Frisbee Network Manager - Beta 2

Configuration wizards, scanners, remote desktop, etc.
Message
Author
User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

Re: Another frisbee update

#221 Post by mavrothal »

rerwin wrote: - The "restart X" problem is managed; to restore normal network_tray function after an X restart, go into frisbee and click on either "Restart Networks" or "Restart DHCP".
That was always the case and does not look like a proper solution.
As far as I can see the problem is the call from /root/Startup/network_tray_modeset of frisbee_mode_enable.
$(ls /usr/local/lib/X11/mini-icons/networkdead-eth* 2>/dev/null) is always true since is a symlink to networkdead.xpm and thus "killall -35 network_tray" is executed. The last command is the offending one that actually turns the networkup icon to networkdead.
There are a couple of strange things with this. "killall -l" does not list "35" as an option so I'm not sure what is it doing. :?
Issuing "killall -35 network_tray" from the terminal can turn networkup to networkdead only soon (<30sec) after X restart but not latter :? :?
Setting permissions of /root/Startup/network_tray_modeset to 000 , "solves" this problem but generates another. Now network tray icon shows as connected as soon as the interface is up regardless of an established connection. :roll:
What "killall -35 network_tray" is doing exactly?...
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

Re: Another frisbee update

#222 Post by rerwin »

mavrothal wrote:That was always the case and does not look like a proper solution.
As far as I can see the problem is the call from /root/Startup/network_tray_modeset of frisbee_mode_enable.
$(ls /usr/local/lib/X11/mini-icons/networkdead-eth* 2>/dev/null) is always true since is a symlink to networkdead.xpm and thus "killall -35 network_tray" is executed. The last command is the offending one that actually turns the networkup icon to networkdead.
There are a couple of strange things with this. "killall -l" does not list "35" as an option so I'm not sure what is it doing. :?
Issuing "killall -35 network_tray" from the terminal can turn networkup to networkdead only soon (<30sec) after X restart but not latter :? :?
Setting permissions of /root/Startup/network_tray_modeset to 000 , "solves" this problem but generates another. Now network tray icon shows as connected as soon as the interface is up regardless of an established connection. :roll:
What "killall -35 network_tray" is doing exactly?...
mavrothal,
Thanks for your comments. From your mention of network_tray_modeset, I find that I missed an update to use the frisbee.conf file instead of the flags. So, the script does not recognize that frisbee mode is in effect, so does not run frisbee_mode_enable. Fixed for next upload (20130320).

The "/usr/local/lib/X11/mini-icons/networkdead-eth*" test is to determine whether the network_tray program can accept the signals created for frisbee. network_tray-2.5 cannot, and would terminate if signaled. With 2.6 as the current version, yes, the test will always be true.

Re: "killall -l" does not list "35", that is because it is user defined. Although I cannot right now find my source, I found that the named signals are numbered below 32; numbers 32-or-33 (it is ambiguous) through 63-or-64 are left to users to define. I define 35-40 for frisbee to use to communicate with network_tray. They are not valid in any other context. I found that the prefix SIGRT is used for such signal numbers. As the name SIGRTfrisbeeup implies, it tells network_tray that frisbee mode is in effect, so the extended icons are to be used. In my next upgrade of network_tray, I plan to drop that signal and SIGRTfrisbeedown (40) and always display the full set of icons (once the logic is cleaned up).

Re: the restarts after restarting X, actually, they have been ineffective because a newly started network_tray does not know about frisbee mode until told through SIGRTfrisbeeup, and that was not done by the two restart functions. Now, they always send that signal. The bottom line is that without that signal after an X restart, network_tray behaves as it always has for standard puppies.
Richard

EDIT: I just now tried X restart on a clean puppy with only the woof_updates package installed. After restarting X, the tray icon has the red X. This may have been due to the installing of the package without a subsequent reboot. After I restarted networks and restored the correct icon state, a subsequent X restart did not result in the "X" icon. I have not sorted that out. So, please disregard the X icon after woof_update installation followed by a "restart X". The restart-networks/DHCP should straighten things out. The corrected frisbee kit is now uploaded:
http://www.murga-linux.com/puppy/viewto ... 212#693212

Atle
Posts: 596
Joined: Wed 19 Nov 2008, 12:38
Location: Oslo, Norway
Contact:

#223 Post by Atle »

Frisbee also let you use a Android phone as a router, connecting using USB.

In the Andriod menu, you need to be there and ready to press the icon where as USB connection is set, within a second or two, if you got a storage device in your phone. This as you insert the USB cable to the PC or Android.

Otherwise the storage device will snag the USB connection. Androids WITHOUT a storage device can take their time and press the check box for USB connection in what i call a normal pace...

Your android will route a Wlan connection or a Mobile internet connection trough the USB cable.

The big benefit of course is that there is no such thing as running out of battery:-)

User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

Re: Another frisbee update

#224 Post by mavrothal »

rerwin wrote: The corrected frisbee kit is now uploaded:
http://www.murga-linux.com/puppy/viewto ... 212#693212
I think there is a problem with the upload. There is no frisbee pet in it.
Here is what I get expanding the tarball

Code: Select all

# tar xvzf frisbee-1.0_puppy-5.5_kit-20130320.tar.gz 
dhcpcd-5.6.4-patched_dropwait-i486.pet
woof_updates-20130314.pet
frisbee-1.0_puppy-5.5_kit-20130320.tar.gz
# tar xvzf frisbee-1.0_puppy-5.5_kit-20130320.tar.gz 
dhcpcd-5.6.4-patched_dropwait-i486.pet
woof_updates-20130314.pet
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

#225 Post by mavrothal »

Tried 0314 frisbee on precise 5.5 with ubuntu openvpn and was poor.
ie as soon as it connected to tun0 dropped the wlan0 connection and then was trying to reconnect to tun0 instead of wlan0.
Maybe was a "misbehaving" open AP, but did anyone tried openvpn with frisbee-1.0-<date> on precise OK?
(Precise 5.4.3 with Fribee-beta2 was OK with openvpn)
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#226 Post by rerwin »

Atle wrote:Frisbee also let you use a Android phone as a router, connecting using USB.

In the Andriod menu, you need to be there and ready to press the icon where as USB connection is set, within a second or two, if you got a storage device in your phone. This as you insert the USB cable to the PC or Android.

Otherwise the storage device will snag the USB connection.
That sounds like something usb_modeswitch should handle. If you have an Android with the storage device, could you post the hardware ID of the storage device part as well as that of the phone/router part? We can watch for the possibility that usb_modeswitch eventually can switch to the nonstorage mode.

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#227 Post by rerwin »

mavrothal,
Thanks for catching my mistake in updating the kit. I have corrected that, after 2 days away working another project.
http://www.murga-linux.com/puppy/viewto ... 212#693212
mavrothal wrote:Tried 0314 frisbee on precise 5.5 with ubuntu openvpn and was poor.
ie as soon as it connected to tun0 dropped the wlan0 connection and then was trying to reconnect to tun0 instead of wlan0.
Maybe was a "misbehaving" open AP, but did anyone tried openvpn with frisbee-1.0-<date> on precise OK?
(Precise 5.4.3 with Fribee-beta2 was OK with openvpn)
This is new territory for me, but one I want to resolve. I may have found a cause for the reconnection attempts. At least I prevent the repeated testing of the enable-wifi checkbox that might cause a change in connections. Or, at least we can rule that out as a cause. The change is in the 20130322 version of frisbee-1.0.

If the problem continues, could you send me a pdiag (--wpa) file so I can try to understand what is happening. Any further information you have on that operation might be helpful. Also helpful would be a pnetworkdiag of your beta2 run. (Both of those diag files can be initiated from F/frisbee's Diagnostic dialog.)

Does the same thing happen if you connect via ethernet instead of wireless? (Probably not, I expect.)
Richard

An afterthought: The change I made stops an every-15-second test of the connection status and recovery if the connection drops. So, the connection may not be as resilient as before. The change may have to come out because of that. R

User avatar
peebee
Posts: 4370
Joined: Sun 21 Sep 2008, 12:31
Location: Worcestershire, UK
Contact:

#228 Post by peebee »

Hi Richard

Tested the above package on my laptop.

Seems to be working as designed for me.

My thoughts:

if you select "no" on initial install it seems a bit odd that you get both the "detecting devices" and "scanning for signals" popup messages but then get a screen with no detected signals and both wifi tick boxes disabled.

if after a "no" install I click the connect icon on the desktop and then deliberately choose to try Frisbee then it would be better if the 2 wifi tick boxes were enabled by default.

or...I agree with mavrothal that a "Scan for wifi" button which also enabled the 2 tick boxes would be a more intuitive interface.

Just "My Thoughts"
Cheers
PeeBee
ImageLxPup = Puppy + LXDE
Main version used daily: LxPupSc; Assembler of UPups, ScPup & ScPup64, LxPup, LxPupSc & LxPupSc64

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#229 Post by rerwin »

peebee wrote:if you select "no" on initial install it seems a bit odd that you get both the "detecting devices" and "scanning for signals" popup messages but then get a screen with no detected signals and both wifi tick boxes disabled.

if after a "no" install I click the connect icon on the desktop and then deliberately choose to try Frisbee then it would be better if the 2 wifi tick boxes were enabled by default.

or...I agree with mavrothal that a "Scan for wifi" button which also enabled the 2 tick boxes would be a more intuitive interface.
Well...I caved. I see your point in that there should be no impediment to the first encounter of frisbee after installation. I now suppress the "scanning for signals" message, but keep the "detecting" message, which covers the long frisbee startup, whatever it is doing. The text is probably not accurate, but I am open to better text. I think frisbee is mainly preparing its dialogs.

I am still not completely satisfied with the "don't auto-connect" logic because it actually just does not start wpa_supplicant and thus does not display the available networks without connecting to the last-known network. I need to see if there is a way to start wpa_supplicant but prevent it from connecting.

Also, I was bothered by the tray icons in precise pup 5.5, so created replacements and added that package to the kit. The green added to the towers and ethernet cable indicate that the interfaces are "bound", which I interpret as meaning that they have IP addresses.

Thank you for your thoughts on this. Try version 20130324 at the usual place on page 15. More work to do for the don't-connect situation, in due time.
Richard

User avatar
peebee
Posts: 4370
Joined: Sun 21 Sep 2008, 12:31
Location: Worcestershire, UK
Contact:

#230 Post by peebee »

rerwin wrote:Try version 20130324 at the usual place on page 15. More work to do for the don't-connect situation, in due time.
Richard
All looks good to me - many thanks
Peebee
ImageLxPup = Puppy + LXDE
Main version used daily: LxPupSc; Assembler of UPups, ScPup & ScPup64, LxPup, LxPupSc & LxPupSc64

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#231 Post by rerwin »

peebee,
Thanks for checking out yesterday's product. Although I implied that I would take my time in working the "don't-auto-connect" logic, I worked on it today and have uploaded the result, as kit 30130325.

Since I could not find any way to stop wpa_supplicant from connecting to the best network it sees, I decided to use a temporary configuration file for it when automatic is deselected. That way, wpa_supp starts with an empty conf file at startup. That is also useful for avoiding retention of the configuration information after a session is ended, for those with security concerns.

I also fixed a problem I introduced yesterday, where the non-frisbee tray icons are used at startup instead of the frisbee set. Fortunately, you are the only person to use that version. I hope that today's kit completes the necessary fixes for awhile. But I am ready to pounce if you find anything wrong with it. Thanks for your continuing help.
Richard

User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

#232 Post by mavrothal »

Hi Richard,
on the road without a lot of free time but plenty of diverse conditions for frisbee to test (v0325-included).
Re tray icon:It still shows wifi as disconnected after an X restart

Re VPN: Though no time to test properly and produce logs, what I see is that when wlan0 is down tun0 shows as still up in ifconfig (and tray icon shows frisbee as connected). When wlan0 is reconnected tun0 although disconnected from the host does not reconnect. Thus you are out of network since you try to tunnel through a broken VPN.
A network reset remedies the problem reconnecting both wlan0 and then tun0 as needed.
More as time permits.

So it would appear that still changes in the network status are not properly/timely detected.
Is network status detection runs as a daemon or is expecting changes input from other apps?
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#233 Post by rerwin »

mavrothal wrote:Re tray icon:It still shows wifi as disconnected after an X restart
That will have to be the case until I update network_tray, which I expect will not need the signals to set frisbee mode and identify active interfaces after an X restart. My intent is to eliminate all of the 6 signals and support only an extended set of "frisbee mode" icons, which will show traffic correctly when both wifi and ethernet interfaces are installed.
Re VPN: Though no time to test properly and produce logs, what I see is that when wlan0 is down tun0 shows as still up in ifconfig (and tray icon shows frisbee as connected). When wlan0 is reconnected tun0 although disconnected from the host does not reconnect. Thus you are out of network since you try to tunnel through a broken VPN.
That, too, is a network_tray issue that I would like to address in my update to it. I suspect it treats tun0 as an ethernet interface, so needs a way to discriminate it from the others (other than depending on the "tun", unless there is no other way). To help me with that, could you gather some info on what network_tray would be looking at? Specifically, please run these commands and send/post their output:
  • ls /sys/class/net/tun0/ -l
    cat /sys/class/net/tun0/dormant 2>&1
(That is ls...dash-ell.) Please get this info for both cases: connected successfully and with wlan0 down. One problem I can imagine is how to know which wlan/eth interface is handling the VPN, if both are being used, to know whether the VPN is affected by one of them going down.
A network reset remedies the problem reconnecting both wlan0 and then tun0 as needed.
Do both reconnect automatically with the reset, or do you have to re-establish the VPN connection? I suppose that if tun0 is considered still active, then reconnecting wlan0 might be all that is needed. But that seems a stretch.
So it would appear that still changes in the network status are not properly/timely detected.
Is network status detection runs as a daemon or is expecting changes input from other apps?
It doesn't work quite that way. The network_tray program runs as a daemon that accepts signals 35-40 to determine the situation. Restarting X apparently kills and restarts network_tray, losing the knowledge it had accumulated from the signals over time. So, it thinks there are no active interfaces. It thinks it is not in frisbee mode until something tells it by a new signal (35) -- I am not sure where that comes from, so added it to the two restart buttons. Before that signal, the original, non-frisbee case is in effect, which sees whatever network is up. When switched to frisbee mode, it does not see any active interfaces (red X) because it needs the signals to set indicators. Either of the restarts causes dhcpcd to restart, thus triggering the signals that interfaces are up. My plan for my network_tray update is to use the "old" way of detecting active interfaces, but to discriminate between ethernet and wireless without depending on the signals.
Richard

User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

#234 Post by mavrothal »

rerwin wrote: Specifically, please run these commands and send/post their output:
  • ls /sys/class/net/tun0/ -l
    cat /sys/class/net/tun0/dormant 2>&1

Code: Select all

# ls /sys/class/net/tun0/ -l
total 0
-r--r--r-- 1 root root 4096 Mar 26 21:21 address
-r--r--r-- 1 root root 4096 Mar 26 21:21 addr_len
-r--r--r-- 1 root root 4096 Mar 26 21:21 broadcast
-r--r--r-- 1 root root 4096 Mar 26 21:21 carrier
-r--r--r-- 1 root root 4096 Mar 26 21:21 dev_id
-r--r--r-- 1 root root 4096 Mar 26 21:21 dormant
-r--r--r-- 1 root root 4096 Mar 26 21:21 duplex
-r--r--r-- 1 root root 4096 Mar 26 21:21 features
-rw-r--r-- 1 root root 4096 Mar 26 21:21 flags
-r--r--r-- 1 root root 4096 Mar 26 21:21 group
-rw-r--r-- 1 root root 4096 Mar 26 21:21 ifalias
-r--r--r-- 1 root root 4096 Mar 26 21:21 ifindex
-r--r--r-- 1 root root 4096 Mar 26 21:21 iflink
-r--r--r-- 1 root root 4096 Mar 26 21:21 link_mode
-rw-r--r-- 1 root root 4096 Mar 26 21:21 mtu
-r--r--r-- 1 root root 4096 Mar 26 21:21 operstate
-r--r--r-- 1 root root 4096 Mar 26 21:21 owner
drwxr-xr-x 2 root root    0 Mar 26 21:21 power
-r--r--r-- 1 root root 4096 Mar 26 21:21 speed
drwxr-xr-x 2 root root    0 Mar 26 21:21 statistics
lrwxrwxrwx 1 root root    0 Mar 26 21:21 subsystem -> ../../../../class/net
-r--r--r-- 1 root root 4096 Mar 26 21:21 tun_flags
-rw-r--r-- 1 root root 4096 Mar 26 21:21 tx_queue_len
-r--r--r-- 1 root root 4096 Mar 26 21:21 type
-rw-r--r-- 1 root root 4096 Mar 26 21:21 uevent
# cat /sys/class/net/tun0/dormant 2>&1
0
#
# ifconfig wlan0 down
# ls /sys/class/net/tun0/ -l
total 0
-r--r--r-- 1 root root 4096 Mar 26 21:21 address
-r--r--r-- 1 root root 4096 Mar 26 21:21 addr_len
-r--r--r-- 1 root root 4096 Mar 26 21:21 broadcast
-r--r--r-- 1 root root 4096 Mar 26 21:21 carrier
-r--r--r-- 1 root root 4096 Mar 26 21:21 dev_id
-r--r--r-- 1 root root 4096 Mar 26 21:21 dormant
-r--r--r-- 1 root root 4096 Mar 26 21:21 duplex
-r--r--r-- 1 root root 4096 Mar 26 21:21 features
-rw-r--r-- 1 root root 4096 Mar 26 21:21 flags
-r--r--r-- 1 root root 4096 Mar 26 21:21 group
-rw-r--r-- 1 root root 4096 Mar 26 21:21 ifalias
-r--r--r-- 1 root root 4096 Mar 26 21:21 ifindex
-r--r--r-- 1 root root 4096 Mar 26 21:21 iflink
-r--r--r-- 1 root root 4096 Mar 26 21:21 link_mode
-rw-r--r-- 1 root root 4096 Mar 26 21:21 mtu
-r--r--r-- 1 root root 4096 Mar 26 21:21 operstate
-r--r--r-- 1 root root 4096 Mar 26 21:21 owner
drwxr-xr-x 2 root root    0 Mar 26 21:21 power
-r--r--r-- 1 root root 4096 Mar 26 21:21 speed
drwxr-xr-x 2 root root    0 Mar 26 21:21 statistics
lrwxrwxrwx 1 root root    0 Mar 26 21:21 subsystem -> ../../../../class/net
-r--r--r-- 1 root root 4096 Mar 26 21:21 tun_flags
-rw-r--r-- 1 root root 4096 Mar 26 21:21 tx_queue_len
-r--r--r-- 1 root root 4096 Mar 26 21:21 type
-rw-r--r-- 1 root root 4096 Mar 26 21:21 uevent
# cat /sys/class/net/tun0/dormant 2>&1
0
#ifconfig
lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:784 errors:0 dropped:0 overruns:0 frame:0
          TX packets:784 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:59936 (58.5 KiB)  TX bytes:59936 (58.5 KiB)

tun0      Link encap:UNSPEC  HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00  
          inet addr:10.66.0.26  P-t-P:10.66.0.25  Mask:255.255.255.255
          UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1500  Metric:1
          RX packets:120 errors:0 dropped:0 overruns:0 frame:0
          TX packets:84663 errors:0 dropped:82731 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          RX bytes:88978 (86.8 KiB)  TX bytes:125512995 (119.6 MiB)

#
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#235 Post by rerwin »

mavrothal,
Thanks for the listings. I see these differences between the directories:
  • /sys/class/net/tun0/ has:
    features
    group
    owner
    tun_flags


    /sys/class/net/eth0/ has:
    addr_assign_type
    device ->
    netdev_group
    queues

    ifconfig tun0 has:
    UP POINTOPOINT RUNNING
The presence of the "tun_flags" file seems like a good signature of a point-to-point interface. Maybe, too, the absence of the "device" link. When you get a chance, could you do a "cat /sys/class/net/tun0/... for the 4 files unique to tun0? I am saving this info for when I get to network_tray.
Richard

User avatar
mavrothal
Posts: 3096
Joined: Mon 24 Aug 2009, 18:23

#236 Post by mavrothal »

rerwin wrote:When you get a chance, could you do a "cat /sys/class/net/tun0/... for the 4 files unique to tun0? I am saving this info for when I get to network_tray.
Richard
There you go.

Code: Select all

######### waln0 up #########

# pwd
/sys/class/net/tun0
# cat features
0x0
# cat group
-1
# cat owner
-1
# cat tun_flags
0x3001
#
# ifconfig wlan0 down
#
# cat features
0x0
# cat group
-1
# cat owner
-1
# cat tun_flags
0x3001


# ifconfig wlan0 up
# ls -al
total 0
drwxr-xr-x 4 root root    0 Mar 28 17:50 .
drwxr-xr-x 4 root root    0 Mar 28 17:35 ..
-r--r--r-- 1 root root 4096 Mar 28 17:51 address
-r--r--r-- 1 root root 4096 Mar 28 17:51 addr_len
-r--r--r-- 1 root root 4096 Mar 28 17:52 broadcast
-r--r--r-- 1 root root 4096 Mar 28 17:52 carrier
-r--r--r-- 1 root root 4096 Mar 28 17:52 dev_id
-r--r--r-- 1 root root 4096 Mar 28 17:50 dormant
-r--r--r-- 1 root root 4096 Mar 28 17:52 duplex
-r--r--r-- 1 root root 4096 Mar 28 17:52 features
-rw-r--r-- 1 root root 4096 Mar 28 17:52 flags
-r--r--r-- 1 root root 4096 Mar 28 17:52 group
-rw-r--r-- 1 root root 4096 Mar 28 17:52 ifalias
-r--r--r-- 1 root root 4096 Mar 28 17:52 ifindex
-r--r--r-- 1 root root 4096 Mar 28 17:52 iflink
-r--r--r-- 1 root root 4096 Mar 28 17:52 link_mode
-rw-r--r-- 1 root root 4096 Mar 28 17:52 mtu
-r--r--r-- 1 root root 4096 Mar 28 17:52 operstate
-r--r--r-- 1 root root 4096 Mar 28 17:52 owner
drwxr-xr-x 2 root root    0 Mar 28 17:52 power
-r--r--r-- 1 root root 4096 Mar 28 17:52 speed
drwxr-xr-x 2 root root    0 Mar 28 17:50 statistics
lrwxrwxrwx 1 root root    0 Mar 28 17:50 subsystem -> ../../../../class/net
-r--r--r-- 1 root root 4096 Mar 28 17:52 tun_flags
-rw-r--r-- 1 root root 4096 Mar 28 17:52 tx_queue_len
-r--r--r-- 1 root root 4096 Mar 28 17:52 type
-rw-r--r-- 1 root root 4096 Mar 28 17:50 uevent
# cat *
0

1
0x0
0
full
0x0
0x1091
-1
4
4
0
1500
unknown
-1
10
0x3001
100
65534
INTERFACE=tun0
IFINDEX=4
Note: /sys/class/net/tun0/address is always empty (so you can align the output with the files)
== [url=http://www.catb.org/esr/faqs/smart-questions.html]Here is how to solve your[/url] [url=https://www.chiark.greenend.org.uk/~sgtatham/bugs.html]Linux problems fast[/url] ==

User avatar
ASRI éducation
Posts: 3197
Joined: Sat 09 May 2009, 12:10
Location: France
Contact:

#237 Post by ASRI éducation »

I do not get involved in this conversation, because I am not competent in the field of network.
However, I regularly read and I hope you will succeed in us a "Top Frisbee." For now, it seems to me that Frisbee is the best solution "network" for novices.
Sincerely,

User avatar
Sky Aisling
Posts: 1368
Joined: Sat 27 Jun 2009, 23:02
Location: Port Townsend, WA. USA

Frisbee Network Manager - Beta 2

#238 Post by Sky Aisling »

Hi,
I'm struggling to get Frisbee to work on Puppy-431.
Will post details after I know ...

Is http://puppylinux.org/wikka/Frisbee the correct place to download the latest Frisbee?
Install Puppy Lucid (5.2x) and Puppy 4
1. Install the following Software Package(s):
Frisbee-beta-2.pet Frisbee-beta-2.pet
wpa_supplicant-0.6.9-i486.4X.pet wpa_supplicant-0.6.9-i486.4X.pet
2. Install the correct dhcpcd package for the Puppy version:

* dhcpcd-5.2.9-Puppy5.pet {Puppy 5}
* dhcpcd-5.2.9-i486.pet {Puppy 4, Fluppy, Puppeee}
Thank you

User avatar
Sky Aisling
Posts: 1368
Joined: Sat 27 Jun 2009, 23:02
Location: Port Townsend, WA. USA

Frisbee Network Manager - Beta 2

#239 Post by Sky Aisling »

Hello again, :)

Question: Is this an appropriate thread to give you feedback on Frisbee?
If so, then here is what I am finding with this situation:

Machine:
Toshiba Satellite P10-s429 (with old bios)
512 mg RAM, screen resolution 1280x800
OS:
Puppy-Linux 431 frugally installed on HDD (using Shinobar's frugal install)
Seamonkey 2.16.2 loaded as an .sfs on HDD (using Shinobar's .sfs downloads)
Wifi Modem:
Home - close to machine

Download source for Frisbee taken 3/31/13 from:
http://puppylinux.org/wikka/Frisbee

At start up Frisbee shows on JWM panel as unconnected.
*Sound* icon on JWM panel not shown until *restart X server* is activated.
*Connect* desktop icon initiates Frisbee scan of networks.
Highlighting a network of choice then choosing *wireless network reset* activates Frisbee (most the time).
Frisbee maintains connection after closing and reopening browser within a session.
No *pup bubbles* showing progress of Frisbee show at any time.

How to make Frisbee automatically search and connect on startup?
How to fix the missing JWM sound icon?


Thank you so much for Frisbee.
Frisbee works without issue on my setups of 528.005 and SalukiNOP. I give these machines that I setup to people who are new to using Linux and need the machine to auto connect to the wifi without hassle. Frisbee works well for this purpose. However, when I encounter an older machine, I find Puppy 431 is the best choice of Puppys. So, I'm hoping Frisbee will work smooth on Puppy 431.

User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

#240 Post by rerwin »

mavrothal, peebee & frisbee-1.0 users,
I have discovered that the "dropwait" feature added to woof and precise pup is not being used, due to an oversight in the script that detects the feature. That might explain why mavrothal was seeing dropped connections, since the feature is intended to keep weak connections more stable. I updated the frisbee kit in the usual place:
http://www.murga-linux.com/puppy/viewto ... 212#693212
Sky Aisling wrote:Question: Is this an appropriate thread to give you feedback on Frisbee?
Yes, it is. But for the "beta" versions of Frisbee, we rely on peebee and pemasu for assistance. My focus is on the "integrated" and upgraded version called frisbee-1.0 for puppies built with the current levels of "woof". My intent is to back-port it to lucid pup 528 and (if anyone cares) to 4.3.1 (although that might be impossible due to dependencies).
Richard

Post Reply