| uvos | Wizzup: D3 is Solana btw if you want to check moto sources for ram map | 00:09 |
|---|---|---|
| Wizzup | uvos: Solana? | 00:10 |
| Wizzup | uvos: we might need modified kexecboot and such as well, no? | 00:10 |
| uvos | d3's code name | 00:10 |
| Wizzup | right | 00:10 |
| Wizzup | first thing I did was externally charge the battery so that it boots | 00:11 |
| uvos | in source d4 = masarati d3 = solana | 00:11 |
| uvos | Wizzup: sure | 00:11 |
| Wizzup | it'd be fun to make it all work, but I'd need hand holding | 00:11 |
| uvos | id start with safestrap | 00:11 |
| uvos | and get something to boot there | 00:11 |
| Wizzup | it's booted to android atm | 00:11 |
| Wizzup | ok | 00:11 |
| uvos | id guess that d3 has the cpcap setup of bionic | 00:12 |
| uvos | just becuase its closer to that than d4 dates wise | 00:12 |
| Wizzup | ok | 00:13 |
| Wizzup | uvos: so I'd have to google around and find stuff like this? https://sites.google.com/site/sdshadowscollection/droid-3-xt862 | 00:15 |
| uvos | at first yeah | 00:16 |
| uvos | unless the d3 has the utags tag like d4 | 00:16 |
| uvos | you have to get mainline to work with safestrap | 00:16 |
| Wizzup | btw this is probably the US variant | 00:16 |
| uvos | first | 00:16 |
| Wizzup | XT862 | 00:16 |
| uvos | find a utags.bin (in a moto update) and link it somehwere | 00:17 |
| Wizzup | what do you mean, moto update? | 00:17 |
| uvos | a update .zip | 00:17 |
| uvos | for the stock fw | 00:17 |
| Wizzup | this also seems helpful: https://forum.xda-developers.com/t/index-motorola-droid-3-2011.3114734/ | 00:18 |
| Wizzup | ok | 00:18 |
| Wizzup | seems to contain links to tools and fw | 00:18 |
| uvos | https://firmware.center/firmware/ has solana fws | 00:21 |
| lel | MerlijnWajer opened an issue: https://github.com/maemo-leste/bugtracker/issues/544 (Potentially look into Motorola Droid 3 support) | 00:24 |
| Wizzup | uvos: I remember now also that someone bought a phone like the droid 4 and it accidentily booted with our d4 image | 00:25 |
| Wizzup | not sure if that was bionic or something else | 00:25 |
| Wizzup | iirc it was something else | 00:25 |
| uvos | Wizzup: that was a bionic | 00:25 |
| uvos | thats how we came to support this device | 00:25 |
| Wizzup | ok | 00:26 |
| Wizzup | :p | 00:26 |
| uvos | lucky for him and us that was before tmlind implmented setting up the cpcap regs in the mainline kernel so bionic just worked with d4 dts at that time on the bootloader setup. | 00:28 |
| Wizzup | right | 00:29 |
| Wizzup | As you saw, I made a ticket for supporting it | 00:29 |
| uvos | no utags :( | 00:33 |
| Wizzup | mhm | 00:33 |
| uvos | solana cdt.bin shows that mbm dosent use utags on this device | 00:33 |
| Wizzup | ok | 00:36 |
| * Wizzup zzz | 01:01 | |
| freemangordon | Wizzup: https://pastebin.com/ZC8Y4w37 | 15:36 |
| Wizzup | freemangordon: is this a change in the new wpasupplicant? | 15:39 |
| Wizzup | dpkg -l | grep supplicant | 15:40 |
| freemangordon | yes | 15:40 |
| Wizzup | ok, I only installed the new client part of it | 15:40 |
| freemangordon | 2:2.9.0-21+2m7.1 | 15:40 |
| Wizzup | I can look at this after I am done biking | 15:40 |
| freemangordon | sure, no hurry | 15:40 |
| Wizzup | wait, Scan also is not present | 15:40 |
| freemangordon | just wanted to let you know | 15:40 |
| freemangordon | mhm | 15:40 |
| Wizzup | I think something else is up | 15:40 |
| freemangordon | no idea | 15:40 |
| Wizzup | is your interface visible? | 15:40 |
| freemangordon | it should be | 15:41 |
| freemangordon | lemme check | 15:41 |
| Wizzup | ok, well, I will try with new wpasupplicant after this, and if it works for me, we can figure out what's up | 15:41 |
| Wizzup | semi afk for the next 2 hours first though | 15:41 |
| freemangordon | take your time, I'll not be available as well in the next few hours | 15:42 |
| freemangordon | maybe USB is misbehaving | 15:43 |
| Wizzup | freemangordon: maybe check if interface is enabled and visible | 15:47 |
| freemangordon1 | hmm, this wifi adapter took my PC down :) | 15:48 |
| freemangordon | Wizzup: yep, wlan0 is there | 15:52 |
| uvos | tmlind: so VRZ_9.8.2OT_127_MZ609_CFC_1FF.xml contains mbmloader.bin | 18:48 |
| uvos | tmlind: but no allow-mbmloader-flashing-mbm.bin | 18:48 |
| uvos | so how is moto performing the update? | 18:48 |
| uvos | is mbm.bin maybe just more permissive on xyboard? | 18:49 |
| Wizzup | freemangordon: ok, will take a look momentarily | 18:56 |
| Wizzup | uvos: maybe we can diff the d4 .bin files and see what allow changes? | 18:56 |
| Wizzup | uvos: that's also possible @ permissive | 18:59 |
| uvos | im not sure what diffing the files would accive | 19:00 |
| uvos | if you have the device you can just test and see what it allows | 19:00 |
| uvos | you cant change it anyhow as its singed | 19:00 |
| uvos | it might also be that mbm.bin in VRZ_9.8.2OT_127_MZ609_CFC_1FF.xml specificly is more permissive | 19:00 |
| uvos | i also found allow-mbmloader-flashing-mbm.bin ostensibly for MZ608 here: https://forum.xda-developers.com/t/guide-xoom2-xyboard-how-to-update-xyboard-wifi-mz607-to-jelly-bean-4-1-2.2591388/page-6 | 19:02 |
| uvos | ie the european variant of xyboard | 19:02 |
| uvos | tmlind ^^^ | 19:03 |
| tmlind | uvos: ok probably the mz608 binary won't work on anything else though | 19:40 |
| Wizzup | uvos: how can I test? | 19:40 |
| tmlind | Wizzup: hold vol down + power until you get to flash mode | 19:41 |
| tmlind | then from pc with usb cable connected, fastboot flash mbm allow-mbmloader-flashing-mbm.bin | 19:41 |
| tmlind | battery must be well chargerd, it will probably produce an error and then you have to reflash the original mbm.bin | 19:42 |
| tmlind | with fastboot flash mbm mbm.bin | 19:42 |
| Wizzup | tmlind: but there is no allow-mbmloader-flashing-mbm.bin - right? | 19:43 |
| tmlind | Wizzup: oh i though you wanted to try out the mz608 file on mz609 | 19:45 |
| Wizzup | well, uvos said that maybe the default is already permissive | 19:45 |
| tmlind | at least on mz617 you can only flash to cache partition pretty much, not sure about the user partition | 19:46 |
| tmlind | i just did mkfs.ext4 on the preinstall partition and untarred my rootfs there, seems to work just fine | 19:47 |
| tmlind | while trying to flash preinstall with an unsigned image produces an error and then you have reflash the original preinstall | 19:47 |
| Wizzup | freemangordon: ok yeah so this is something that changed in their dbus interface and it means wifi is broken in -devel | 20:00 |
| tmlind | uvos: looks like mmcblk1p15 cid partition on mz617 at least has a signature, maybe earlier models did not if that can be swapped like the link above says | 20:03 |
| uvos | tmlind: well since VRZ_9.8.2OT_127_MZ609_CFC_1FF.xml contains a mbmloader.bin update and the purpose of allow-mbmloader-flashing-mbm.bin is to allow flashing mbmloader it would stand to reason that at least mbmloader partition is unlock in mbm.bin from VRZ_9.8.2OT_127_MZ609_CFC_1FF.xml | 20:53 |
| uvos | so maybe Wizzup should try this file | 20:53 |
| uvos | i know for a fact that on bionic motorola dosent blow any fuses to prevent you from downgrading | 20:53 |
| uvos | so you can flash any old mbm.bin | 20:53 |
| uvos | then if that dosent work maybe just maybe the mz608 file will work | 20:54 |
| uvos | obv. all this is a bit risky | 20:56 |
| uvos | but so is dding the rootfs and utags around | 20:56 |
| tmlind | uvos: worth a try, cheap toys to play with at least :) | 21:06 |
| tmlind | uvos: but hmm so if mbm.bin allows reflashing mbmloader, it still does not allow flashing other partitions with random images | 21:08 |
| uvos | tmlind: no but maybe mbm.bin from that file is in fact configured like allow-mbmloader-flashing-mbm.bin | 21:09 |
| uvos | ie also more permissive in other ways | 21:09 |
| tmlind | uvos: maybe, but i think on xt894 the difference was that flashing kexecboot to bpsw is allowed with allow-mbmloader-flashing-mbm.bin | 21:10 |
| uvos | there is no real reason why allow-mbmloader-flashing-mbm.bin is more permissve that mbm.bin in flashing bspw | 21:10 |
| uvos | right | 21:10 |
| tmlind | hmm or maybe it was needed to reflash utags actually | 21:11 |
| uvos | both i think | 21:11 |
| uvos | iirc | 21:11 |
| tmlind | on mz617, i noticed i had to use fastboot flash:raw boot boot.img to reflash the signed boot.img | 21:12 |
| tmlind | but i don't think flash:raw allows reflashing random images again | 21:13 |
| freemangordon | Wizzup: I think that's another reason to proxy it - if we use it in more than one place, we shall change all of the users next time they break the interface | 21:14 |
| Wizzup | freemangordon: this was more a code deficiency in my code | 21:21 |
| Wizzup | freemangordon: I just assumed the interface path would not change, but they made it zero indexed | 21:21 |
| Wizzup | freemangordon: so we make a proxy that does the same for multiple interfaces? wouldn't solve the problem per se | 21:21 |
| freemangordon | this is interface change by all means | 21:21 |
| uvos | i dont think its senseble to proxy deamon everything for this reason | 21:21 |
| Wizzup | freemangordon: | 21:22 |
| Wizzup | -#define TEST_INTERFACE_PATH WPA_DBUS_INTERFACES_OPATH "/1" | 21:22 |
| Wizzup | +#define TEST_INTERFACE_PATH WPA_DBUS_INTERFACES_OPATH "/0" | 21:22 |
| freemangordon | that's why I said "another reason" :) | 21:22 |
| freemangordon | Wizzup: yeah, got it | 21:22 |
| Wizzup | freemangordon: that's the only change required to make it work for now | 21:22 |
| freemangordon | but, I have to change connui as well | 21:22 |
| Wizzup | shared lib maybe makes sense, proxy not so much imho | 21:22 |
| freemangordon | ok | 21:23 |
| Wizzup | Of course I can find the first interface and just use it, but as it stands, icd2 plugin cannot deal with more than one wifi interface anyway | 21:23 |
| uvos | we should do that... | 21:23 |
| uvos | and switch to presistant names | 21:23 |
| freemangordon | BTW, that's something we shall fix | 21:23 |
| uvos | instead of wlan0 wlan1 etc | 21:23 |
| freemangordon | I think we can use service_id, but not sure | 21:24 |
| uvos | since using only one interface, but allways the same one is better than using a random one depeding on what driver gets loaded first... | 21:24 |
| Wizzup | pull request welcome for that | 21:25 |
| Wizzup | Looks like I can use this: mdbus2 -s fi.w1.wpa_supplicant1 /fi/w1/wpa_supplicant1 fi.w1.wpa_supplicant1.GetInterface "wlan0" | 21:27 |
| Wizzup | all of this will get much more complicated now, blergh | 21:45 |
| Wizzup | freemangordon: parazyd: I'm building a new libicd-network-wpasupplicant which deals with the wpa supplicant interface name change, but as the current fix is a hack, it is really important that they stay in -devel for now, and independently *cannot* be moved to non-devel | 21:50 |
| parazyd | okay | 21:52 |
| parazyd | New wpasupplicant is also devel only | 21:52 |
| Wizzup | parazyd: yes, but this is a heads up(tm) | 21:53 |
| parazyd | :) | 21:54 |
| parazyd | Noted | 21:54 |
| Wizzup | I will have to add InterfaceAdded and InterfaceRemoved notifiers, and all the code will have to accept interface data as arg, and it will need to be managed and won't be available on start, so it's all getting much more complicated now unfortunately | 21:54 |
| Wizzup | I can't write that code now | 21:54 |
| parazyd | Can I help? | 21:55 |
| Wizzup | well, it's just that the wpa_supplicant change made this more apparent | 21:56 |
| Wizzup | as long as we push them in sync this shouldn't cause major headaches, but it's something that needs to be done | 21:56 |
| Wizzup | if you want to help, sure, but I imagine it's at least a few hours of work, maybe more | 21:56 |
| Wizzup | and also really we'd probably just pick the first interface we find anyway | 21:56 |
| parazyd | Agree on the interface choice. I guess this is what the kernel sets up first. | 21:57 |
| parazyd | btw. if needed we can also use eudev for iface naming | 21:58 |
| Wizzup | yes, but this is about the wpa supplicant dbus path for the interface | 21:59 |
| jk | hello #maemo-leste!! | 22:03 |
| jk | rock on! | 22:03 |
| jk | I need some assistance with my dorid4. i beleive I can't gert my wlan0 up to find and connect to wifi and lan | 22:05 |
| Wizzup | jk: hi, are you on -devel? | 22:06 |
| jk | eehmm.. I think not. I am new to irc | 22:07 |
| Wizzup | ok, so this has been a consistent problem since you tried maemo leste? | 22:08 |
| jk | nop, I tried to change my user password to be able to ssh as non-root with the command passwd | 22:09 |
| Wizzup | ok | 22:09 |
| Wizzup | what image did you flash? | 22:10 |
| jk | the problems started after that | 22:10 |
| jk | let me check... | 22:10 |
| jk | ... I flashed it last year, may I think... things have been going smooth. even the beowulf upgrade was fine | 22:11 |
| Wizzup | heh, neat | 22:12 |
| Wizzup | so some recent upgrade broke wifi for you? | 22:12 |
| jk | no,I don't thik it was an upgrade. I was connecting over ssh form my laptop but only as root, as I had no "user" password (I still only have the root-user and the user-user)) | 22:15 |
| jk | so I restarted my Droid4, booted to the emergency shell or something like that | 22:15 |
| jk | logged in as root and ran passwd. I rewrite the root password to the same (not surre if this changed anything in the deeper level) and then changed the user password... once throrough passwd -expire but HIldon could not start automatically, so I changed it back again to empty | 22:17 |
| jk | this last time again over the leste emergency shell. with passwd user (if I remember right) and let it blank and also delited the hash from the shadowfile following some internet guide | 22:20 |
| jk | ...using nano | 22:20 |
| freemangordon | Wizzup: ok, so, lets agree at least that icd2 plugin will expose dbus interface to get the current wlan interface, otherwise we'll have to duplicate the same code in connui | 22:21 |
| freemangordon | or if you have a better idea, please share | 22:21 |
| uvos | why is connui interacting with the interface directly anyhow? | 22:25 |
| uvos | thats icd's job no? | 22:25 |
| Wizzup | jk: hm, ok. is the wlan0 interface visible in 'ifconfig'? and it wpa_supplicant running? ('ps xua | grep wpa') | 22:27 |
| Wizzup | uvos: all of this stems from the hidden wlan ap hell | 22:27 |
| Wizzup | uvos: freemangordon wants to port the interactive UI to add a hidden AP to the settings without using the settings | 22:27 |
| Wizzup | uvos: this requires active scans for the hidden ap, which means the UI has to interface with something | 22:28 |
| uvos | yeah no | 22:28 |
| uvos | lets not leak the implentation details of the layer below icd to the ui | 22:28 |
| uvos | why cant icd2 just get a new function to scann for hidden aps | 22:28 |
| Wizzup | so the maemo fremantle UI also violated this by just issuing a scan directly using wlancond | 22:29 |
| Wizzup | uvos: yes, I was thinking that as well- let's do this in icd2 | 22:29 |
| uvos | yeah well fremantle is badly desinged allmost everywhere | 22:29 |
| uvos | Wizzup: ok that sounds sane | 22:29 |
| Wizzup | I think this is also what freemangordon is suggesting | 22:29 |
| Wizzup | I am just not sure if it should be in the plugin or in icd2 itself | 22:29 |
| uvos | jk: did your device reboot during a recent update? | 22:31 |
| sicelo | Danct12: you said Dillo does JS? last i recall it didn't :-/ | 22:31 |
| jk | @uvos: not that I remember | 22:31 |
| uvos | i had to run dpkg --configure -a to get wifi back on bionic after a recent update rebooted before the pacakges where configured | 22:32 |
| jk | @Wizzup: I am checking... | 22:32 |
| uvos | jk: try that | 22:32 |
| jk | ok! | 22:32 |
| jk | dpkg --configure -a run successfully as root | 22:33 |
| jk | wpa_supplicant found on htop (with several flags) trying ps now... | 22:33 |
| uvos | if dpkg --configure -a did anything you have to reboot | 22:34 |
| uvos | also check ip addr for the wlan0 interface | 22:34 |
| Wizzup | parazyd: if this is the case, we really need to figure out that random reboot ^ | 22:35 |
| jk | ps xua | grep wpa shows also /sbin/wpa_supplicant -s -P /run/wpa_supplicant.wlan0.pid -i wlan0 -u -D nl80211,w ext -c /etc/wpa_supplicant/wpa_supplicant.conf | 22:35 |
| jk | ... rebooting now | 22:36 |
| sicelo | i don't recall leste having an /etc/wpa_supplicant/wpa_supplicant.conf ... | 22:37 |
| sicelo | s/having/having or using/ | 22:38 |
| jk | ... extra info: with vipw -s on the line of user i have "user::" 'cuz I deleted the password there after my manual passwd user change aiming for ssh from laptop... most entries under vipw -s have "<name>:*:..." | 22:40 |
| freemangordon | Wizzup: icd is by no means wlan only daemon, neither we shall tie it to wpa_supplicant, that's why I suggested either a daemon that exposes a common interface to both plugin and ui or dbus interface in the plugin itself | 22:41 |
| jk | ip addr shows me wlan0 as number 3: | 22:42 |
| freemangordon | To me having a separate daemon is better, but... | 22:42 |
| jk | 3: wlan0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000 | 22:43 |
| uvos | jk: i assum wifi dosent work after reboot | 22:43 |
| uvos | jk: please descibe symptoms in detail | 22:43 |
| uvos | jk: what dose not work mean | 22:43 |
| uvos | dose it not scann | 22:43 |
| uvos | not connect etc | 22:43 |
| jk | no scan so no connect. I can't get it up | 22:44 |
| uvos | ok so the network dialog dosent show anything | 22:45 |
| uvos | right? | 22:45 |
| jk | it usually connected automatically after boot (some seconds later) no I see no available connections on the gui | 22:45 |
| jk | jup, Select connection is empty | 22:45 |
| uvos | ok | 22:45 |
| jk | "No connections available" | 22:46 |
| uvos | what dose maserati-callibrate say | 22:46 |
| Wizzup | jk: did a reboot help after the dpkg --configure -a ? | 22:46 |
| jk | as root? | 22:46 |
| Wizzup | jk: the reboot? you can use the UI for it too | 22:47 |
| jk | no change after dpkg --configure -a and reboot | 22:47 |
| Wizzup | freemangordon: are you up for writing the code for that interface? | 22:47 |
| jk | maserati-calibrate as root? | 22:47 |
| uvos | yes | 22:47 |
| Wizzup | freemangordon: I think I'd like to take my hands off of this hidden AP business now, icd2 should support it OK now if it is configured from "Internet Connections" UI | 22:47 |
| jk | is shows MAC addr from NVS: ... | 22:47 |
| jk | Calibration already performed | 22:48 |
| uvos | ok | 22:48 |
| Wizzup | freemangordon: I'm fine with whatever solution you prefer, as long as it doesn't mean I have to write more dbus code | 22:48 |
| freemangordon | Wizzup: yes, why not | 22:48 |
| jk | I did that las year... | 22:48 |
| uvos | so its not that | 22:48 |
| freemangordon | I was thinking to reuse wlancond interface | 22:48 |
| uvos | jk ok yeah i thought cal might have goten lost somehow | 22:48 |
| freemangordon | afc I will first check if it fits for what we need it | 22:48 |
| uvos | im all for Wizzup writing more dbus code | 22:48 |
| freemangordon | :) | 22:49 |
| Wizzup | jk: so the sympton is that nothing shows up in the 'Internet Connections' status dialog? | 22:49 |
| uvos | dbus everywhere | 22:49 |
| uvos | everything dbus | 22:49 |
| Wizzup | uvos: I am fine if others write dbus code, but it breaks my brain | 22:49 |
| jk | yes, that is the symptom | 22:49 |
| uvos | jk: can you use wpacli to scann for networks | 22:49 |
| uvos | ? | 22:49 |
| Wizzup | that's wpa_cli | 22:49 |
| Wizzup | the commands are 'scan' and 'scan_results' | 22:49 |
| freemangordon | how's that different from calling dbus? | 22:50 |
| Wizzup | freemangordon: this is us helping jk debug his wifi issue, not related to our discussion :) | 22:50 |
| jk | i did # wpa_cli | 22:50 |
| jk | > scan | 22:50 |
| jk | OK | 22:50 |
| jk | <mobile.scjorge@gmail.com><3>CTRL-EVENT-SCAN-STARTED | 22:51 |
| Wizzup | did you also see 3>CTRL-EVENT-SCAN-RESULTS ? | 22:51 |
| jk | <3>CTRL-EVENT-SCAN-RESULTS | 22:51 |
| freemangordon | Wizzup: ah, ok :) | 22:51 |
| jk | <3>WPS-AP-AVAILABLE | 22:51 |
| jk | <3>CTRL-EVENT-NETWORK-NOT-FOUND | 22:51 |
| Wizzup | jk: ok, does scan_results list anything? | 22:52 |
| Wizzup | no need to paste/type it all here | 22:52 |
| jk | > scan result gives me the same 4 lines | 22:52 |
| Wizzup | jk: btw, you could maybe use usbnet (if you know how to) to ssh into the device for easier access/debugging | 22:52 |
| Wizzup | scan_results ? | 22:52 |
| Wizzup | (note the 's' at the end) | 22:52 |
| jk | thank for the hint on the s! | 22:52 |
| jk | still the same 5 lines (the same 4 plus an OK on the first position) | 22:53 |
| Wizzup | hm, I think from the '<3>WPS-AP-AVAILABLE' we know it does find networks | 22:54 |
| jk | I don't know usbnet... | 22:54 |
| Wizzup | I am trying to think what the next steps debugging wise are. I don't really know. Maybe this: | 22:55 |
| Wizzup | dpkg-reconfigure libicd-network-wpasupplicant icd2 | 22:55 |
| uvos | "scan result gives me the same 4 lines" | 22:55 |
| Wizzup | is 'icd2' running ? | 22:55 |
| uvos | what four lines? | 22:55 |
| Wizzup | uvos: I think the entries might have expired perhaps | 22:55 |
| jk | 4 lines: <3>CTRL-EVENT-SCAN-STARTED \ <3>CTRL-EVENT-SCAN-RESULTS \ <3>WPS-AP-AVAILABLE \ <3>CTRL-EVENT-NETWORK-NOT-FOUND | 22:56 |
| sicelo | if wpa_cli has 'issues', then the problem is unlikely to be in libicd-* or icd* | 22:57 |
| jk | oh! scan results is actually scan_results?! | 22:59 |
| Wizzup | jk: yes | 23:00 |
| Wizzup | freemangordon: something else I noticed (of course), a dbus-issued scan is not the same as a scan issued over the control socket interface it seems | 23:01 |
| Wizzup | one calls wpa_supplicant_trigger_scan and the others calls wpa_supplicant_req_scan | 23:01 |
| Wizzup | freemangordon: and of course only the control socket 'scan' seems to do the right thing, I believe | 23:01 |
| jk | ok... I do see connections, wifis | 23:02 |
| freemangordon | Wizzup: I see | 23:03 |
| jk | bssid / frequenc / signal level / flags / ssid | 23:03 |
| freemangordon | are those blocking calls? | 23:03 |
| Wizzup | jk: ok, so we need to figure out why the data is not making it to the UI | 23:06 |
| Wizzup | freemangordon: no | 23:06 |
| Danct12 | sicelo: i never said dillo supports js | 23:07 |
| sicelo | Ok. I must have misunderstood. Apologies | 23:08 |
| freemangordon | Wizzup: ok, I'll think about it for a while, but maybe implementing wlancond interface by using socket interface looks sane, albeit lots of work. | 23:09 |
| freemangordon | in a separate daemon that is | 23:09 |
| Wizzup | freemangordon: the socket interface is extremely simple, but you might have the same perms problems | 23:10 |
| freemangordon | that way we'll be able to use Nokia icd2 wlan plugin. And your work won;t be in vain, as most of it will go in that daemon | 23:10 |
| freemangordon | that daemon will run as root | 23:10 |
| Wizzup | what do you mean? nokia icd2 wlan plugin? | 23:10 |
| sicelo | wlancond, iiuc | 23:10 |
| jk | Wizzup: ok. we can check that. so the password overwrites that I did have no importance? | 23:10 |
| freemangordon | Wizzup: https://github.com/community-ssu/libicd-network-wlan | 23:11 |
| uvos | jk: yes | 23:11 |
| uvos | you can set user pw with no problem | 23:11 |
| uvos | but leste currenly explodes if you dont have sudo without pw | 23:11 |
| Wizzup | freemangordon: if we really go down that route I think we just want to look at NetworkManager | 23:12 |
| uvos | +1 | 23:12 |
| jk | I did rewrire the sudo password to the standard | 23:12 |
| uvos | uh | 23:12 |
| uvos | yeah that causes lots of issues | 23:12 |
| freemangordon | Wizzup: why is that? | 23:12 |
| uvos | jk: please restore default sudo config | 23:13 |
| freemangordon | I mean - it is obvious that wpa_supplicant dbus interface has issues, no? | 23:13 |
| uvos | sudo usage in leste is badly broken | 23:13 |
| jk | ehm the root password I mean! | 23:13 |
| uvos | oh | 23:13 |
| uvos | ok | 23:13 |
| uvos | thats fine | 23:13 |
| Wizzup | freemangordon: there is a lot of work going into rewriting what essentially already exists in some other form, and it took me several months to write the wpa supplicant plugin for icd2, and I did it because you wanted it that way (which is fine), but I really don't understand why we would need to give wlancond some wpa supplicant interface and then switch to the other nokia plugin which likely will | 23:13 |
| Wizzup | have way more issues and backwards things | 23:13 |
| freemangordon | Wizzup: hold on, I was just sharing an idea | 23:14 |
| Wizzup | ok, sorry, I'm not upset | 23:14 |
| Wizzup | it just feels like a very niche interface | 23:14 |
| freemangordon | on the other hand we still don;t have solution for coule of things | 23:14 |
| freemangordon | not to say we still don;t support ad-hoc | 23:14 |
| jk | ok. I only rewrote the "su"-password. sudo is still possible without password | 23:14 |
| uvos | jk: thats fine | 23:15 |
| uvos | and not the issue | 23:15 |
| Wizzup | freemangordon: It actually is not very hard to support ad hoc I believe | 23:15 |
| uvos | networkmanager supports adhoc | 23:15 |
| Wizzup | freemangordon: I think the wpa_supplicant dbus interface isn't that bad, but apparently it's awkward regarding hidden APs | 23:15 |
| uvos | and bt pan and ..... | 23:15 |
| freemangordon | uvos: please, stop that | 23:15 |
| freemangordon | I guess wicd does it too | 23:15 |
| freemangordon | none of them is mobile aware | 23:16 |
| uvos | bullshit | 23:16 |
| freemangordon | they don;t integrate with mce for example | 23:16 |
| uvos | mce has no role in this | 23:16 |
| freemangordon | yes, it does | 23:16 |
| uvos | no | 23:16 |
| uvos | it just consumes the offline state | 23:16 |
| uvos | it dose nothing with it | 23:16 |
| freemangordon | ok, it seems leste is going in a direction I am not sure I support | 23:17 |
| freemangordon | the project was started as a try to bring fremantle experience based on modern distro | 23:18 |
| Wizzup | I am not advocating for using networkmanager per se, I am just frustrated with the amount of work required for a very niche feature (hidden aps) when I'd rather focus on bigger things | 23:18 |
| uvos | i dont see why fremantle experiance has anything to do with what bacgriound deamons we use | 23:18 |
| uvos | we want fremantles ui over a modern distro no? | 23:18 |
| freemangordon | uvos: you dont; have experience with fremantle, right? | 23:19 |
| uvos | best case scenario we dont have to maintain anything but the ui | 23:19 |
| uvos | ofc thats not feasble in all cases | 23:19 |
| uvos | but the less we have to maintain the better | 23:19 |
| freemangordon | in which case it will turn into a mock of fremantle | 23:19 |
| sicelo | No. We wanted fremantle UX, at least I did. Not just UI | 23:19 |
| freemangordon | just a shell | 23:19 |
| freemangordon | sicelo: exactly my point | 23:19 |
| freemangordon | and I still want it | 23:19 |
| uvos | right hildon-desktop should just be a xdg shell (or session in xdg language) | 23:20 |
| freemangordon | (fremantle UX) | 23:20 |
| Wizzup | freemangordon: I think the current icd2 mostly supports the fremantle ui/ux, just not yet with hidden aps | 23:20 |
| freemangordon | uvos: again, you have no experience with fremantle, I wonder how you are so sure what HD should become? | 23:20 |
| Wizzup | if there are specific UX you would like, it would be good to write them out clearly because otherwise there's no discussion to be had | 23:21 |
| freemangordon | fremantle is way more then just HD | 23:21 |
| uvos | i ment "hildon" not hd | 23:21 |
| uvos | as in all of the ui stuff | 23:21 |
| Wizzup | jk: did going to passwordless sudo solve your problem? | 23:21 |
| freemangordon | Wizzup: part of the UX we're discussing is the ability to add hidden wlan from the wizard | 23:21 |
| uvos | Wizzup: no it dident he had passwordless sudo allready | 23:21 |
| freemangordon | niche or not, it is there and users try to use it | 23:21 |
| freemangordon | see the issue we have opened | 23:22 |
| Wizzup | freemangordon: let me just note that with -devel you can add it to "Internet Connections" and it will work | 23:22 |
| Wizzup | you just check the checkbox that says it's hidden | 23:22 |
| freemangordon | sure, but thats not what I am talking about | 23:22 |
| Wizzup | I agree there are outstanding issues (and we can get back to discussing those when you want), but I think if we make a poll, more users will care about some other parts :P | 23:22 |
| uvos | freemangordon: one thing im sure of is that we are like 4 people | 23:23 |
| jk000 | ... sorry if I had someone else's nickname! :P | 23:23 |
| freemangordon | uvos: so? it was only 1 at the very beginning, how that changes the things? | 23:23 |
| uvos | and what 4 people can credbly maintian and improve in spare time is.. not alot | 23:23 |
| uvos | so we must share code with others as mutch as possible | 23:23 |
| freemangordon | uvos: then we'll fail | 23:23 |
| uvos | that meas shareing with deskop where possible | 23:24 |
| freemangordon | not the end of the world | 23:24 |
| uvos | and with phosh and plamo where not | 23:24 |
| freemangordon | but better fail than romise and not deliver what is promised | 23:24 |
| Wizzup | freemangordon: uvos: we don't know if networkmanager is no longer mobile aware and what we can do with it, but from a practical point of view I think we should stick with what we have until we have a more functional experience | 23:24 |
| Wizzup | freemangordon: uvos: I don't think this discussion about UX is relevant, I think we more or less agree on the goals, and if someone magically did all the work to support fremantle interfaces (functionally equivalent) with networkmanager backend we'd use it, but it's not there, and it will also take a lot of work to get there | 23:25 |
| Wizzup | freemangordon: uvos: other mobile distros will also try to make it as mobile as possible, and some of them use networkmanager, so I am sure they will eventually have to tackle some of the same problems | 23:25 |
| Wizzup | freemangordon: uvos: Personally I am totally fine with telling the user to configure it through the Settings -> Internet Connections, and I'm pretty sure the user would be happy too. | 23:26 |
| Wizzup | freemangordon: uvos: My point really was that I don't -personally- want to invest more time in a niche side of an already niche feature that we've really now implemented already | 23:26 |
| uvos | sure that sounds mostly sane | 23:27 |
| uvos | im not sure on the same goals part tho | 23:27 |
| Wizzup | So I'd be fine with leaving the bug report open specifically for the easy wlan wizard with hidden ap support | 23:27 |
| uvos | i think freemangordon has other goals | 23:27 |
| freemangordon | Wizzup: UI is just missing a single call (be it dbus or whatever) to get hidden wlan properties at it should work. We just need to agree upon how that call shall be issued and I'll implement it | 23:27 |
| Wizzup | freemangordon: sounds good | 23:27 |
| freemangordon | *and it | 23:27 |
| Wizzup | jk000: I am trying to think of how to help you debug this. is icd2 running? | 23:28 |
| Wizzup | jk000: 'ps xua | grep icd2' | 23:28 |
| jk000 | thanx a lot!! | 23:28 |
| jk000 | let's see | 23:28 |
| freemangordon | but, I involved the other things that we miss currently (like reg domain, adhoc etc) so while we plan how to implement that call which is needed for hiden wlans, if possible, to fix or at least make those issues fixanle with less effort | 23:29 |
| jk000 | yes, it is | 23:29 |
| Wizzup | freemangordon: I think we can really make all of that work with the icd2 plugin I made | 23:29 |
| freemangordon | ok, lets spend some time tomorrow to agree upon the interface | 23:30 |
| Wizzup | jk000: so I am thinking that if I had access locally, start icd2 manually with low log level, and monitor the logs, but I am not really sure what to tell you to look for, and given that you have no connectivity, if I find it hard to figure out how to continue from here | 23:30 |
| freemangordon | UI->icd2 plugin that is | 23:30 |
| Wizzup | freemangordon: ok | 23:30 |
| Wizzup | jk000: so if you run this as root '/etc/init.d/icd2 stop' and 'icd2 -l0' icd2 will write info to /var/log/daemon.log, but other services also log to it | 23:31 |
| Wizzup | you could look at those logs in real time with 'tail -f /var/log/daemon.log' and maybe see if you can spot the problem, but all of this is hard on the phone screen | 23:31 |
| Wizzup | jk000: do you think you could set up usbnet? | 23:31 |
| Wizzup | https://leste.maemo.org/Status/USB_Peripheral | 23:31 |
| jk000 | I beleive I can setup usbnet! | 23:32 |
| jk000 | thanxs for the link!! ... I am getting sleepy will continue another day | 23:33 |
| Wizzup | freemangordon: fwiw I just connected my droid4 to my hidden ap | 23:34 |
| Wizzup | (using packages from -devel) | 23:34 |
| Wizzup | jk000: we can continue this tomorrow if you'd like, usbnet is powerful tool in any case | 23:35 |
| Wizzup | (I'm also getting tired) | 23:35 |
| jk000 | yeah! I am hapy to learn usbnet! i don't know when I will be ble to connect again... the droid 4 is a for me a side project, but a very interesting one! I just got this (https://duckduckgo.com/?q=droid++bionic+hd+dock&t=qupzilla&iar=images&iax=images&ia=images) today! mouse and keyboard work simultaneously | 23:39 |
| Wizzup | oh, that's cool | 23:40 |
| Wizzup | So you're saying that the usb host most just works with leste on the d4? | 23:40 |
| Wizzup | If you could make photo or video of that, we'd love to see it | 23:40 |
| jk000 | ... still struggling with my DAC to get highest quality audio | 23:40 |
| jk000 | I beleive usbnet will work i just gotta learn it. which will be added to my toolbox! yeah! | 23:41 |
| jk000 | I'll try to make a video, where do I send it?? | 23:41 |
| Wizzup | jk000: conceptually it should be as simple as 'connect the droid 4 to a PC', 'set up the IP on the PC side', ssh to the phone | 23:41 |
| Wizzup | jk000: if you can upload it somewhere and link it, that's perhaps easiest | 23:41 |
| Wizzup | I don't know what you preferred method is, you can also upload it to some site like youtube or the foss ones | 23:42 |
| jk000 | jup, and in theory that is extensible to many devices, right? i will read into that! | 23:42 |
| Wizzup | jk000: all leste supported devices should expose it, I use it all the time for debugging | 23:42 |
| jk000 | alright... I'll see what I can do and get back here. I am still catching up on irc-tech. 1st time runnign Hexchat and seems preatty easy. with irssi I had some issues as I am a newbie | 23:44 |
| jk000 | thanks, leste-community!! keep up the pioneer work and enable freedom further! | 23:48 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!