Bug 633874

Summary: NetworkManager fails some times after a suspend (probably ath9k driver???IRQ Shared)
Product: [Fedora] Fedora Reporter: Maximiliano Castanon <maximi89>
Component: NetworkManagerAssignee: Dan Williams <dcbw>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: 12CC: dcbw
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2010-10-18 17:57:29 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
/var/log/messages|grep -i irq none

Description Maximiliano Castanon 2010-09-14 15:10:26 UTC 
Description of problem:
When i resume from suspend state, some times, not all, i get a NetworkManager who can't initiate a connection to wifi, because probably a rfkill state indicate it are blocked, i get this from:

BAD:
[root@gateway foo2zjs]# cat /sys/devices/platform/acer-wmi/rfkill/rfkill0/state 
1
[root@gateway foo2zjs]# cat /sys/devices/pci0000\:00/0000\:00\:06.0/0000\:09\:00.0/ieee80211/phy0/rfkill2/state
2


NORMAL:
[root@gateway foo2zjs]# cat /sys/devices/platform/acer-wmi/rfkill/rfkill0/state 
1
[root@gateway foo2zjs]# cat /sys/devices/pci0000\:00/0000\:00\:06.0/0000\:09\:00.0/ieee80211/phy0/rfkill2/state
1


I do a "killall NetworManager" to get it back... and then i call it again "NetworkManager", and works normally.

Version-Release number of selected component (if applicable):


How reproducible:
some times

Steps to Reproduce:
1.suspend/probably hibernate, but i don't do any test in hibernate state...
2.resume
3.and some times NetworkManager remains blocked.
  
Actual results:
it fails...

Expected results:
to works normally :D

Additional info:
BAD:
[root@gateway foo2zjs]# cat /sys/devices/platform/acer-wmi/rfkill/rfkill0/state 
1
[root@gateway foo2zjs]# cat /sys/devices/pci0000\:00/0000\:00\:06.0/0000\:09\:00.0/ieee80211/phy0/rfkill2/state
2

you can check the rfkill numbers...

PD: i found a relationed bug with RFKILL
https://bugzilla.redhat.com/show_bug.cgi?id=533601
Comment 1 Maximiliano Castanon 2010-09-14 15:10:53 UTC 
The package who i have it's 

[maximi89@gateway foo2zjs]$ rpm -q NetworkManager
NetworkManager-0.8.1-6.git20100831.fc12.i686
Comment 2 Maximiliano Castanon 2010-09-15 03:57:26 UTC 
Not a rfkill problem, it happened again, i checked the rfkill states, both at 1, but removing the module and loading it again the network start to work...
Comment 3 Maximiliano Castanon 2010-09-15 04:00:36 UTC 
or probably more than 1 problem...
Comment 4 Maximiliano Castanon 2010-09-16 04:32:49 UTC 
Well, i was doing an investigation about what cause it, and looking at 
/var/log/message |grep -i IRQ,  i found this:

http://fpaste.org/LAMz/
Or check the file or read a part of the log... 

here:

Sep 14 23:41:20 gateway kernel: radeon 0000:01:05.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
Sep 14 23:41:20 gateway kernel: HDA Intel 0000:01:05.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19
Sep 14 23:41:20 gateway kernel: ath9k 0000:09:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
Sep 14 23:41:22 gateway kernel: irq 18: nobody cared (try booting with the "irqpoll" option)
Sep 14 23:41:22 gateway kernel: [<c048c9d8>] __report_bad_irq+0x33/0x74
Sep 14 23:41:22 gateway kernel: [<c048d0e8>] handle_fasteoi_irq+0x83/0xa2
Sep 14 23:41:22 gateway kernel: [<c040b19f>] handle_irq+0x40/0x4d
Sep 14 23:41:22 gateway kernel: [<c040a965>] do_IRQ+0x46/0x9a
Sep 14 23:41:22 gateway kernel: [<c06aa027>] (usb_hcd_irq+0x0/0x6f)
Sep 14 23:41:22 gateway kernel: [<f7fe8728>] (radeon_driver_irq_handler_kms+0x0/0x19 [radeon])
Sep 14 23:41:22 gateway kernel: Disabling IRQ #18
Comment 5 Maximiliano Castanon 2010-09-16 04:33:53 UTC 
Created attachment 447628 [details]
/var/log/messages|grep -i irq

Sep 15 22:02:46 gateway kernel: irq 18: nobody cared (try booting with the "irqpoll" option)
Sep 15 22:02:46 gateway kernel: [<c048c9d8>] __report_bad_irq+0x33/0x74
Sep 15 22:02:46 gateway kernel: [<c048d0e8>] handle_fasteoi_irq+0x83/0xa2
Sep 15 22:02:46 gateway kernel: [<c040b19f>] handle_irq+0x40/0x4d
Sep 15 22:02:46 gateway kernel: [<c040a965>] do_IRQ+0x46/0x9a
Sep 15 22:02:46 gateway kernel: [<c06aa027>] (usb_hcd_irq+0x0/0x6f)
Sep 15 22:02:46 gateway kernel: [<f7fe8728>] (radeon_driver_irq_handler_kms+0x0/0x19 [radeon])
Sep 15 22:02:46 gateway kernel: Disabling IRQ #18
Comment 6 Dan Williams 2010-10-15 15:16:55 UTC 
When this happens, can you run:

nmcli nm

and report that output?
Comment 7 Maximiliano Castanon 2010-10-18 17:40:43 UTC 
[maximi89@gateway gnash2]$ cat ./../nmcli
EJECUTANDO      ESTADO          HARDWARE-WIFI   WIFI       HARDWARE-WWAN   WWAN      
ejecutando      conectando      habilitado      habilitado habilitado      habilitado
[maximi89@gateway gnash2]$ cat ./../nmcli2
EJECUTANDO      ESTADO          HARDWARE-WIFI   WIFI       HARDWARE-WWAN   WWAN      
ejecutando      dormido         habilitado      habilitado habilitado      habilitado
Comment 8 Dan Williams 2010-10-18 17:57:29 UTC 
Ok, this has recently been fixed and an update has been proposed for F12, F13, and F14.  Look for details in the bug that this one is duplicated to.  Or check out the update notification here:

https://admin.fedoraproject.org/updates/NetworkManager-0.8.1-9.git20100831.fc12

*** This bug has been marked as a duplicate of bug 638640 ***