Opened 3 months ago

Last modified 2 months ago

#18800 assigned bug

DHCP with Telekom Speedport 4 not working

Reported by: marcusathome Owned by: nobody
Priority: normal Milestone: Unscheduled
Component: Drivers/Network/iprowifi3945 Version: R1/beta4
Keywords: Cc:
Blocked By: Blocking:
Platform: All

Description

Connecting my laptop Lenovo ThinkPad R61i to the Network Router Telekom Speedport 4 thru WIFI works, I can see data is constantly being transmitted. It does not, however, get an IP address and everything else assigned from the router. The NetworkStatus forever shows “Configuring…” I was partly successful by entering the IP address, netmask, gateway and DNS-Servers (as found on the router for that machine) manually and can ping other local machines on the network, but not getting onto the Internet. Syslog shows that these is negotiation taking place between router and local machine but times out and then fails (syslog attached). DHCP assignment on that machine works fine out of Linux.

Attachments (1)

syslog (46.1 KB ) - added by marcusathome 3 months ago.

Download all attachments as: .zip

Change History (13)

by marcusathome, 3 months ago

Attachment: syslog added

comment:1 by waddlesplash, 3 months ago

Component: Network & Internet/WirelessDrivers/Network/iprowifi3945
Owner: changed from mmlr to nobody
Status: newassigned

comment:2 by waddlesplash, 3 months ago

Does ethernet work?

comment:3 by marcusathome, 3 months ago

I don't know. Have to find a long enough cable. I can try maybe tomorrow. Will let you know then.

comment:4 by marcusathome, 3 months ago

Test with ethernet connection is positive, all is working as expected. The router assigns IP address and anything else that is needed. Next test my side, I'm gonna manually configure the wireless network with same values, except for the IP address, and see if this works.

comment:5 by marcusathome, 3 months ago

Partly success! Using the values from the ethernet, except for the wireless adapter (which has a fixed IP assigned on the router and this is what I used here), it now connects to the Internet and I am able to open any website and also performed a system update. Still, that doesn't solve the reported issue, that is, DHCP is not working. MTU was set to 2294 by the system. Not sure if playing around with this might change anything in the behaviour. Additionally, the network preferences dialog often completely freezes when switching between DHCP and static.

comment:6 by pulkomandy, 3 months ago

The DHCP server is receiving the request, but does not accept it.

DAEMON 'DHCP': /dev/net/iprowifi3945/0: Received DHCP_NACK from 192.168.2.1

From the attached syslog, it's hard to say exactly why, as the start of the syslog is missing (because of too many logs from the wifi).

At the wifi level there are a lot of these in the syslog:

KERN: wlan_control: 9235, 15
KERN: wlan_control: 9235, 76

Also there seem to be a firmware error (twice in the log):

142	KERN: [iprowifi3945] (wpi) fatal firmware error
143	KERN: firmware error log (count = 1):
144	KERN:   error type = "SYSASSERT" (0x00000005)
145	KERN:   error data      = 0x0000010B
146	KERN:   branch link     = 0x000008B60000035E
147	KERN:   interrupt link  = 0x0000032000000000
148	KERN:   time            = 2589921694
149	KERN: driver status:
150	KERN:   tx ring  0: qid=0  cur=33  queued=0
151	KERN:   tx ring  1: qid=1  cur=0   queued=0
152	KERN:   tx ring  2: qid=2  cur=0   queued=0
153	KERN:   tx ring  3: qid=3  cur=4   queued=0
154	KERN:   tx ring  4: qid=4  cur=48  queued=0
155	KERN:   rx ring: cur=34
156	KERN: [net/iprowifi3945/0] stop running, 1 vaps running
157	KERN: [net/iprowifi3945/0] ieee80211_new_state_locked: RUN -> INIT (nrunning 0 nscanning 0)
158	KERN: ieee80211_notify_ifnet_change not implemented, yet.
159	KERN: [net/iprowifi3945/0] down parent
160	KERN: ieee80211_notify_scan_done
161	KERN: [net/iprowifi3945/0] ieee80211_newstate_cb: RUN -> INIT arg -1
162	KERN: [net/iprowifi3945/0] sta_newstate: RUN -> INIT (-1)
163	KERN: [net/iprowifi3945/0] ieee80211_ref_node (ieee80211_send_mgmt:2647) 0xffffffff805d8000<50:6f:0c:de:4f:0f> refcnt 4
164	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] send station disassociate (reason: 8 (sending STA is leaving/has left BSS))
165	KERN: ieee80211_notify_node_leave
166	KERN: /dev/net/iprowifi3945/0: link down, media 0x10080 quality 1000 speed 0
167	KERN: [net/iprowifi3945/0] node_reclaim: remove 0xffffffff805d8000<50:6f:0c:de:4f:0f> from station table, refcnt 3
168	KERN: [net/iprowifi3945/0] ieee80211_alloc_node 0xffffffff80658000<00:1c:bf:85:c2:52> in station table
169	KERN: [net/iprowifi3945/0] [00:1c:bf:85:c2:52] ieee80211_alloc_node: inact_reload 2
170	KERN: ieee80211_notify_radio not implemented, yet.
171	KERN: [net/iprowifi3945/0] start running, 0 vaps running
172	KERN: ieee80211_notify_ifnet_change not implemented, yet.
173	KERN: [net/iprowifi3945/0] ieee80211_start_locked: up parent
174	KERN: ieee80211_notify_radio not implemented, yet.
175	KERN: [net/iprowifi3945/0] start running, 1 vaps running
176	KERN: wlan_control: 9235, 76
177	KERN: wlan_control: 9234, 20
178	KERN: Last message repeated 2 times.
179	KERN: wlan_control: 9234, 16
180	KERN: wlan_control: 9234, 17
181	KERN: wlan_control: 9234, 26
182	KERN: wlan_control: 9234, 103
183	KERN: [net/iprowifi3945/0] ieee80211_new_state_locked: INIT -> SCAN (nrunning 0 nscanning 0)
184	KERN: [net/iprowifi3945/0] ieee80211_newstate_cb: INIT -> SCAN arg 0
185	KERN: [net/iprowifi3945/0] sta_newstate: INIT -> SCAN (0)
186	KERN: ieee80211_notify_scan_done
187	KERN: wlan_control: 9235, 76
188	KERN: wlan_control: 9234, 18
189	KERN: wlan_control: 9234, 7
190	KERN: wlan_control: 9234, 95
191	KERN: wlan_control: 9234, 17
192	KERN: wlan_control: 9234, 26
193	KERN: wlan_control: 9234, 21
194	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] station assoc via MLME
195	KERN: [net/iprowifi3945/0] ieee80211_alloc_node 0xffffffff805d8000<50:6f:0c:de:4f:0f> in station table
196	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] ieee80211_alloc_node: inact_reload 2
197	KERN: [net/iprowifi3945/0] node_reclaim: remove 0xffffffff80658000<00:1c:bf:85:c2:52> from station table, refcnt 0
198	KERN: [net/iprowifi3945/0] set WME_AC_BE (chan) [acm 0 aifsn 3 logcwmin 4 logcwmax 10 txop 0]
199	KERN: [net/iprowifi3945/0] set WME_AC_BE (bss ) [acm 0 aifsn 3 logcwmin 4 logcwmax 10 txop 0]
200	KERN: [net/iprowifi3945/0] set WME_AC_BK (chan) [acm 0 aifsn 7 logcwmin 4 logcwmax 10 txop 0]
201	KERN: [net/iprowifi3945/0] set WME_AC_BK (bss ) [acm 0 aifsn 7 logcwmin 4 logcwmax 10 txop 0]
202	KERN: [net/iprowifi3945/0] set WME_AC_VI (chan) [acm 0 aifsn 2 logcwmin 3 logcwmax 4 txop 94]
203	KERN: [net/iprowifi3945/0] set WME_AC_VI (bss ) [acm 0 aifsn 2 logcwmin 3 logcwmax 4 txop 94]
204	KERN: [net/iprowifi3945/0] set WME_AC_VO (chan) [acm 0 aifsn 2 logcwmin 2 logcwmax 3 txop 47]
205	KERN: [net/iprowifi3945/0] set WME_AC_VO (bss ) [acm 0 aifsn 2 logcwmin 2 logcwmax 3 txop 47]
206	KERN: [net/iprowifi3945/0] update WME_AC_BE (chan+bss) [acm 0 aifsn 2 logcwmin 4 logcwmax 10 txop 0]
207	KERN: [net/iprowifi3945/0] ieee80211_wme_updateparams_locked: WME params updated, cap_info 0xfe
208	KERN: [net/iprowifi3945/0] ieee80211_new_state_locked: SCAN -> AUTH (nrunning 0 nscanning 0)
209	KERN: [net/iprowifi3945/0] ieee80211_newstate_cb: SCAN -> AUTH arg 192
210	KERN: [net/iprowifi3945/0] sta_newstate: SCAN -> AUTH (192)
211	KERN: [net/iprowifi3945/0] ieee80211_ref_node (ieee80211_send_mgmt:2647) 0xffffffff805d8000<50:6f:0c:de:4f:0f> refcnt 3
212	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] recv auth frame with algorithm 0 seq 2
213	KERN: [net/iprowifi3945/0] ieee80211_new_state_locked: AUTH -> ASSOC (nrunning 0 nscanning 0)
214	KERN: [net/iprowifi3945/0] ieee80211_newstate_cb: AUTH -> ASSOC arg 0
215	KERN: [net/iprowifi3945/0] sta_newstate: AUTH -> ASSOC (0)
216	KERN: [net/iprowifi3945/0] ieee80211_ref_node (ieee80211_send_mgmt:2647) 0xffffffff805d8000<50:6f:0c:de:4f:0f> refcnt 3
217	KERN: [net/iprowifi3945/0] ieee80211_parse_wmeparams: WME: 0: acm=0 aifsn=3 logcwmin=4 logcwmax=10 txopLimit=0
218	KERN: [net/iprowifi3945/0] ieee80211_parse_wmeparams: WME: 1: acm=0 aifsn=7 logcwmin=4 logcwmax=10 txopLimit=0
219	KERN: [net/iprowifi3945/0] ieee80211_parse_wmeparams: WME: 2: acm=0 aifsn=2 logcwmin=3 logcwmax=4 txopLimit=94
220	KERN: [net/iprowifi3945/0] ieee80211_parse_wmeparams: WME: 3: acm=0 aifsn=2 logcwmin=2 logcwmax=3 txopLimit=47
221	KERN: [net/iprowifi3945/0] ieee80211_wme_updateparams_locked: WME params updated, cap_info 0x4
222	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] assoc success at aid 9: short preamble, short slot time, QoS
223	KERN: [net/iprowifi3945/0] ieee80211_new_state_locked: ASSOC -> RUN (nrunning 0 nscanning 0)
224	KERN: [net/iprowifi3945/0] ieee80211_newstate_cb: ASSOC -> RUN arg 16
225	KERN: [net/iprowifi3945/0] sta_newstate: ASSOC -> RUN (16)
226	KERN: ieee80211_notify_node_join
227	KERN: [net/iprowifi3945/0] [50:6f:0c:de:4f:0f] ieee80211_node_authorize: inact_reload 20
228	KERN: wlan_control: 9235, 15
229	KERN: wlan_control: 9235, 1
230	KERN: /dev/net/iprowifi3945/0: link up, media 0x81008f quality 1000 speed 36000000
231	KERN: wlan_control: 9235, 1
232	KERN: wlan_control: 9235, 15
233	KERN: wlan_control: 9234, 19

We should probably solve these before we look at the DHCP level of things? Even if it mostly works, the connection seems to be quite unstable?

MTU was set to 2294 by the system

This seems appropriate for WLAN, but maybe you can try forcing it to 1490 which would allow packets to travel between Wifi, Ethernet and other networks without any splitting and reassembly.

comment:7 by waddlesplash, 3 months ago

The firmware errors are also seen in #14260. I think this driver may have used to work better, years ago before FreeBSD refactored things...

comment:8 by marcusathome, 3 months ago

The workaround I implemented is fine for me now (I can use Haiku) but I'm happy to help with further error tracing, where needed. Re the MTU: Is there a config file to set the MTU?

comment:9 by waddlesplash, 3 months ago

To confirm: the workaround is to set a lower MTU? Which one, specifically? (Or do you need to set a static IP *and* the lower MTU? Or just the static IP?)

If the lower MTU fixes DHCP, that seems very odd. Wouldn't the DHCP packets be much smaller than the MTU?

comment:10 by marcusathome, 2 months ago

The MTU is still unchanged. I was just thinking that this is one parameter to play around with and see if it causes any changes. Here is what I did: 1) Disabled the wifi adapter 2) enabled the ethernet adapter, set it to DHCP and connected it. 3) Once the Internet connection was working, write down all the connection parameters. 4) set the router for the MAC address of the wifi adapter to static, to lock to one address (to prevent later address overlap). 5) disable the ethernet adapter, set the wifi to 'Static' and enter IP, Netmask and gateway, check the DNS servers. 6) to make sure reboot. 7) enable the wifi adapter and make sure it is set to 'Static'. Nevertheless, I can see that the wifi gets dropped easily while syncing Nextcloud (although it reconnects automatically), so I now run a permanent ping to a local machine to keep the adapter busy. Even on the pings I can see that the network gets dropped frequently, so I still don't have a 100% reliable connection.

comment:11 by marcusathome, 2 months ago

Syncing Nextcloud, which is quite of a stress test due to the data volume, is not going well. The network constantly gets dropped, re-connects, get dropped a couple of minutes later etc. Shall I get a syslog tail when it does that?

comment:12 by waddlesplash, 2 months ago

You can, but I suspect it's just more firmware panics.

Note: See TracTickets for help on using tickets.