Cacko 1.23 & Wifi Signal Strength Options

[charlesa]

Hi -

I have a permanent red network icon in my taskbar, and popup message telling me that "The signal level is low". The network icon occasionally flashes to blue for a few seconds. This is despite the fact I am sitting next to the WiFi access point. Is this "normal"? I did not see behavior like this in previous cacko releases.

I have an 860 running Cacko 1.23 heavy and a DLink 660 WiFi card. I recently reflashed and just upgraded the DLink firmware to 1.7.4. The previous version of firmware, 1.3.x was having probs connecting with Cacko 1.23, thus the upgrade.


# hostap_diag wlan0
Host AP driver diagnostics information for 'wlan0'

NICID: id=0x800c v1.0.0 (PRISM II (2.5) PCMCIA (SST parallel flash))
PRIID: id=0x0015 v1.1.1
STAID: id=0x001f v1.7.4 (station firmware)
#


I have /etc/pcmcia/hostap_cs.conf as the following:


card "D-Link DWL-660 11Mbps WLAN Card"
version "D-Link", "DCF-660W", ""
# manfid 0xd601, 0x0005
bind "hostap_cs"


I have tried enabling the manfid 0xd601, 0x0005; sometimes it connects, other times it can not, so I have gone back to the default and commented that line out.

Now I can connect to a 802.11 network, but the signal strength is always red in the toolbar, and often complaining the signal is "low".



# iwconfig
lo no wireless extensions.

Warning: Driver for device wifi0 recommend version 18 of Wireless Extension,
but has been compiled with version 16, therefore some driver features
may not be available...

wifi0 IEEE 802.11b ESSID:"KTHNOKIA" Nickname:"zaurus"
Mode:Managed Frequency:2.437 GHz Access Point: 00:15:E9:77:91:70
Bit Rate:11 Mb/s Sensitivity=1/3
Retry min limit:8 RTS thr:off Fragment thr:off
Encryption key:6161-6131-3233-3435-3637-3839-31 Security mode:open
Power Management:off
Link Quality:51 Signal level:0 Noise level:204
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:2400 Missed beacon:0

wlan0 IEEE 802.11b ESSID:"KTHNOKIA" Nickname:"zaurus"
Mode:Managed Frequency:2.437 GHz Access Point: 00:15:E9:77:91:70
Bit Rate:11 Mb/s Sensitivity=1/3
Retry min limit:8 RTS thr:off Fragment thr:off
Encryption key:6161-6131-3233-3435-3637-3839-31 Security mode:open
Power Management:off
Link Quality:51 Signal level:0 Noise level:204
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:2400 Missed beacon:0

#



What is strange is that this is happening even while sitting next to the WiFi access point (can reach out & touch it). I never saw this on previous cacko ROMs (1.22 etc). Other devices seem to be connecting just fine with 100% signal strength.

Questions:

- Are any of you also experiencing this persistant red signal strength indicator and "signal level low" message in Cacko 1.23?
- Does the app "Wireless LAN Status" never show ESSID?
- Is the "Wireless LAN Status" information reliable? It does not always seem to correspond with the iwconfig data. Is iwconfig's information relaible? In fact, iwconfig always says the "Signal level:0" even thou there is a link.
- Any other hints?

Thanks in advance!

Charles

[charlesa]

Well, I tried to downgrade to Cacko 1.22 to eliminate the possibility of something in 1.23 (also had bluetooth issues in 1.23). However, I am still getting this signal low problem, and dropped packets which is making the Z almost unusable.

Dmesg tells me the following:

eth0 (WE) : Buffer for request 8B1B too small (0<8)

Does anyone know to what this refers?

TIA,

C.


BTW, I also found a thread on OESF which states the following:

Ok, once again, the signal too low error is a bug in Qtopia/Sharp which doesn't query the driver for the range of expected signal strength values but instead uses a fixed range, which is... wrong.

ref: http://www.oesf.org/forums/lofiversion/index.php?t3287.html

[charlesa]

OK the weirdness continues. Now I can connect fine for about five minutes (blue icon), then get the red icon before dropping packets and the connection downgrades to be unusable. This is happening on both office and home Wireless Lan networks, virtually like clockwork.

Here is a piece of dmesg:

wifi0: hfa384x_cmd_issue: cmd reg was busy for 5000 usec
wifi0: hfa384x_cmd_issue - timeout - reg=0x8888
wifi0: hfa384x_cmd: entry still in list? (entry=c3b658a0, type=0, res=-1)
wifi0: hfa384x_cmd: interrupted; err=-110
wifi0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd51, len=6)
wifi0: hfa384x_cmd_issue: cmd reg was busy for 5000 usec
wifi0: hfa384x_cmd_issue - timeout - reg=0x8888
wifi0: hfa384x_cmd: entry still in list? (entry=c3b658a0, type=0, res=-1)
wifi0: hfa384x_cmd: interrupted; err=-110
wifi0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd51, len=6)
wifi0: hfa384x_cmd_issue: cmd reg was busy for 5000 usec
wifi0: hfa384x_cmd_issue - timeout - reg=0x8888
wifi0: hfa384x_cmd: entry still in list? (entry=c3b658a0, type=0, res=-1)
wifi0: hfa384x_cmd: interrupted; err=-110
wifi0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd51, len=6)
wifi0: hfa384x_cmd_issue: cmd reg was busy for 5000 usec
wifi0: hfa384x_cmd_issue - timeout - reg=0x8888
wifi0: hfa384x_cmd: entry still in list? (entry=c3b658a0, type=0, res=-1)
wifi0: hfa384x_cmd: interrupted; err=-110
wifi0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd51, len=6)
eth0: hfa384x_setup_bap - timeout before
eth0: hfa384x_set_rid (rid=fc0e, len=34) - failed - res=-110
wifi0: hfa384x_cmd_issue: cmd reg was busy for 5000 usec
wifi0: hfa384x_cmd_issue - timeout - reg=0x8888
wifi0: hfa384x_cmd: entry still in list? (entry=c3b658a0, type=0, res=-1)
wifi0: hfa384x_cmd: interrupted; err=-110
wifi0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc28, len=2)
Could not read current WEP flags.
wifi0: encryption setup failed
eth0: set_encryption failed
hostap_cs: CS_EVENT_PM_RESUME
hostap_cs: CS_EVENT_CARD_RESET
prism2_hw_init: initialized in 190 ms
wifi0: NIC: id=0x800c v1.0.0
wifi0: PRI: id=0x15 v1.1.1
wifi0: STA: id=0x1f v1.7.4
wifi0: TXEXC - status=0x0004 ([Discon]) tx_control=000c
retry_count=0 tx_rate=0 fc=0x4108 (Data::0 ToDS)
A1=00:00:00:00:00:00 A2=00:80:c8:2b:7a:f7 A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00
wifi0: LinkStatus=1 (Connected)
wifi0: LinkStatus: BSSID=00:15:e9:e0:7f:55
eth0 (WE) : Buffer for request 8B1B too small (0<9)



Have you guys seen anything like this before? Could it be the Dlink 660W card is failing?

TIA,

C.

[BarryW]

Might be the card. I'm using Cacko full and have had my connection stay up for a days. Using an Ambicom card though.

[agoodm]

Sounds like something is up with your card.