Re: [opensuse] opensuse network fails on reboot
On 04/24/2015 01:33 PM, don fisher wrote:
On 04/24/2015 09:43 AM, Anton Aylward wrote:
What's your use-case that needs BOTH to be active at the same time?
Or if that is not the case, please give a better explanation, because that seems to be how its coming across.
My interpretation of documentation from the yast2 GUI was that if the "Snartnode" was set to plugd, and the priorities were wet to that the cable that preference over wifi, the wifi would run until a cable was plugged into the Ethernet port. The the cable would then be active and the wifi would go to the background. The hardwired port state is setup-in-progress, which I assumed was waiting for the cable to be plugged in. My intent is to have both interfaces in my machine, with the wifi active until I plug a cable into the Ethernet port. The cable usually has better bandwidth, so I wanted it tpreemptmt the wifi when available.
I hope I did not read this incorrectly.
Well, you've jumped in the deep end with that. perhaps the reason I make progress albeit slowly is that I don't bite things off in big chunks. I do 'incrementalism'. If I were faced with that I would first get wired working with wifi completely out of the picture. Then I'd get wifi working with wired completely out of the picture. hey! Wait up! That EXACTLY what I *DID* do! OK so it was the days before wikd, but never the less. The point here is that I *KNEW* for certain what did work as components before trying to combine them. Have you established them BOTH individually as components? Have you established that you can switch between them manually with both configured? I could live with manually disabling wifi when I plugged a cable in. As i said, my use-cases are sufficiently differentiated between home(=wired) and not-home(=wifi) that manually switching is not a big issue, not a deal breaker. Do think about your actual use-case as opposed to a hypothetical. How often, where, when, why, will you be running on wifi then suddenly an Ethernet connection appears? Will manually switching make life impossible for you? -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Hello, I am curious what you may think if the attached output from dmesg. Boot time is Fri Apr 24 08:26:51 2015, just for a reference. The first mention of wlp11s0 is at 08:26:59, 8 seconds after boot. At 08:27:03 it says that is associated (?), then at 08:29:59 it says it is deauthenticating (?). Then the procedure starts again at 08:30:07. That still has to be during the boot process, before I perform if ifdown/ifup sequence, which does not appear to show up in dmesg. The only mention of the hardwired port during this boot is at 08:26:59 where it says "enp9s0: link is not ready". I admit to never having become knowledgeable about the network interface system. In the past I have always lived behind a system where everything was provided. I will try your experiment with no hardwired port at all later. Don
В Fri, 24 Apr 2015 12:48:05 -0700 don fisher <hdf3@comcast.net> пишет:
Hello,
I am curious what you may think if the attached output from dmesg. Boot
[Fri Apr 24 08:31:51 2015 < 74.964715>] mce: [Hardware Error]: Machine check events logged This deserves some investigation. It is usually associated with memory/cache/CPU problem or interconnect between them. You may consider to install mcelog.
time is Fri Apr 24 08:26:51 2015, just for a reference. The first mention of wlp11s0 is at 08:26:59, 8 seconds after boot. At 08:27:03 it says that is associated (?), then at 08:29:59 it says it is deauthenticating (?). Then the procedure starts again at 08:30:07. That still has to be during the boot process, before I perform if ifdown/ifup sequence, which does not appear to show up in dmesg. The only mention of the hardwired port during this boot is at 08:26:59 where it says "enp9s0: link is not ready".
This looks like wpa_supplicant is started and connects then something else (wicked? NetworkManager?) is started and resets wireless interface state to start connection.
I admit to never having become knowledgeable about the network interface system. In the past I have always lived behind a system where everything was provided.
I will try your experiment with no hardwired port at all later.
Don
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 04/24/2015 01:05 PM, Andrei Borzenkov wrote:
This deserves some investigation. It is usually associated with memory/cache/CPU problem or interconnect between them. You may consider to install mcelog.
This looks like wpa_supplicant is started and connects then something else (wicked? NetworkManager?) is started and resets wireless interface state to start connection.
I have been running mcelog, but perhaps the setup is not correct. Form the entire output of dmesg there are two entries: 225:[ 0.026016] mce: CPU supports 9 MCE banks 1136:[ 300.710483] mce: [Hardware Error]: Machine check events logged But I do not know where the errors are logged. I thought it might not be running, and tried to start as --daemon. That at leas got me a /var/log/mcelog file (none present before) with the sole entry "mcelog: mcelog server already running". Does the last line above at 1136 say there were errors, or that it will just log the errors? And if so, where? I have not touched the mcelog.conf file. Don -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (3)
-
Andrei Borzenkov -
Anton Aylward -
don fisher