After chatting with xnox on IRC, he discovered that the output of
catting /proc/modules in 4.15 now depends on who is doing the catting.
So `cat /proc/modules` as a normal user will return null references
while `sudo cat /proc/modules` will return the actual memory offsets,
accurately.
For example:
ubuntu@xwing:~$ sudo cat /proc/modules |grep e1000e
e1000e 249856 0 - Live 0xffffffffc0225000
ptp 20480 2 igb,e1000e, Live 0xffffffffc00a9000
ubuntu@xwing:~$ cat /proc/modules |grep e1000e
e1000e 249856 0 - Live 0x (null)
ptp 20480 2 igb,e1000e, Live 0x (null)
So for the cert tooling, this should be a simple fix then, just to
ensure that the module script runs as root.
** Changed in: plainbox-provider-checkbox (Ubuntu)
Status: Confirmed => Triaged
** Changed in: plainbox-provider-checkbox (Ubuntu)
Status: Triaged => In Progress
** Changed in: plainbox-provider-checkbox (Ubuntu)
Status: In Progress => Fix Committed
** Changed in: plainbox-provider-checkbox (Ubuntu)
Status: Fix Committed => In Progress
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1757143
Title:
/proc/modules has Null references causing python parsing issues
Status in linux package in Ubuntu:
Incomplete
Status in plainbox-provider-checkbox package in Ubuntu:
In Progress
Bug description:
EDIT: this originally looked like some sort of regression in the
module_resource script of plainbox-provider-resource, however, on
diving in a bit it seems the root cause is that /proc/modules in 4.15
is different than in previous kernels.
This is a line from 4.15's /proc/modules data:
e1000e 249856 0 - Live 0x (null)
And this is the same module info from 4.13:
e1000e 249856 0 - Live 0x0000000000000000
That null character at the end appears to be causing the script to
choke.
For that reason, because /proc/modules has somehow changed to include
null references (0xNULL is invalid, where 0x0000000 is parsable), I've
added a kernel task to this bug.
Original Summary:
I ran into this issue (and came across some further glitches) while
running canonical-certification-server on Ubuntu Server 18.04 on s390x
(having 16.04 full selected in canonical-certification-server user
interface):
$ /usr/lib/plainbox-provider-resource-generic/bin/module_resource
Traceback (most recent call last):
File "/usr/lib/plainbox-provider-resource-generic/bin/module_resource",
line 73, in <module>
sys.exit(main())
File "/usr/lib/plainbox-provider-resource-generic/bin/module_resource",
line 62, in main
for module in modules:
File "/usr/lib/plainbox-provider-resource-generic/bin/module_resource",
line 57, in get_modules
yield get_module(line)
File "/usr/lib/plainbox-provider-resource-generic/bin/module_resource",
line 49, in get_module
"offset": int(offset, 16)}
ValueError: invalid literal for int() with base 16: '0x'
result file is attached to the ticket, as well as the console log ...
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1757143/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
