Intel 5300 disconnect problems after upgrading 11.3

English Network/Internet
#1

I use to have 11.2 on my Dell Studio XPS 1645 and the wireless worked fine. I had the hard drive fail so after I had the drive replaced I installed 11.3 and now my wireless doesn’t work in linux. I’m running openSUSE 11.3 with KDE 4.5 (I had this same problem with 4.4).

Network Manager will connect to my router and work fine at first, but somewhere between 1 and 30 minutes it’ll stop working. The network manager still shows that it’s connected, but it doesn’t work, web pages won’t load, I can’t connect to anything, ping can’t ping anything on my local network anymore (it says, “Destination Host Unreachable” when trying to ping the ip address of my wireless router), etc.

Here is the “lspci -nnk” output:


05:00.0 Network controller [0280]: Intel Corporation Ultimate N WiFi Link 5300 [8086:4235]
    Subsystem: Intel Corporation Device [8086:1121]
    Kernel driver in use: iwlagn

Here is the relevant part of my the dmesg output. It shows the time from when my wireless connected to when it stopped working:


   46.253359] ADDRCONF(NETDEV_UP): wlan0: link is not ready
   46.267942]   alloc irq_desc for 40 on node -1
   46.267947]   alloc kstat_irqs on node -1
   46.267969] tg3 0000:0b:00.0: irq 40 for MSI/MSI-X
   46.375280] ADDRCONF(NETDEV_UP): eth0: link is not ready
   46.437655] NET: Registered protocol family 17
   58.138882] lo: Disabled Privacy Extensions
   78.498170] bootsplash: status on console 0 changed to on
  112.650073] wlan0: authenticate with 00:1c:f0:f5:cb:ba (try 1)
  112.652498] wlan0: authenticated
  112.652523] wlan0: associate with 00:1c:f0:f5:cb:ba (try 1)
  112.656367] wlan0: RX AssocResp from 00:1c:f0:f5:cb:ba (capab=0x431 status=0 aid=3)
  112.656371] wlan0: associated
  112.663954] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
  116.327498] martian source 255.255.255.255 from 192.168.24.253, on dev wlan0
  116.327504] ll header: ff:ff:ff:ff:ff:ff:00:1c:f0:f5:cb:ba:08:00
  121.237825] martian source 255.255.255.255 from 192.168.24.253, on dev wlan0
  121.237831] ll header: ff:ff:ff:ff:ff:ff:00:1c:f0:f5:cb:ba:08:00
  123.627084] wlan0: no IPv6 routers present
  188.553168] martian source 255.255.255.255 from 192.168.24.253, on dev wlan0
  188.553176] ll header: ff:ff:ff:ff:ff:ff:00:1c:f0:f5:cb:ba:08:00
  188.656506] martian source 255.255.255.255 from 192.168.24.253, on dev wlan0
  188.656513] ll header: ff:ff:ff:ff:ff:ff:00:1c:f0:f5:cb:ba:08:00
  207.830531] iwlagn 0000:05:00.0: iwl_tx_agg_start on ra = 00:1c:f0:f5:cb:ba tid = 0
  452.977208] SFW2-OUT-ERROR IN= OUT=wlan0 SRC=192.168.24.100 DST=66.102.7.83 LEN=52 TOS=0x00 PREC=0x00 TTL=64 ID=41457 DF PROTO=TCP SPT=38880 DPT=443 WINDOW=501 RES=0x00 ACK FIN URGP=0 OPT (0101080A000258437FA089E1)

Does anyone know of a way I can fix this?

Thanks,
Darren

#2

I’m not a wireless Guru, so I can’t help solve your problem, but given the driver for the Intel 5300 comes with the kernel, could you also advise as to what kernel version you are using? ie provide output of:

uname -a

There is an ‘lspci’ command that is helpful, but I can’t recall exact arguments. … … it may be something like this, … and it may be different (sorry - bad memory on my part):

su -c 'lspci -vnk'

For example, on my Dell Studio 1537 with an Intel 5300AGN wireless (similar/same as yours) on 11.3 my wireless is working. But my kernel may be different. I have:

su -c 'lspci -vnk'
..........
04:00.0 0280: 8086:4235
        Subsystem: 8086:1121
        Flags: bus master, fast devsel, latency 0, IRQ 31
        Memory at f8000000 (64-bit, non-prefetchable) [size=8]
        Capabilities: [c8] Power Management version 3
        Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [e0] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number 00-16-ea-ff-ff-ed-80-76
        Kernel driver in use: iwlagn

and I have:

uname -a
Linux dell.darmstadt 2.6.34-12-default #1 SMP 2010-06-29 02:39:08 +0200 x86_64 x86_64 x86_64 GNU/Linux

Maybe that extra information, if you provide it, may help others assist you ?

In addition to my knowing nothing about wireless, I’m also on vacation in Asia 1/3 of the world away from home, with limited to no internet access, so my ability to monitor this thread is severely limited currently.[/size]

#3

I’m bad … your " lspci -nnk " is likely the command I could not remember. Here is what I get for same command:

04:00.0 Network controller [0280]: Intel Corporation Ultimate N WiFi Link 5300 [8086:4235]
        Subsystem: Intel Corporation Device [8086:1121]
        Kernel driver in use: iwlagn

where my wireless works. i.e. I think we have the same wireless hardware.

But I have not updated my kernel to the latest version, and I have seen SuSE-GmbH kernel updates break this driver in the past (that was the case for 11.2 where an update broke the driver).

#4

I forget to mention, I’m using a wireless connection with this laptop to type this post. … I have had the connection up for hours with no noticeable disconnect. … Hence my suspicion that this could be a kernel update issue (?) , or a configuration issue.

#5

Thanks for your help. Here is the output from uname -a:

Linux XPS16 2.6.34.7-0.2-desktop #1 SMP PREEMPT 2010-09-14 14:21:06 +0200 x86_64 x86_64 x86_64 GNU/Linux

I downgraded the kernel to 2.6.34-12-dekstop (the version that ships on the 11.3 dvd) and that appears to have fixed the problem. I guess I will just have to hope that the intel driver gets fixed at some point.

Thanks,
Darren

#6

It looks like I was a bit premature with my post. After installing the older kernel I still have the same problem.

-Darren

#7

This does not seem to be specific to a particular driver or chipset. These other posts ( including mine) indicate the same issue-

What’s wrong with wireless in OpenSUSE?
wireless issues in the kernel (?!) in opensuse 11.3

And there are others, if you search. It also does not seem to be specific to opensuse but kernels after 2.6.34 ( incl) appear to have this problem.

#8

Neither of those URLs point to the same hardware.

I appreciate the effort, but just blindly post URLs of users with completely different hardware is not IMHO helpful. I can point to dozens of posts of users with the same audio hardware as my pc who can’t get basic sound working from xine, but that does not mean there is a problem with xine. Rather the users have misconfigured the software completely unintentionally.

I supported the milestone release phase of openSUSE-11.3, testing with the Intel 5300 wireless for practically every release with the identical hardware that user Nobled has, and it worked with every version that I tested. Every version! That leads me to believe that if user Nobled as the same hardware as I have on my laptop, and every indication is his wireless hardware is IDENTICAL then this is a configuration issue on his PC (given he now has the same kernel version which means the same driver). Its possible there is other support software in which he has installed a broken version ? Maybe iw or libiw30 ? I confess this speculation is probably not helpful, and we need a real Guru of the wireless to step in here and help.

As an aside, I note on my PC (with IDENTICAL hardware where wireless ‘just works’ in 11.3):

oldcpu@dell:~> rpm -qa '*iw*'
iw-0.9.18-3.1.x86_64
libiw30-30.pre8-11.2.x86_64
#9

Any chance the problem is the router? or some hacker launching a denial of service on your wireless connection to crack into the wpa or wep connection you have setup.

#10

This laptop use to have 11.2 on it and it would stay connected for days. It also has Windows 7 on it and the wireless works fine in windows. I also have 2 other laptops and my android phone using this wireless router without issue (I’m posting this from my phone on the router right now). My laptop also has the same disconnect problem on my work’s wireless network. I will check the versions of iw and libiw as soon as I can.

Thanks,
Darren

#11

I checked the version of iw and libiw and they appear to be the same as yours:

<darren@XPS16:~> rpm -qa '*iw*'                                                                                                                                                                                                                                         
libiw30-30.pre8-11.2.x86_64
iw-0.9.18-3.1.x86_64

Does anyone else have more ideas?

Thanks,
Darren

#12

Not an identical card but I have the intel5100 card so from the same family. My card’s data is as follows:

05:00.0 Network controller [0280]: Intel Corporation WiFi Link 5100 [8086:4232]
        Subsystem: Intel Corporation WiFi Link 5100 AGN [8086:1201]
        Kernel driver in use: iwlagn


05:00.0 0280: 8086:4232
        Subsystem: 8086:1201
        Flags: bus master, fast devsel, latency 0, IRQ 30
        Memory at febfe000 (64-bit, non-prefetchable) [size=8]
        Capabilities: [c8] Power Management version 3
        Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [e0] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number 00-26-c6-ff-ff-cf-44-ca
        Kernel driver in use: iwlagn


mediacentre@mediacentre:~> rpm -qa '*iw*'
libiw30-30.pre8-11.2.x86_64
iw-0.9.18-3.1.x86_64

As you can see it’s virtually identical to the 5300 card.

Does your card connect immediately after boot up or is there a delay? I ask because with my card there seems to be a delay before it actually connects and after it connects it will drop the connection about 50-60 seconds afterwards. Looking at the dmesg logs shows that it disconnects a few times because of reason 6 initially

   32.788652] eth0: no link during initialization.
   32.789762] ADDRCONF(NETDEV_UP): eth0: link is not ready
   32.899333] ADDRCONF(NETDEV_UP): wlan0: link is not ready
   36.271387] wlan0: authenticate with 00:14:85:ce:98:1e (try 1)
   36.273814] wlan0: authenticated
   36.273917] wlan0: associate with 00:14:85:ce:98:1e (try 1)
   36.276740] wlan0: RX AssocResp from 00:14:85:ce:98:1e (capab=0x431 status=0 aid=1)
   36.276752] wlan0: associated
   36.283920] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
   43.451123] wlan0: deauthenticated from 00:14:85:ce:98:1e (Reason: 6)
   43.455667] cfg80211: Calling CRDA to update world regulatory domain
   43.467327] cfg80211: World regulatory domain updated:
   43.467337]     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
   43.467347]     (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   43.467355]     (2457000 KHz - 2482000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
   43.467363]     (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
   43.467371]     (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   43.467379]     (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   46.681905] wlan0: authenticate with 00:14:85:ce:98:1e (try 1)
   46.684200] wlan0: authenticated
   46.684293] wlan0: associate with 00:14:85:ce:98:1e (try 1)
   46.687130] wlan0: RX AssocResp from 00:14:85:ce:98:1e (capab=0x431 status=0 aid=1)
   46.687141] wlan0: associated
   53.838106] wlan0: deauthenticated from 00:14:85:ce:98:1e (Reason: 6)
   53.842954] cfg80211: Calling CRDA to update world regulatory domain
   53.863453] cfg80211: World regulatory domain updated:
   53.863466]     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
   53.863475]     (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   53.863484]     (2457000 KHz - 2482000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
   53.863492]     (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
   53.863499]     (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   53.863507]     (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
   57.121402] wlan0: authenticate with 00:14:85:ce:98:1e (try 1)
   57.123730] wlan0: authenticated
   57.123816] wlan0: associate with 00:14:85:ce:98:1e (try 1)
   57.126629] wlan0: RX AssocResp from 00:14:85:ce:98:1e (capab=0x431 status=0 aid=1)
   57.126639] wlan0: associated
   59.075278] Intel AES-NI instructions are not detected.
   59.223989] padlock: VIA PadLock not detected.
   59.395818] RPC: Registered udp transport module.
   59.395825] RPC: Registered tcp transport module.
   59.395830] RPC: Registered tcp NFSv4.1 backchannel transport module.
   59.852035] Slow work thread pool: Starting up
   59.852894] Slow work thread pool: Ready

Then about 60 seconds later the wifi deauhenticates with reason 3.

  121.585484] BIOS EDD facility v0.16 2004-Jun-25, 1 devices found
  123.519085] mac80211-phy0: failed to remove key (0, 00:14:85:ce:98:1e) from hardware (-22)
  123.527266] wlan0: deauthenticating from 00:14:85:ce:98:1e by local choice (reason=3)
  123.564068] cfg80211: Calling CRDA to update world regulatory domain
  123.583158] cfg80211: World regulatory domain updated:
  123.583170]     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
  123.583179]     (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  123.583187]     (2457000 KHz - 2482000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
  123.583194]     (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
  123.583202]     (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  123.583210]     (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  123.815771] ADDRCONF(NETDEV_UP): wlan0: link is not ready
  127.363328] bootsplash: status on console 0 changed to on
  132.683235] wlan0: authenticate with 00:14:85:ce:98:1e (try 1)
  132.685584] wlan0: authenticated
  132.685656] wlan0: associate with 00:14:85:ce:98:1e (try 1)
  132.688370] wlan0: RX AssocResp from 00:14:85:ce:98:1e (capab=0x431 status=0 aid=1)
  132.688378] wlan0: associated
  132.690918] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
  132.800780] eth0: no link during initialization.
  132.801960] ADDRCONF(NETDEV_UP): eth0: link is not ready
  132.815097] wlan0: deauthenticating from 00:14:85:ce:98:1e by local choice (reason=3)
  132.850757] cfg80211: Calling CRDA to update world regulatory domain
  132.863428] cfg80211: World regulatory domain updated:
  132.863439]     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
  132.863447]     (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  132.863455]     (2457000 KHz - 2482000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
  132.863462]     (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
  132.863469]     (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  132.863477]     (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
  132.914201] ADDRCONF(NETDEV_UP): wlan0: link is not ready
  133.017080] ADDRCONF(NETDEV_UP): wlan0: link is not ready
  133.110862] wlan0: authenticate with 00:14:85:ce:98:1e (try 1)
  133.113307] wlan0: authenticated
  133.113418] wlan0: associate with 00:14:85:ce:98:1e (try 1)
  133.116213] wlan0: RX AssocResp from 00:14:85:ce:98:1e (capab=0x431 status=0 aid=1)
  133.116224] wlan0: associated
  133.119715] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready

I have no idea why this is happening and don’t see it happening on my laptops using intel abg3945 cards.

I then seem to experience random signal drops there after but I haven’t been able to pin point a pattern yet.

Does your dmesg show anything similar with the deauthentications?[/size]

#13

Sorry that I can not be of help. Here is what I have with a functional Intel 5300 (from dmesg):

    5.873427] iwlagn: Intel(R) Wireless WiFi Link AGN driver for Linux, in-tree:d
    5.873430] iwlagn: Copyright(c) 2003-2010 Intel Corporation
    5.873494] iwlagn 0000:04:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
    5.873502] iwlagn 0000:04:00.0: setting latency timer to 64
    5.874479] iwlagn 0000:04:00.0: Detected Intel Wireless WiFi Link 5300AGN REV=0x24
..............
    5.896458] iwlagn 0000:04:00.0: Tunable channels: 13 802.11bg, 24 802.11a channels
..............
    5.896458] iwlagn 0000:04:00.0: Tunable channels: 13 802.11bg, 24 802.11a channels
    5.896543]   alloc irq_desc for 31 on node -1
    5.896545]   alloc kstat_irqs on node -1
    5.896585] iwlagn 0000:04:00.0: irq 31 for MSI/MSI-X
    5.982409] iwlagn 0000:04:00.0: firmware: requesting iwlwifi-5000-2.ucode
    5.984444] iwlagn 0000:04:00.0: loaded firmware version 8.24.2.12
................
    6.063389] phy0: Selected rate control algorithm 'iwl-agn-rs'
................
   16.158525] ADDRCONF(NETDEV_UP): wlan0: link is not ready
   16.171671]   alloc irq_desc for 34 on node -1
   16.171678]   alloc kstat_irqs on node -1
   16.171712] tg3 0000:08:00.0: irq 34 for MSI/MSI-X
   16.274199] ADDRCONF(NETDEV_UP): eth0: link is not ready
   16.370539] NET: Registered protocol family 17
..................
   53.706613] wlan0: authenticate with 00:19:cb:85:2e:4b (try 1)
   53.708319] wlan0: authenticated
   53.708371] wlan0: associate with 00:19:cb:85:2e:4b (try 1)
   53.712967] wlan0: RX AssocResp from 00:19:cb:85:2e:4b (capab=0x421 status=0 aid=3)
   53.712976] wlan0: associated
   53.717628] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
   53.717744] cfg80211: Calling CRDA for country: AT
   53.725851] cfg80211: Current regulatory domain updated by AP to: AT
   53.725854]     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
   53.725856]     (2402000 KHz - 2482000 KHz @ 40000 KHz), (N/A, 2000 mBm)
  784.489088] CE: hpet increased min_delta_ns to 7500 nsec

I may have missed some lines , … I have a slow dial up, and the wife is rushing me to close this post and give her a hand … so must run …

#14

actually, blindly replying to posts without reading them does not help either.

My POINT was that this is happening for different hardware - not JUST intel. That might be a clue to someone.
The 2 other links are essentially the SAME issue - you do not recongnize it. I have read these posts and recongnize them to be the same issue. There is a problem in the kernel code, PERIOD.

#15

clearly, you never read the post before you started to reply - that seems to be a common problem on forums, leading to too much noise and chatter of little consequence.

#16

clearly, you never read the original post or the links I posted before you started to reply.

#17

I think this problem isn’t unique to Linux. INTEL 5100/5300 Network Dropouts/Disconnects - lenovo community
People seem to be effected on XP as well. They were able to eliminate most of the problem by installing the latest Intel driver.

#18

Well, for me this problem is unique to openSUSE 11.3. With 11.2 and Windows 7 my wifi on this same machine can stay connected for days at a time, after upgrading to 11.3 I’m lucky to have it stay connected for more then a minute or 2. So obviously there is something wrong / different with 11.3.

#19

I believe that link is out of date and not applicable to the kernel in openSUSE-11.3. The kernel in 11.3 has code that is supposed to have improved the Intel 5100/5300 connection in 11.3 over 11.2. In fact it has done so on my PC as I noted above. I’m not a network Guru, so I can not give you specific advice on how to troubleshoot your configuration, … but I can say that it IS possible to have a functional connection in 11.3 without continual noticeable dropouts (because I have such a configuration). Until a network guru (of which I am not one) steps in and helps debug your configuration, there is nothing more I can add to this thread other than what I have already noted. For me, 11.3 has no such problem, and it works well.

… but I do note my Intel 5300 with 11.2 only worked with the original 11.2 kernel and was broken and did NOT work with any of the newer 11.2 kernels. I wrote a bug report on the Intel 5300 on 11.2.

#20

Here are two bug reports to look at: