#1895 PR merged
: Avoid false alarm during md5sum verification (related to issue1883)¶
Labels: enhancement
, fixed / solved / done
, minor bug
jsmeix opened issue at 2018-08-08 15:36:¶
-
Type: Enhancement
-
Impact: Normal
Technically the impact is zero (only false alarm)
but to the user the impact is 'normal' because false alarm confuses the user.
I think essentially false alarm is like a WARNING message, cf.
https://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html -
Reference to related issue (URL):
https://github.com/rear/rear/issues/1883#issuecomment-409875733 -
How was this pull request tested?
By me on my SLES12 KVM system. -
Brief description of the changes in this pull request:
Now also exclude all regular files in /dev/
(device nodes get already excluded by 'find -type f')
because sometimes it could happen that there are
regular files in /dev/ in ROOTFS_DIR
which won't get copied into the recovery system
so that those regular files in /dev/ are missing
when etc/scripts/system-setup tries to verify their md5sums,
for example see
https://github.com/rear/rear/issues/1883#issuecomment-409875733
and
https://github.com/rear/rear/issues/1891#issue-347952166
that both read (excerpts):
md5sum: ./dev/.SRC-Semaphore: No such file or directory ./dev/.SRC-Semaphore: FAILED open or read md5sum: WARNING: 1 listed file could not be read
where /dev/.SRC-Semaphore seems to be related to
IBM Reliable Scalable Cluster Technology (RSCT), see
http://www-01.ibm.com/support/docview.wss?uid=isg1IV35736
Additionally /etc/udev/rules.d/70-persistent-net.rules is always
excluded
to avoid false alarm because it seems it can be modified
even before the md5sum verification runs, see
https://github.com/rear/rear/issues/1883#issuecomment-409875733
that reads (excerpt):
./etc/udev/rules.d/70-persistent-net.rules: FAILED md5sum: WARNING: 1 computed checksum did NOT match
Nice WARNING messages from md5sum ;-)
jsmeix commented at 2018-08-08 15:41:¶
@rear/contributors
if there are no objections from you I would like to "just merge" it
tomorrow
because things work well for me and I cannot imagine regressions
(of course the one who makes a change can never imagine regressions ;-)
[Export of Github issue for rear/rear.]