Network not working after reboot

Enrico1989 · 2020-06-02 13:57:22

Foreword:

I have searched and found several related questions on the web, but I haven't found a solution to the issue. I apologize in advance if I have missed some important thread on this or other forums, but at the moment my only connection to the web is through the small screen of my phone. The issue described in the following prevents also from using the phone as router.

Here's what happened:

The system became unresponsive while I was in a MS Teams call (through the browser), so I rebooted it.
It got stuck (black screen with white blinking cursor on the top left for some minute) during the process, so I powered it off by holding the power button down.
Then I waited for it to cool down a bit (yeah, lately it's always on 70-80°C, and during MS Teams calls it goes as high as 95-102°C), and booted it again.

Surprise:

The network does not work.

My research & troubleshooting:

I use connman as a network manager, so I verified that "connmanctl services" shows empty output. Then I searched the web with the phone and found some possible solutions; mainly, Then “RTNETLINK answers: Input/Output error” given when trying to bring interface up points to several other related questions.

One of the suggestions is to execute

sudo ip link set dev wlp3s0 up

which for me gives

RTNETLINK answers: Input/output error

Another suggestion comes from iwlwifi: Failed to run INIT ucode: -5, but that did not work either for me.

At the moment it's a bit difficult to provide other details if I don't know what can be useful for you to help me, since I'm writing from the phone, and I have to manually copy whichever output you'll ask for, but if you suggest that I grep some command for specific lines, I can do that.

Again, I apologize, but I really need help on this, as I am working from home (well I should be, but I am not, because of this), and I only have my phone to ask for it

Last edited by Enrico1989 (2020-06-02 14:03:35)

seth · 2020-06-02 14:11:27

There's https://bbs.archlinux.org/viewtopic.php?id=256175 so if you're usign the LTS kernel…

Otherwise please post a complete system journal, you can use https://wiki.archlinux.org/index.php/Tethering to share your phones network connection or just copy the output to your phone and upload it from there.

nick_0278 · 2020-06-02 14:12:08

Wich kernel do you use (uname -r)?
If it's a Laptop: Is only wifi not working, or LAN too?

Last edited by nick_0278 (2020-06-02 14:23:46)

Enrico1989 · 2020-06-02 14:30:20

uname -r
5.6.15-arch1-1

Yes, it is a laptop. How should I check if the LAN is working, if I have no wires to connect the computer to the network?

I'm readin the links seth suggested.

seth, by system journal you mean the full output of journalctl?

Last edited by Enrico1989 (2020-06-02 14:32:33)

nick_0278 · 2020-06-02 14:33:47

I'm readin the links seth suggested.

The first won't help you, because you are already using the standard kernel.

seth wrote:

Otherwise please post a complete system journal, you can use https://wiki.archlinux.org/index.php/Tethering to share your phones network connection or just copy the output to your phone and upload it from there.

Post how he said a complete journal (journalctl -b)

Last edited by nick_0278 (2020-06-02 14:35:04)

seth · 2020-06-02 14:35:13

Unless your user is in the %wheel or %adm group, make it

sudo journalctl -b

(Otherwise you're only getting the session journal)

@nick_0278, the second link should still be vastly helpful ;-)

nick_0278 · 2020-06-02 14:36:13

Thats why I corrected my message ^^

Enrico1989 · 2020-06-02 14:49:53

Luckily, I was reading the second, and now I'm using the laptop, connect through the phone's wifi by USB. here's the journal.

So I huge first thank you for helping me being online again. I could have probably looked at the wiki myself, but I was in total panic.

Now I'm half in panic.

How can I get my network working again? Do you think it could be a hardware issue?

Last edited by Enrico1989 (2020-06-02 14:57:08)

seth · 2020-06-02 15:06:48

If you look at the older logs ("… -b -5" for the 5th last log), do you get the i915/drm crashes there?

a) You do not run a parallel windows installation, do you?
b) boot some live distro (grml, knoppix, ubuntu - if you must ;-) to check whether the i915/iwlwifi issues occur w/ a completely different SW stack.

I'm not gonna lie: if two different errors (in the IGP and the wifi) pop up together after a possible heat incident, that's bad :-(

If the temperature is constantly high (esp. directly after a cold boot), ensure it's not a D.U.S.T. issue: https://fixedlocally.co.uk/wp-content/u … 040446.jpg

Enrico1989 · 2020-06-02 16:19:41

seth, I get a lot of matches when doing for instance

sudo journalctl -b -5 | grep 'i915'

and I don't know how to interpret them. Maybe I searching for a specific regex could be more productive?

(update: when I check earlier, the sudo journalctl -b -5 command generated something about 6 millions lines long, now it generates just less than 2300 lines. Can that output become that big and then reduce so much after a reboot?)

a) No, I've got only linux on this laptop
b) I will have to find a falsh USB somewhere to do this

What is IGP?

Given the attached picture, I understand that D.U.S.T. is just "dust" with more emphasis on the importance of the issue, and not a true acronym, isn't it?

Concerning the temporary workaround of using the WiFi connection from the phone through the USB, I can't seem to get it working automatically. What I mean is that after physically connecting the cable, I see the change in the following output

$ ip address show 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp5s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 14:da:e9:36:4f:ea brd ff:ff:ff:ff:ff:ff
3: wlp3s0: <BROADCAST,MULTICAST,DYNAMIC> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 78:92:9c:24:9e:a0 brd ff:ff:ff:ff:ff:ff
5: enp4s0u1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 02:44:f1:66:a7:c7 brd ff:ff:ff:ff:ff:ff

where enp4s0u1 appears just after I connect the cable, but I have to enter

sudo ip link set dev enp4s0u1 up

to get it up. I thought this section was exactly to have the connection happen automatically when I physically connect the USB cable. Have I misunderstood something?

Last edited by Enrico1989 (2020-06-02 16:54:59)

seth · 2020-06-02 16:49:32

"Integrated graphics processor" and D.U.S.T. is just dust, yes.
You're looking for stuff like

Jun 02 13:29:02 redviper kernel: ------------[ cut here ]------------

Jun 02 13:29:02 redviper kernel: WARN_ON(clock->n + 2 == 0 || clock->p == 0)

Jun 02 13:29:02 redviper kernel: WARNING: CPU: 2 PID: 345 at drivers/gpu/drm/i915/display/intel_display.c:592 i9xx_calc_dpll_params+0x3f/0xb0 [i915]

Jun 02 13:29:02 redviper kernel: Modules linked in: mac80211 i915(+) libphy ttm snd_hda_core libarc4 joydev mousedev snd_hwdep snd_pcm i2c_algo_bit drm_kms_helper snd_timer snd iwlwifi crct10dif_pclmul mxm_wmi cec crc32_pclmul input_leds asus_nb_wmi ghash_clmulni_intel mac_hid rc_core asus_wmi aesni_intel soundcore iTCO_wdt iTCO_vendor_support intel_gtt crypto_simd sparse_keymap syscopyarea sysfillrect cryptd mei_hdcp sysimgblt cfg80211 fb_sys_fops glue_helper evdev psmouse rfkill mei_me intel_cstate intel_uncore mei lpc_ich wmi pcspkr i2c_i801 intel_rapl_perf battery ac drm agpgart ip_tables x_tables usbhid hid ext4 crc32c_generic crc16 mbcache jbd2 serio_raw atkbd libps2 xhci_pci crc32c_intel xhci_hcd sr_mod ehci_pci cdrom ehci_hcd i8042 serio

Jun 02 13:29:02 redviper kernel: CPU: 2 PID: 345 Comm: systemd-udevd Not tainted 5.6.15-arch1-1 #1

Jun 02 13:29:02 redviper kernel: Hardware name: ASUSTeK Computer Inc. K53SC/K53SC, BIOS K53SC.216 09/26/2011

Jun 02 13:29:02 redviper kernel: RIP: 0010:i9xx_calc_dpll_params+0x3f/0xb0 [i915]

Jun 02 13:29:02 redviper kernel: Code: 0c 8b 4e 0c 0f af 4e 10 89 56 1c 89 4e 20 85 c9 74 05 83 ff fe 75 18 48 c7 c6 d8 c2 c7 c0 48 c7 c7 9e cd c9 c0 e8 c8 cd 4f cf <0f> 0b 31 c0 c3 0f af c2 83 c7 02 85 c0 41 0f 9e c0 85 ff 0f 9f c2

Jun 02 13:29:02 redviper kernel: RSP: 0018:ffffb3e9805ff820 EFLAGS: 00010282

Jun 02 13:29:02 redviper kernel: RAX: 0000000000000000 RBX: 0000000001000000 RCX: 0000000000000000

Jun 02 13:29:02 redviper kernel: RDX: 0000000000000001 RSI: ffffffff91b0e2eb RDI: 0000000000000246

Jun 02 13:29:02 redviper kernel: RBP: ffff912681eeb000 R08: 00000003f013d68e R09: 000000000000002b

Jun 02 13:29:02 redviper kernel: R10: 0000000000000159 R11: 0000000000000001 R12: 000000000001d4c0

Jun 02 13:29:02 redviper kernel: R13: 0000000000000000 R14: 0000000000000001 R15: ffff91267cc107b0

Jun 02 13:29:02 redviper kernel: FS:  00007f73ab717a80(0000) GS:ffff912687880000(0000) knlGS:0000000000000000

Jun 02 13:29:02 redviper kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033

Jun 02 13:29:02 redviper kernel: CR2: 0000562f1ba4fdc8 CR3: 0000000243a1a003 CR4: 00000000000606e0

Jun 02 13:29:02 redviper kernel: Call Trace:

Jun 02 13:29:02 redviper kernel:  i9xx_crtc_clock_get.isra.0+0x134/0x2a0 [i915]

Jun 02 13:29:02 redviper kernel:  ilk_get_pipe_config+0x2ea/0x370 [i915]

Jun 02 13:29:02 redviper kernel:  intel_modeset_setup_hw_state+0x1a9/0x1670 [i915]

Jun 02 13:29:02 redviper kernel:  ? _raw_spin_lock_irqsave+0x26/0x50

Jun 02 13:29:02 redviper kernel:  ? drm_modeset_lock_all_ctx+0x1be/0x2e0 [drm]

Jun 02 13:29:02 redviper kernel:  intel_modeset_init+0xa5c/0x1e40 [i915]

but if this is a temperature issue, the errors can be random, so don't look for specific patterns, but crashes involving the i915 module.

Edit: that section will require you to adapt the vendor/product ID and the resulting action (since you're using connman rather than systemd-networkd?)

Last edited by seth (2020-06-02 16:52:19)

Enrico1989 · 2020-06-02 17:04:52

Sorry, I might be dumb, but I do have that file, so those lines are indeed there. Maybe you mean that I should look for those lines when I run that command from a live distro? If that's the case, then I will update you on this as soon as I can try this.

Assuming the errors will be somewhat random, what would searching for crashes involving the i915 module mean? As in, in the very chunk of the journal that you quoted, what is/are the part(s) that indicate a crash involving that module (and not something fine about that module)?

Can you help me understand how I adapt the suggestions from that section to the my case? (Yes, I'm using connaman).

Last edited by Enrico1989 (2020-06-02 17:06:03)

seth · 2020-06-02 20:20:30

No, in an older journal, eg. "sudo journalctl -b -5"

kernel: ------------[ cut here ]------------

is indicative of *some* kernel module crash. What it actually is, is determined by the backtrace below - at least some of the lines would end in "[i915]" or "[drm]".

Enrico1989 · 2020-06-02 20:40:14

Ok, that is what is in the journal that I have shared:

Jun 02 13:29:02 redviper kernel: ------------[ cut here ]------------
Jun 02 13:29:02 redviper kernel: pipe_off wait timed out
Jun 02 13:29:02 redviper kernel: WARNING: CPU: 2 PID: 345 at drivers/gpu/drm/i915/display/intel_display.c:1093 intel_disable_pipe+0x218/0x220 [i915]
Jun 02 13:29:02 redviper kernel: Modules linked in: iwldvm(+) realtek mac80211 i915(+) libphy ttm snd_hda_core libarc4 joydev mousedev snd_hwdep snd_pcm i2c_algo_bit drm_kms_helper snd_timer snd iwlwifi crct10dif_pclmul mxm_wmi cec crc32_pclmul input_leds asus_nb_wmi ghash_clmulni_intel mac_hid rc_core asus_wmi aesni_intel soundcore iTCO_wdt iTCO_vendor_support intel_gtt crypto_simd sparse_keymap syscopyarea sysfillrect cryptd mei_hdcp sysimgblt cfg80211 fb_sys_fops glue_helper evdev psmouse rfkill mei_me intel_cstate intel_uncore mei lpc_ich wmi pcspkr i2c_i801 intel_rapl_perf battery ac drm agpgart ip_tables x_tables usbhid hid ext4 crc32c_generic crc16 mbcache jbd2 serio_raw atkbd libps2 xhci_pci crc32c_intel xhci_hcd sr_mod ehci_pci cdrom ehci_hcd i8042 serio
Jun 02 13:29:02 redviper kernel: CPU: 2 PID: 345 Comm: systemd-udevd Tainted: G        W         5.6.15-arch1-1 #1
Jun 02 13:29:02 redviper kernel: Hardware name: ASUSTeK Computer Inc. K53SC/K53SC, BIOS K53SC.216 09/26/2011
Jun 02 13:29:02 redviper kernel: RIP: 0010:intel_disable_pipe+0x218/0x220 [i915]

,
the "cut here" line followed by some lines ending by [i915], so we where looking for stuff like that, and we have found it. Now what are we doing with this?

At the moment I cannot afford disassemble the computer to clean it, because if something goes wrong (as I am not a technician), then I cannot work. I can definitely stress it less, for instance using the smartphone for MS Teams or other programs that, more or less expectedly, use a lot of CPU.

However, I would like to get the internet working again (not tethering the wifi from the phone). Do we have something else we can try to diagnose the problem?

seth · 2020-06-02 20:49:45

There's probably some language barrier here…

That's from the most recent boot, "sudo journalctl -b" - equivalent to "sudo journalctl -b 0".
We're interested in an older boot, eg. "sudo journalctl -b -5" to go 5 boots back in time.
Because we want to know whether the problems w/ the IGP that exist now also existed in the past.
If they did, they're insignificant.
If not, this raises chances that your HW took some hit.

Enrico1989 · 2020-06-02 22:15:10

Ok, in this case here's the output of sudo journalctl -b -5 a few hours ago. It contains that line followed by some lines ending with [drm] or [i915]. Even If I run the command "... -b -5" now, I still see that kind of pattern. Based on your last message, this does not add much to the investigation, right?

seth · 2020-06-03 05:32:56

It also has the iwlwifi crashes, you'll have to go further back in time until you reach a date before the wifi issues popped up.
Notice that the offset is relative, the more you reboot, the bigger it will have to be.
You can also use --since and --until to select a timeframe, see https://wiki.archlinux.org/index.php/Sy … ing_output

Enrico1989 · 2020-06-03 07:26:18

Based on what I read from the man page of journalctl, the following output could help.

$ sudo journalctl --list-boots
 -10 6fdba210eaa747d8a599150740aa9d06 Fri 2020-05-22 18:32:43 BST—Fri 2020-05-22 18:33:09 BST
  -9 95b9a894e40a4e038eedbbe45d8a1b73 Fri 2020-05-22 18:34:53 BST—Tue 2020-05-26 12:01:23 BST
  -8 ee36291e2e2f4e518f2c380e73e2081d Tue 2020-05-26 12:51:14 BST—Tue 2020-05-26 17:03:37 BST
  -7 234f1a2593e5480d875af4d323d38ed6 Tue 2020-05-26 20:59:54 BST—Wed 2020-05-27 13:34:23 BST
  -6 1df14a782ac34a2583066e590989a5e5 Wed 2020-05-27 14:12:04 BST—Tue 2020-06-02 10:41:39 BST # This is when I rebooted;
  -5 18c4df3df5ab42aa832b8f15e3056e6b Tue 2020-06-02 10:44:22 BST—Tue 2020-06-02 10:54:01 BST # then I found that the wifi did not work
  -4 633ff5a22d9541198ffcc9f8f811d78c Tue 2020-06-02 10:54:22 BST—Tue 2020-06-02 11:16:01 BST
  -3 6552e25d14f74c2e95fc329a5a6a4377 Tue 2020-06-02 11:17:56 BST—Tue 2020-06-02 11:46:57 BST
  -2 78b489a36a414b58b0e21815847de0ed Tue 2020-06-02 11:48:59 BST—Tue 2020-06-02 12:24:13 BST
  -1 19d431a6a49c45dfa64441b053db719f Tue 2020-06-02 13:28:49 BST—Tue 2020-06-02 17:34:47 BST
   0 7b6698e06d8945db89ddb61b8d49d964 Tue 2020-06-02 17:35:08 BST—Wed 2020-06-03 07:51:14 BST

However, I've checked 100 entries for matches of "[ cut here ]" and the matches go this way

...
sudo -E journalctl -b -56 ===> 4 matches
sudo -E journalctl -b -57 ===> 4 matches
sudo -E journalctl -b -58 ===> 4 matches
sudo -E journalctl -b -59 ===> 4 matches
sudo -E journalctl -b -60 ===> 4 matches
sudo -E journalctl -b -61 ===> 0 matches
sudo -E journalctl -b -62 ===> 1 matches # here after the [ cut here ] line, there's a line ending with [iwlwifi], but no line ending with [i915]
sudo -E journalctl -b -63 ===> 0 matches
sudo -E journalctl -b -64 ===> 0 matches
sudo -E journalctl -b -65 ===> 0 matches
sudo -E journalctl -b -66 ===> 0 matches
sudo -E journalctl -b -67 ===> 0 matches
...

So here are the journals for -61, -62, -63.

Does this help understand if it's possible to recover the wifi?

seth · 2020-06-03 11:39:58

Those journals are from late March and afaiu the problem only showed up very recently?
So

  -8 ee36291e2e2f4e518f2c380e73e2081d Tue 2020-05-26 12:51:14 BST—Tue 2020-05-26 17:03:37 BST
  -7 234f1a2593e5480d875af4d323d38ed6 Tue 2020-05-26 20:59:54 BST—Wed 2020-05-27 13:34:23 BST
  -6 1df14a782ac34a2583066e590989a5e5 Wed 2020-05-27 14:12:04 BST—Tue 2020-06-02 10:41:39 BST # This is when I rebooted;

are far more interesting.

Enrico1989 · 2020-06-03 13:09:37

Yes, they are from March, and, since then, there's always neen 1 or more lines containing "[ cut here ]".

The heating issue, indeed, has been a problem for a long time.

However, today I'm not using MS Teams at all (or any other CPU intensive program), and the temperature is not going above 55°C, which I think is fine.

The problem with not being able to use the network happened the first time when I opened this thread; never had issues before.

If those 3 lines are interesting, then I will post the corresponding files after work. Thank you for the patience.

Enrico1989 · 2020-06-04 20:49:20

Hi seth, I'm sorry for the late update.

However, this are -8 and -7, as I have some difficulty in uploading the -6, as it is 6 million lines for almost 900MB. Any suggestions on which part I should look at? There are 75 [ cut here ] lines in it.

seth · 2020-06-04 21:03:44

-8 from May 26 12:51 and -7 from later that day already have 4 i915/drm crashes each, so that's probably insignificant.
Did you meanwhile try a live distro to rule out HW failure?

Enrico1989 · 2020-06-06 16:39:50

I couldn't try it yet. I will update you after June the 13th.

Enrico1989 · 2020-06-16 18:32:24

It does not start form a live usb

I tried with ubuntu, but when I boot it from the usb, I reach the GRUB screen, choose Ubuntu, and then I get a black screen (turned on, I mean) nothing happens, except the fan goes faster and the temperature goes high.

nick_0278 · 2020-06-16 18:33:31

Did you try other live systems or USBs?

Last edited by nick_0278 (2020-06-16 18:33:44)

Arch Linux

#1 2020-06-02 13:57:22

Network not working after reboot

#2 2020-06-02 14:11:27

Re: Network not working after reboot

#3 2020-06-02 14:12:08

Re: Network not working after reboot

#4 2020-06-02 14:30:20

Re: Network not working after reboot

#5 2020-06-02 14:33:47

Re: Network not working after reboot

#6 2020-06-02 14:35:13

Re: Network not working after reboot

#7 2020-06-02 14:36:13

Re: Network not working after reboot

#8 2020-06-02 14:49:53

Re: Network not working after reboot

#9 2020-06-02 15:06:48

Re: Network not working after reboot

#10 2020-06-02 16:19:41

Re: Network not working after reboot

#11 2020-06-02 16:49:32

Re: Network not working after reboot

#12 2020-06-02 17:04:52

Re: Network not working after reboot

#13 2020-06-02 20:20:30

Re: Network not working after reboot

#14 2020-06-02 20:40:14

Re: Network not working after reboot

#15 2020-06-02 20:49:45

Re: Network not working after reboot

#16 2020-06-02 22:15:10

Re: Network not working after reboot

#17 2020-06-03 05:32:56

Re: Network not working after reboot

#18 2020-06-03 07:26:18

Re: Network not working after reboot

#19 2020-06-03 11:39:58

Re: Network not working after reboot

#20 2020-06-03 13:09:37

Re: Network not working after reboot

#21 2020-06-04 20:49:20

Re: Network not working after reboot

#22 2020-06-04 21:03:44

Re: Network not working after reboot

#23 2020-06-06 16:39:50

Re: Network not working after reboot

#24 2020-06-16 18:32:24

Re: Network not working after reboot

#25 2020-06-16 18:33:31

Re: Network not working after reboot

Board footer