[Openswan Users] Openswan 2.6.14/15 on kernel 2.6.24.7

Wed Aug 6 09:34:11 EDT 2008

On Tue, 5 Aug 2008, Mario Lobo wrote:

> Sorry for replying to my own post.
>
> Ok Paul. I know you're busy so this won't be long.
>
> As an addition to what's bellow (case my last e-mail got into your stack), I
> tested OSw 2.6.15dr2 with kernels 2.6.25 and 2.6.25.9 today and the same
> error came up with both.
>
> The log bellow is from a  2.6.24.7 kernel. Kernel 2.6.24 was also tested with
> the same results.

We are releasing a new version later today, hopefully that will fix most people's
compile/load problems.

Paul

> Thanks,
> -- 
> Mario Lobo
> http://www.mallavoodoo.com.br
>
>
> On Monday 04 August 2008 15:30:01 Mario Lobo wrote:
>> On Monday 04 August 2008, Mario Lobo wrote:
>>> Hi;
>>>
>>> I am trying to make this work with KLIPS (which I had working perfectly
>>> in kernel 2.6.14 + Oswan 2.4.8) but I can't get past the loading of
>>> ipsec.ko.
>>>
>>> It compiles with no errors, either through the kernel compilation or
>>> through the make module command in openswan source.
>>>
>>> patches are applied with no failures.
>>>
>>> The problem is that if I load the module, the kernel panics.
>>>
>>> if I compile with the (*) option instead of (M), the kernel panics at
>>> boot.
>>>
>>> Any suggestions on what I should do to fix this?
>>>
>>> Thanks
>>
>> Here is what I get right after a "modprobe ipsec". The machine keeps
>> working but the network goes crazy and stops. A reboot is needed.
>>
>> ===========
>> Aug  4 14:30:18 Spyket-embratel kernel: klips_info:ipsec_init: KLIPS
>> startup, Openswan KLIPS IPsec stack version: 2.6.15dr2
>> Aug  4 14:30:18 Spyket-embratel kernel: NET: Registered protocol family 15
>> Aug  4 14:30:18 Spyket-embratel kernel: ------------[ cut here
>> ]------------ Aug  4 14:30:18 Spyket-embratel kernel: kernel BUG at
>> net/core/dev.c:3609! Aug  4 14:30:18 Spyket-embratel kernel: invalid
>> opcode: 0000 [#1]
>> Aug  4 14:30:18 Spyket-embratel kernel: Modules linked in: ipsec sunrpc
>> ipt_REJECT ipt_LOG xt_limit xt_state xt_tcpudp iptable_filter
>> iptable_mangle iptable_nat ip_tables x_tables nf_nat_ftp nf_nat
>> nf_conntrack_ftp
>> nf_conntrack_ipv4 nf_conntrack uhci_hcd usbcore i2c_i801 i2c_core rng_core
>> intel_agp agpgart 8139too tg3 3c59x floppy
>> Aug  4 14:30:18 Spyket-embratel kernel:
>> Aug  4 14:30:18 Spyket-embratel kernel: Pid: 5533, comm: modprobe Not
>> tainted (2.6.24.7 #1)
>> Aug  4 14:30:18 Spyket-embratel kernel: EIP: 0060:[<c054a772>] EFLAGS:
>> 00010246 CPU: 0
>> Aug  4 14:30:18 Spyket-embratel kernel: EIP is at
>> register_netdevice+0x1e2/0x2f0
>> Aug  4 14:30:18 Spyket-embratel kernel: EAX: 00000000 EBX: f7f3ec00 ECX:
>> 00000000 EDX: 00000000
>> Aug  4 14:30:18 Spyket-embratel kernel: ESI: f7f3ec00 EDI: f7f3ef50 EBP:
>> 00000000 ESP: f6641e1c
>> Aug  4 14:30:18 Spyket-embratel kernel:  DS: 007b ES: 007b FS: 0000 GS:
>> 0033 SS: 0068
>> Aug  4 14:30:18 Spyket-embratel kernel: Process modprobe (pid: 5533,
>> ti=f6640000 task=f7f3b550 task.ti=f6640000)
>> Aug  4 14:30:18 Spyket-embratel kernel: Stack: ffffffff ffffffff f7f3ec10
>> 0000000a 00000010 f7f3ec00 f8fd642b f7f3ec00
>> Aug  4 14:30:18 Spyket-embratel kernel:        f7f3ec00 f7f3ef50 00000000
>> c054a8b7 f7f3ec00 00000000 f8f9b80a f7f3ec00
>> Aug  4 14:30:18 Spyket-embratel kernel:        00000010 f8fd6425 00000000
>> f8f95326 00000000 0805a0e0 00000000 f6640000
>> Aug  4 14:30:18 Spyket-embratel kernel: Call Trace:
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<c054a8b7>]
>> register_netdev+0x37/0x50
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<f8f9b80a>]
>> ipsec_mast_createnum+0x8a/0xe0 [ipsec]
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<f8f95326>]
>> ipsec_tunnel_init_devices+0x16/0x50 [ipsec]
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<f8f9b9a7>]
>> ipsec_mast_init_devices+0x7/0x10 [ipsec]
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<f8f8d0e0>]
>> ipsec_klips_init+0xe0/0x1e0 [ipsec]
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<c01a56d3>]
>> sysfs_add_file+0x53/0x90 Aug  4 14:30:18 Spyket-embratel kernel:
>> [<c01a7401>] create_files+0x31/0x60 Aug  4 14:30:18 Spyket-embratel kernel:
>>  [<c01568cc>] __vunmap+0xdc/0xf0 Aug  4 14:30:18 Spyket-embratel kernel:
>> [<c013c5fd>] load_module+0xa9d/0xbb0 Aug  4 14:30:18 Spyket-embratel
>> kernel:  [<c0132657>]
>> blocking_notifier_call_chain+0x17/0x20
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<f8f8d415>] init_module+0x5/0x10
>> [ipsec]
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<c013c7f9>]
>> sys_init_module+0xc9/0x120
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<c0103fe6>]
>> sysenter_past_esp+0x5f/0x85
>> Aug  4 14:30:18 Spyket-embratel kernel:  [<c05a0000>]
>> __find_acq_core+0x220/0x350
>> Aug  4 14:30:18 Spyket-embratel kernel:  =======================
>> Aug  4 14:30:18 Spyket-embratel kernel: Code: c6 0f 84 9e fe ff ff 89 f6 7e
>> 16 be fb ff ff ff eb 0f be ef ff ff ff 8b 93 d8 01 00 00 85 d2 75 45 83 c4
>> 1c 89 f0 5b 5e 5f 5d c3 <0f> 0b eb fe 0f 0b eb fe 8d b6 00 00 00 00 e8 6b
>> 69 00 00 c7 04
>> Aug  4 14:30:18 Spyket-embratel kernel: EIP: [<c054a772>]
>> register_netdevice+0x1e2/0x2f0 SS:ESP 0068:f6641e1c
>> Aug  4 14:30:18 Spyket-embratel kernel: ---[ end trace 978155c3692bc231
>> ]--- ===========
>>
>> Here is a quote from Linus Torvalds about this:
>>
>> ==========
>>
>> On Mon, 21 Jul 2008, Ingo Molnar wrote:
>>> [   24.434799] kernel BUG at net/core/dev.c:1328!
>>
>> Ok, it is now a WARN_ON_ONCE() in my tree (which I _just_ pushed out).
>>
>> So it's going to cause irritating messages (once), but the machine should
>> hopefully work.
>>
>>> Should i spend time on bisecting this, or is this known already?
>>
>> It's going to bisect down to the same commit you already bisected once,
>> it's the networking code that changed some of the rules, so various
>> network drivers that didn't follow the expected rules are now unhappy.
>>
>> Maybe the network drivers are few enough that it will get fixed, or maybe
>> the WARN_ON_ONCE() will just be removed and the rule not reinforced.
>>
>> I personally suspect the latter, since it seems to happen with just about
>> _any_ random network driver, including the common and well-maintained ones
>> (ie the Gods only help us for the truly odd/random cases)
>>
>> 			Linus
>> ==========
>>
>> don't the ipsec drivers fit in this?
>>
>> Thanks
>
> _______________________________________________
> Users at openswan.org
> http://lists.openswan.org/mailman/listinfo/users
> Building and Integrating Virtual Private Networks with Openswan:
> http://www.amazon.com/gp/product/1904811256/104-3099591-2946327?n=283155
>