ambiguous error message "sh: echo: I/O error"

[Karl-Philipp Richter]

[SRU justification]

[Impact]

Without this patch, trying to add a bond when this one is already defined leads to the following error :

Example :

# ifenslave bond0 eth1 eth2
sh: echo: I/O error
eth1: could not add interface

[Test Case]

For info, the ambiguous message only appears when using ifenslave on a bond that is already configured :

# ifconfig eth1 up
# ifconfig eth2 up
# ifconfig bond0 up
# ifenslave bond0 eth1 eth2 <<<< No error on the first execution
# ifenslave bond0 eth1 eth2
sh: echo: I/O error
eth1: could not add interface

[Regression Potential]

None expected as the same commit is already in use in the ifenslave package for Wily and in the upstream debian package.

[Original description of the problem]

adding an interface to a bonding interface can result in the error message "sh: echo: I/O error" which is highly ambiguous (did the I/O error occur in sh, echo or is it an I/O error realted to the command/driver? - at least change this to "I/O error occured" which is still not helpful ("I/O occured in the kernel driver, please make sure it supports XY, check with <Z>" would be nice), but at least not confusing.

reproduction:
/etc/wpa_supplicant/wpa_supplicant.conf
<code>
network={
    ssid="assid"
    key_mgmt=WPA-EAP
    eap=TTLS
    proto=WPA RSN
    <email address hidden>"
    <email address hidden>"
    ca_cert="/etc/wpa_supplicant/deutsche-telekom-root-ca-2.crt"
    phase2="auth=PAP"
}
</code>
/etc/network/interfaces
<code>
iface main inet dhcp
    wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

iface bond0 inet manual
    down ifenslave -d bond0 wlan0 wlan1
    bond-mode active-backup
    bond-miimon 100
    bond-slaves wlan0 wlan1

allow-bond0 wlan0
iface wlan0 inet manual
    wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf
    bond-master bond0

allow-bond0 wlan1
iface wlan1 inet manual
    wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf
    bond-master bond0

iface eth1 inet dhcp
</code>
my commands
<code>
sudo modprobe bonding
sudo ifup wlan0
sudo ifup wlan1
sudo ifconfig bond0 # no DHCP example, so I've not idea how to get this running :( just providing a way to reproduce the error message (which should in no case be shown with its current content)
sudo ifenslave bond0 wlan0 wlan1
</code>

I'm not requesting support here, just to fix the bug which should make the former unnecessary anyway :)

[Marcus Furlong (furlongm)]

Also seeing this message. Can it be ignored?

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Louis Bouchard (louis)]

As it turns out, this highlights a bug in the ifenslave script.

Starting with kernel 3.13 (commit http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=5831d66e8097aedfa3bc35941cf265ada2352317), /sys/class/net/{bond}/slave_{ifname} (notice the slave_) no longer exist.

I am checking with a colleague to confirm but from what I can tell, it has been replaced by /sys/class/net/{bond}/lower_{ifname}.

If I am exact, I will bring up the bug to the author with a fix that seems to work on my side. Once that fix makes it into Wily, I will SRU to Trusty.

[Louis Bouchard (louis)]

For info, the ambiguous message only appears when using ifenslave on a bond that is already configured :

# ifconfig eth1 up
# ifconfig eth2 up
# ifconfig bond0 up
# ifenslave bond0 eth1 eth2 <<<< No error on the first execution
# ifenslave bond0 eth1 eth2
sh: echo: I/O error
eth1: could not add interface

[Louis Bouchard (louis)]

I have just opened a bug on the debian ifenslave package : http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=791906 with a proposed patch that seems to fix the issue.

Will wait to see if the maintainer responds.

[Louis Bouchard (louis)]

Quick update : the fix is now in Debian ifenslave-2.7 . It needs to be merged in Wily so we can backport the fix to trusty.

[Louis Bouchard (louis)]

The fixed ifenslave-2.7 is now in Wily. Marking it fixed released. Next step is to SRU to Trusty

[Eric Desrochers (slashd)]

debdiff from the fix in Wily

[Eric Desrochers (slashd)]

Here is a new copy of my .debdiff with the adjustment suggested by the SRU Team.

Thanks !

[Brian Murray (brian-murray)]

I've uploaded this to the Trusty SRU queue for review by the SRU team.

[Chris J Arges (arges)]

Does this need to be fixed in Vivid as well?

[Chris J Arges (arges)]

Hello Karl-Philipp, or anyone else affected,

Accepted ifenslave into trusty-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/ifenslave/2.4ubuntu1.1 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

[Eric Desrochers (slashd)]

Yes, it need to be fixed in Vivid as well.

Will start the sru process for Vivid soon.

[Eric Desrochers (slashd)]

debdiff for Vivid

[Simon Déziel (sdeziel)]

Verified to work on Trusty, thanks!

[Brian Murray (brian-murray)]

The verification of the Stable Release Update for ifenslave has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.

[Brian Murray (brian-murray)]

I've uploaded this to the queue for the SRU team to review for Vivid.

[Brian Murray (brian-murray)]

The bug task as not closed by janitor because it was about the wrong package name.

[Chris J Arges (arges)]

Hello Karl-Philipp, or anyone else affected,

Accepted ifenslave into vivid-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/ifenslave/2.5ubuntu1.1 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

[Ubuntu Foundations Team Bug Bot (crichton)]

The fix for this bug has been awaiting testing feedback in the -proposed repository for vivid for more than 90 days. Please test this fix and update the bug appropriately with the results. In the event that the fix for this bug is still not verified 15 days from now, the package will be removed from the -proposed repository.

[Launchpad Janitor (janitor)]

This bug was fixed in the package ifenslave - 2.5ubuntu1.1

---------------
ifenslave (2.5ubuntu1.1) vivid; urgency=medium

  * Fix change in /sys/class/net/{bond}/slave_ naming after kernel v3.13
    (Closes: #791906)

    Starting in kernel 3.13, slave_{ifname} is no longer present and
    has been replaced by lower_{ifname}. Take this change into account
    to avoid ambiguous error when ifenslave is run with existing bond
    defined. (LP: #1326854)

 -- Eric Desrochers <email address hidden> Wed, 12 Aug 2015 14:08:15 -0400