Networking - Destination unreachable (Fragmentation needed)

[Arthur Fragoso]

A few years ago, I was using Fedora 16 and I was unable to do any HTTP POST at the place where I work. So I tried Ubuntu 12.04 (or 12.10, I'm not quite sure) and it worked perfectly.
Recently I updated it to 14.04 and the same problem I had in Fedora came back.
I'm not an expert in understanding these network packages, but I read that it could be that the network manager could had blocked the ICMP MTU dicovery package. I can check and if it is so I can ask him to unblock it, but I think that more people could also have this problem, so I wonder how it was working in the previous 12.04/12.10 version and if it should work the same in the new version.
Or maybe it's a different problem, I only have this problem at the place where I work.

$ cat /proc/version_signature
Ubuntu 3.13.0-32.57-generic 3.13.11.4

$ lsb_release -rd
Description: Ubuntu 14.04.1 LTS
Release: 14.04

ProblemType: Bug
DistroRelease: Ubuntu 14.04
Package: linux-image-3.13.0-32-generic 3.13.0-32.57
ProcVersionSignature: Ubuntu 3.13.0-32.57-generic 3.13.11.4
Uname: Linux 3.13.0-32-generic x86_64
NonfreeKernelModules: nvidia
ApportVersion: 2.14.1-0ubuntu3.2
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: arthur 3196 F.... pulseaudio
 /dev/snd/controlC0: arthur 3196 F.... pulseaudio
 /dev/snd/pcmC0D0p: arthur 3196 F...m pulseaudio
CurrentDesktop: Unity
Date: Sat Jul 26 17:16:31 2014
HibernationDevice: RESUME=UUID=2a262003-49b8-4b6d-b2a5-4b93996c92cb
InstallationDate: Installed on 2013-08-09 (351 days ago)
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Release amd64 (20120425)
MachineType: Dell Inc. XPS L501X
ProcFB:

ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-3.13.0-32-generic root=/dev/mapper/fedora_xps-ubuntu_root ro quiet splash
RelatedPackageVersions:
 linux-restricted-modules-3.13.0-32-generic N/A
 linux-backports-modules-3.13.0-32-generic N/A
 linux-firmware 1.127.5
SourcePackage: linux
UpgradeStatus: Upgraded to trusty on 2014-03-15 (133 days ago)
dmi.bios.date: 07/26/2011
dmi.bios.vendor: Dell Inc.
dmi.bios.version: A08
dmi.board.name: 00YWG2
dmi.board.vendor: Dell Inc.
dmi.board.version: A08
dmi.chassis.type: 8
dmi.chassis.vendor: Dell Inc.
dmi.chassis.version: A08
dmi.modalias: dmi:bvnDellInc.:bvrA08:bd07/26/2011:svnDellInc.:pnXPSL501X:pvrA08:rvnDellInc.:rn00YWG2:rvrA08:cvnDellInc.:ct8:cvrA08:
dmi.product.name: XPS L501X
dmi.product.version: A08
dmi.sys.vendor: Dell Inc.

[Arthur Fragoso (arthur-p)]

wireshark output

[Brad Figg (brad-figg)]

This change was made by a bot.

[penalvch (penalvch)]

Arthur Fragoso, thank you for reporting this and helping make Ubuntu better. Could you please test the latest upstream kernel available from the very top line at the top of the page (not the daily folder) following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-3.16-rc6

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description.

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

[Arthur Fragoso (arthur-p)]

I tested on 3.16-rc6-utopic and the problem was still there.

I also downloaded an old kernel, the v3.4-rc6-precise to test if it was working in a past kernel, and it was working without problems.

I'm going to test other kernel versions to see in what version the bug was introduced.

[Arthur Fragoso (arthur-p)]

I did more tests, and here is the result:
The version v3.5.7.33-quantal and older versions works without problems.
The version v3.6-quantal and newer has the problem/bug.

[penalvch (penalvch)]

Arthur Fragoso, the next step is to fully commit bisect the kernel in order to identify the offending commit. Could you please do this following https://wiki.ubuntu.com/Kernel/KernelBisection ?

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

[Arthur Fragoso (arthur-p)]

Sorry for being so late, my disk was full and I procrastinated to get the about 10 GB for compiling the kernel.

I started doing the bisection:

arthur@XPS:/media/arthur/Storage/kernel/ubuntu-raring$ git bisect start Ubuntu-3.6.0-0.1 Ubuntu-3.5.0-5.5
Bisecting: a merge base must be tested
[84a1caf1453c3d44050bd22db958af4a7f99315c] Linux 3.5-rc7
arthur@XPS:/media/arthur/Storage/kernel/ubuntu-raring$ git bisect good

Bisecting: 5456 revisions left to test after this (roughly 13 steps)
[b13bc8dda81c54a66a1c84e66f60b8feba659f28] Merge tag 'staging-3.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
arthur@XPS:/media/arthur/Storage/kernel/ubuntu-raring$ git bisect bad

Bisecting: 2384 revisions left to test after this (roughly 12 steps)
[3c4cfadef6a1665d9cd02a543782d03d3e6740c6] Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
arthur@XPS:/media/arthur/Storage/kernel/ubuntu-raring$ git bisect bad

Bisecting: 1953 revisions left to test after this (roughly 11 steps)
[e017507f37d5cb8b541df165a824958bc333bec3] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
arthur@XPS:/media/arthur/Storage/kernel/ubuntu-raring$ git bisect good

Bisecting: 966 revisions left to test after this (roughly 10 steps)
[0cd06647b7c24f6633e32a505930a9aa70138c22] Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next

And I got a kernel panic from the 0cd066 build, so I can't proceed by saying "bisect good or bad", what should I do?

[Arthur Fragoso (arthur-p)]

Sorry, I just read in the docs, `git bisect --help`

I can skip some versions, I will try `git reset --hard HEAD~50`

[Arthur Fragoso (arthur-p)]

I did the bisection and the results are:

c0efc887dcadbdbfe171f028acfab9c7c00e9dde is good.

97bab73f987e2781129cd6f4b6379bf44d808cc6 - gives me a kernel panic when trying to do an upload of data.

7b34ca2ac7063f4ebf07f85fd75253ed84d5c648 - same problem, kernel panic when trying to do an upload of data.

and I think the kernel panic goes up to 352e04b9111d608bd89ba7bd8070846d4f97d104

352e04b9111d608bd89ba7bd8070846d4f97d104 - doesn't give me a kernel panic, but it just hangs when trying to upload data, and this is the problem that I currently face in the upstream kernel.

Those commits were made by David S. Miller, so I will email him asking for a help with this bug.

Attached "git bisect visualize" print.

[Arthur Fragoso (arthur-p)]

Attached BISECT_LOG

[Arthur Fragoso (arthur-p)]

Almost forgot to say,

to be able to compile those commits, I had to apply this patch:

wget https://crewrktablets.arctablet.com:8081/opendev/kernel_rockchip/commit/33d5e332b9c5ce0bf3bfd44ca2127d1773b3f2ad.diff --no-check-certificate
patch -p1 < 33d5e332b9c5ce0bf3bfd44ca2127d1773b3f2ad.diff

[penalvch (penalvch)]

Arthur Fragoso, it is best to both send an e-mail to the maintainer mailing list, as well as CC the person who made the commit. Hence, the issue you are reporting is an upstream one. Could you please report this problem to the appropriate mailing list (netdev) by following the instructions _verbatim_ at https://wiki.ubuntu.com/Bugs/Upstream/kernel ?

Please provide a direct URL to your e-mail to the mailing list once you have made it so that it may be tracked.

Thank you for your understanding.

[Arthur Fragoso (arthur-p)]

I wrote to the netdev mailing list in november ( http://www.spinics.net/lists/netdev/msg305646.html ) and I got no reply, so I'm attaching the result of a test another test that I did that could make it more easy to find the problem.

This test shows the wireshark output of how a "git push" behaves in the kernel that has the bug and in the kernel that doesn't have the bug.

Thanks Christopher for all the help, you really did a great job in taking me through the necessary steps.

I just want to make another notice that I will probably move to a different city around the end of January, so after that it will be very difficult for me to test any possible patch, as I will only be able to test when I visit my old city.

[penalvch (penalvch)]

Arthur Fragoso, you may want to update your test with the latest mainline kernel 3.19-rc2, re-send the e-mail following the format noted in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1349000/comments/14 (November's e-mail did not), and CC both the regression submitter and the maintainer/sub-maintainer of the driver module.

[Arthur Fragoso (arthur-p)]

The problem stopped happening, so I contacted the network manager and he told me that they changed many equipment that was damaged after a lightning in a thunderstorm night, he also told me that it probably was some rule in the firewall of the old router.

Thanks for all the help, and I'm sorry that I can't reproduce it again and nothing can be done here.

[penalvch (penalvch)]

Arthur Fragoso, this bug report is being closed due to your last comment https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1349000/comments/17 regarding you no longer have the original hardware this is reproducible with. For future reference you can manage the status of your own bugs by clicking on the current status in the yellow line and then choosing a new status in the revealed drop down box. You can learn more about bug statuses at https://wiki.ubuntu.com/Bugs/Status. Thank you again for taking the time to report this bug and helping to make Ubuntu better. Please submit any future bugs you may find.