls-base as "maybe root" Feb 26 20:02:08 cshore: http://openwrt.pastebin.com/uK5zNAJv Feb 26 20:02:45 cshore: without that loaded in advance, usbcore won't load due to a missing symbol Feb 26 20:12:17 build #17 of orion is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/orion/builds/17 Feb 26 20:13:13 obsy * r19876 /packages/net/proftpd/Makefile: [packages] proftpd: update to 1.3.2e Feb 26 20:36:34 cshore: you might also want to figure out a way to change the overlay_root option in opkg.conf Feb 26 21:57:28 pong dogge10 and xMff Feb 26 21:57:52 cshore got block-extroot working :) Feb 26 21:58:04 dogge10: excellent Feb 26 21:58:09 build #17 of ar7 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/17 Feb 26 21:59:09 cshore had to change http://openwrt.pastebin.com/uK5zNAJv and the overlay_root option in /etc/opkg.conf Feb 26 22:00:44 hmmm....I don't know why it worked for me (without the nls), but the opkg thing I didn't know about Feb 26 22:01:18 in /etc/opkg.conf i changed: Feb 26 22:01:25 -option overlay_root /jffs Feb 26 22:01:28 +option overlay_root /overlay Feb 26 22:02:00 maybe you can do this via a script somehow Feb 26 22:02:12 yeah, probably a good idea Feb 26 22:03:20 otherwise it block-extroot works fine Feb 26 22:03:36 maybe we should just get rid of /jffs at all, even for non block-extroot Feb 26 22:03:36 with a usb pen drive Feb 26 22:20:20 build #16 of xburst is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/xburst/builds/16 Feb 26 22:20:46 cshore: i just reviewed your recent patch set Feb 26 22:20:50 i think it can go in the way it is Feb 26 22:21:17 there's some leftover crap from the package split, but i'll remove that myself before committing Feb 26 22:24:31 nbd thnx for that Feb 26 22:24:47 nbd maybe you can add a script that changes the overlay_root option in opkg.conf and don't forget the patch from xMff Feb 26 22:25:20 which patch from xMff? Feb 26 22:26:30 nbd http://openwrt.pastebin.com/uK5zNAJv Feb 26 22:27:14 ah Feb 26 22:27:34 cshore: why is block-extroot submitted for /packages while block-mount is in trunk? Feb 26 22:27:46 imho it might make sense to keep both in trunk Feb 26 22:34:12 nbd: back Feb 26 22:35:13 nbd: cool....I did the one in trunk because I didn't know if it'd be considered a 'base' thing....and I have access to 'packages' Feb 26 22:36:26 well, i think it makes sense as base functionality, especially for nas and similar devices Feb 26 22:38:14 nbd: ok Feb 26 22:42:44 nbd: do you want me to update the patch with the nls and doing something about overlay_root, and making it all base? Feb 26 22:43:51 no Feb 26 22:43:56 i've already taken care of that Feb 26 22:44:00 heh, ok Feb 26 22:45:04 I'm puzzled by the nls thing because it worked for me without it Feb 26 22:45:44 maybe it's only required if selected as module Feb 26 22:45:45 nbd * r19877 /trunk/package/ (17 files in 7 dirs): add the block-mount package by Daniel Dickinson (cshore), replacing existing automount functionality Feb 26 22:45:56 nbd * r19878 /trunk/include/kernel.mk: add support for marking specific kernel module packages as potentially being required for mounting the rootfs (patch by cshore) Feb 26 22:46:04 nbd * r19879 /trunk/package/kernel/modules/ (block.mk fs.mk other.mk usb.mk): mark relevant kmod-* packages as being required for rootfs mounting (patch by cshore) Feb 26 22:46:12 nbd * r19880 /trunk/package/ (8 files in 3 dirs): add block-extroot, a package for using an external filesystem as rootfs (patch by cshore) Feb 26 22:46:40 :) Feb 26 22:47:09 nbd: thank you! Feb 26 22:47:15 no problem Feb 26 22:47:23 :) Feb 26 22:48:23 cshore now a app for luci to configure block-extroot :) Feb 26 22:48:48 dogge10: well first the little script I did up at nbd's request Feb 26 22:49:10 nbd: did you get a chance to look a it, btw? Feb 26 22:50:01 not yet Feb 26 22:50:30 nbd: no problem...let me know when you do Feb 26 22:50:33 sure Feb 26 22:51:36 btw. one thing that would also be useful is making sure that block-extroot plays well with sysupgrade Feb 26 22:51:54 i.e. it should grab the config data from jffs2 instead of the extroot overlay Feb 26 22:52:15 nbd: I agree, I had though about that Feb 26 22:52:28 nbd: just not done it yet Feb 26 22:52:40 no problem Feb 26 22:52:47 just a minor wishlist item for making it more practical Feb 26 22:53:01 cshore: interesting work, is there a bit of documentation I can read about how it works? Feb 26 22:53:14 netprince: not yet Feb 26 22:55:45 netprince: but I'll work on that now Feb 26 23:05:22 nbd what about that overlay_root option? Feb 26 23:05:30 in /etc/opkg.conf Feb 26 23:05:54 for that we should move /jffs to /overlay entirely Feb 26 23:06:03 and change opkg.conf at the same time Feb 26 23:06:19 so it's the same overlay dir for extroot vs. no extroot Feb 26 23:06:23 ok Feb 26 23:12:19 nbd: I'll need to make a change in extroot if we do that...right now jffs is mounted /jffs before we do extroot Feb 26 23:12:30 nbd: and unmounted if the extroot is mounted Feb 26 23:12:59 shouldn't be hard to change, right? Feb 26 23:13:11 either way, i need some sleep now, need to get up early tomorrow Feb 26 23:13:18 nbd: no, it should just be fairly straitforward Feb 26 23:13:24 nbd: ok Feb 26 23:30:53 dogge10: oh, did label work? Feb 26 23:32:43 yes, label works. not tested with uuid Feb 26 23:33:09 dogge10: I'm sure that works. I was more concerned about uuid Feb 26 23:33:09 cshore: cool Feb 26 23:33:29 netprince: I'm updating the uci:fstab documentation right now Feb 26 23:37:27 cshore: looks good so far Feb 26 23:39:02 is a crc32 checksum in any way dependent on the byte order? Feb 26 23:42:55 netprince: can you look at the uci:fstab again? I've added the uuid/label documentation Feb 26 23:43:20 rtz2: well the order of the bytes sent to the crc32 matter Feb 26 23:44:01 and the crc32 wlil be shorted in the system's byte order Feb 26 23:44:15 hmmm Feb 26 23:44:16 but if you send the same stream of bytes, you'll get the same number Feb 26 23:44:17 gah Feb 26 23:44:25 i'm to tired for this now Feb 26 23:44:45 rtz2: I can understand Feb 26 23:44:57 endianness warps the brain Feb 26 23:45:10 cshore: I had some comments for your config creation script, but I have to dig them out again Feb 26 23:45:18 rtz2: ok Feb 26 23:45:19 so this will also have to wait till tomorrow Feb 26 23:45:25 rtz2: np Feb 27 00:05:04 cshore: uuid stuff makes sense to me Feb 27 00:09:54 netprince: ok, thanks Feb 27 00:14:19 cshore: perhaps a little more detail on the rootfs feature? for example, does the rootfs get mounted on top of the jffs2 on top of squashfs? does the overlay cover up the /etc/config/fstab from the jffs partition? If so how do you disable the overlay? Feb 27 00:14:44 netprince: I'm doing that as a separate page....I don't think it belongs on fstab Feb 27 00:14:56 cshore: ahh, good idea Feb 27 00:15:53 it looks good to me so far, we have needed this functionality updated for some time Feb 27 00:16:40 netprince: I know, that's partly why I did it, and partly because I need for the project I'm working on Feb 27 00:17:08 netprince: I got to do this as part of my job :) Feb 27 00:17:58 cshore: nice, you building a nas or something? Feb 27 00:18:09 a VoIP PBX Feb 27 00:19:04 our device only has 4MB flash, but it has 128MB usb built in Feb 27 00:19:08 cshore: nice, I've always wanted to try a voip pbx like asterisk, just never had a good reason Feb 27 00:19:36 cshore: if you can get dry DSL, the price factor compared to voice lines is amazing Feb 27 00:19:59 I'll be saving $50-70 per month when I switch Feb 27 00:20:17 what service do you use for the voice uplink? Feb 27 00:20:28 netprince: I'm looking at Unlimitel Feb 27 00:20:41 netprince: haven't done it yet...still working on the PBX Feb 27 00:21:00 netprince: so I'll be using what I produce Feb 27 00:21:19 cool, I'd like to hear how that goes Feb 27 00:21:56 netprince: sure, first I have to get Asterisk and/or FreeSwitch compiling (preferrably both for comparison purposes) Feb 27 00:23:07 netprince: the PBX will include the voicemail as well as control of the extensions Feb 27 00:23:08 You might need to look into qos for that too I suppose? Feb 27 00:23:17 netprince: that is in the plan, yes Feb 27 00:24:24 cshore: sounds like fun, perhaps after you iron out all the bugs, I'll give it a try :) Feb 27 00:24:33 netprince: sounds good Feb 27 00:25:04 cshore: I've done a little with the qos-scripts, if I can be of assistance just let me know Feb 27 00:25:16 netprince: let's just say when I found out about dry DSL I become *really* excited about getting the project into production Feb 27 00:25:25 netprince: ok, thanks Feb 27 00:25:46 netprince: what is your email? Feb 27 00:26:02 netprince (at) vt (dot) edu Feb 27 01:31:46 build #23 of brcm63xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/23 Feb 27 01:46:46 build #17 of au1000 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/17 Feb 27 02:17:15 ping netprince **** ENDING LOGGING AT Sat Feb 27 02:59:56 2010