**** BEGIN LOGGING AT Thu Jun 20 02:59:59 2013 Jun 20 03:57:20 build #265 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/265 Jun 20 04:06:03 build #204 of octeon is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/204 Jun 20 10:17:14 hi Jun 20 10:17:53 is there a way to reflash an image when i flash this device and it doesnt come up? Jun 20 10:17:56 http://wiki.openwrt.org/toh/ubiquiti/unifi Jun 20 10:18:58 <[mbm]> wire up the serial port, tftp a new image from u-boot Jun 20 10:39:00 KanjiMonster: your patch to detect smp in the bcm6328 works (of course) http://pastebin.com/EMvt3f6G Jun 20 10:39:49 Pteridium: well, I tested it on a bcm63283, so it's somewhat expected ;) Jun 20 10:40:15 xD Jun 20 10:43:27 overclocking seems that work, but danitool and me are not sure if the cpu clock is changed really: http://pastebin.com/bvj0dTeJ Jun 20 10:43:54 it's only a pseudobenchmark Jun 20 10:45:33 and forcing smp in the 6328 the router freezes Jun 20 10:49:49 I think overclocking in bcm63xx isn't possible Jun 20 10:54:46 danitool: how can be explained the performance increase? Jun 20 10:55:45 I can manipulate a speedometer of a car changing its values, but it doesn't mean the car will run faster ;) Jun 20 10:57:15 good answer :( Jun 20 11:15:56 <_trine> tripolar, it should be able to accept a tftp Jun 20 11:16:37 <_trine> if its similar to other ubnt stuff Jun 20 11:17:29 <_trine> hold the reset button in while powering it up should put it into tftp mode Jun 20 11:18:12 <_trine> http://wiki.ubnt.com/Firmware_Recovery Jun 20 11:58:10 _trine thanks Jun 20 11:58:34 their windows config tool is crap - at least when you don't run windows at all and from a virtual machine it doesnt work :( Jun 20 12:03:32 <_trine> I have never used any of their tools linux and openwrt has always been enough Jun 20 12:04:46 <_trine> I have recovered some of my nanostation M's using TFTP when they have been non connective by any other means Jun 20 12:05:01 <_trine> seems easy enough Jun 20 12:17:49 _trine but i couldnt log in with ssh because their tool sets the pw Jun 20 12:17:59 its an Ubiquiti UniFi AP Jun 20 12:18:16 <_trine> why would you need to log in for tftp Jun 20 12:18:25 ahh okay Jun 20 12:19:15 <_trine> just press the reset button in and at the same time plug power into it Jun 20 12:19:43 <_trine> on my nano you see the leds flashing in a particular way Jun 20 12:20:00 <_trine> you can tell its waiting for the tftp Jun 20 12:20:22 <_trine> then just use linux to send the factory.bin Jun 20 12:21:22 <_trine> tftp; binary; connect 192.168.1.20; put factoryfirmware.bin Jun 20 12:23:07 here the device boots normal when i press the reset button while powering up Jun 20 12:26:13 <_trine> you're supposed to press the reset button in and keep it pressed in and then power it up still keeping the button pressed for about 8 seconds not longer or shorter Jun 20 12:26:25 <_trine> approximately Jun 20 12:45:32 _trine: the light is flashing fast seems it goes into tftp mode Jun 20 12:45:41 but it doesnt accept the package with Jun 20 12:45:41 atftp --trace --option "timeout 1" --option "mode octet" --put --local-file openwrt-ar71xx-generic-ubnt-unifi-squashfs-factory.bin 192.168.1.20 Jun 20 12:50:04 <_trine> I would just use the manual method and use one command at a time Jun 20 12:50:07 <_trine> tftp Jun 20 12:50:18 <_trine> binary Jun 20 12:50:32 <_trine> connect 192.168.1.20 Jun 20 12:50:32 <_trine> pu your file Jun 20 12:50:34 <_trine> put* Jun 20 12:51:28 <_trine> my nano's default ip address is 192.168.1.20 so its a guess if your is also but i would expect all ubnt stuff to be the same Jun 20 12:52:29 <_trine> my nano's default ip address is 192.168.1.20 so its a guess if yours is also but i would expect all ubnt stuff to be the same Jun 20 12:54:15 but it seems there is no device at 1.20 Jun 20 13:12:03 nbd r36977 trunk/package/mac80211/patches/300-pending_work.patch * ath9k: fix cabq depth accounting Jun 20 13:13:27 _trine: the ip 192.168.1.1 comes up but i is only pingable Jun 20 13:13:35 nmap shows 0 open ports Jun 20 13:14:00 <_trine> does it use airOS Jun 20 13:17:47 <_trine> maybe it is not the same as a nanostation which can use tftp from the boot loader Jun 20 13:18:22 <_trine> in which case you would need to connect the the serial port Jun 20 13:19:08 <_trine> but I would find it strange for ubnt to make a unifi ap diiferent for unbricking purposes than a nanostation Jun 20 13:19:36 <_trine> http://community.ubnt.com/t5/UniFi/HOWTO-Unbrick-your-UniFi-AP/td-p/338684 Jun 20 13:21:46 <_trine> https://github.com/openwisp/OpenWISP-Firmware/wiki/Unifi Jun 20 13:22:52 <_trine> http://www.youtube.com/watch?v=TkWX_fjLjY0 Jun 20 13:29:24 _trine: thanks Jun 20 14:34:01 _trine weired that they removed the tftp server Jun 20 14:41:53 <_trine> tripolar, ubnt seem to be weird period Jun 20 14:42:19 <_trine> and the owner is a loony Jun 20 14:46:39 I'd really like to find an alternate source that produces something in the form factor of the ubiquity picostation Jun 20 14:46:56 there's lots of various smoke detector stuff out there akin to the unifi Jun 20 14:47:23 but, I haven't found anything along the lines of the pico, nice clean small ap in an outdoor case with omni directional antenna Jun 20 14:56:15 <_trine> groz, and dont forget the PoE Jun 20 14:56:48 well trine, consider this with the unifi example Jun 20 14:56:59 passive poe limits it to 100mbit on the ethernet Jun 20 14:57:12 ummm, what's the point of the 300mbit radio, with 100mbit ethernet ? Jun 20 14:57:25 150 in the case of single antenna pico Jun 20 14:57:29 <_trine> I like the nanostation M2's but they could do with more flash and RAM Jun 20 14:57:43 300mbit/150mbit is just the PHY rate Jun 20 14:57:51 I realize that nbd Jun 20 14:57:52 it's not comparable to real throughput numbers Jun 20 14:57:59 also, it's half-duplex Jun 20 14:58:55 so in most cases 100 mbit/s ethernet is enough Jun 20 14:59:04 in most cases, it's more than ample Jun 20 14:59:13 only a few select tuned locations where it's the bottleneck Jun 20 14:59:44 bottleneck for real traffic or for artificial benchmarks? ;) Jun 20 15:00:13 hehe, in one case I'm considering, it's actually real traffic Jun 20 15:00:32 but, we are still debating just bit the bullet Jun 20 15:00:40 dig a trench, and put a fiber in the trench Jun 20 15:01:17 kinda hard to finalize details right now, still negotiating on land purchase Jun 20 15:01:29 we may yet end up in a different spot Jun 20 15:04:59 hauke r36978 trunk/package/mac80211/Makefile Jun 20 15:04:59 mac80211: add rt2x00-mmio package to store the common parts for SoC and PCI access Jun 20 17:21:00 build #268 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/268 Jun 20 17:24:14 build #268 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/268 Jun 20 18:49:07 florian r36979 trunk/target/linux/ brcm63xx/patches-3.9/426-hcs_mac_addr_pool.patch brcm63xx/patches-3.9/419-MIPS-BCM63XX-enable-enet-for-BCM6345.patch brcm63xx/patches-3.9/407-bcm63xx_enet-add-support-for-bcm6368-internal-ethern.patch brcm63xx/patches-3.9/800-wl_exports.patch * brcm63xx: reduce the number of Ethernet DMA channels from 16 to 8 Jun 20 18:57:02 build #259 of xburst is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/259 **** ENDING LOGGING AT Fri Jun 21 02:59:58 2013