Has anyone experienced a problem building with Mock where the buildroot
exists, but can't be found/accessed by the script? It seems like this is
a permission problem of some sort, as the directories it is searching
for exist but can't be written to by the mock group.
Checking /var/lib/mock/, the sticky-bit is set on the mock buildroot
structure, but is not propagating to directories like
/var/lib/mock/epel-6-*/root/var/lib/rpm/ . I get a smattering of tiny
errors laced through mock's build log related to non-existant
directories which do exist, but belong to the root group, not mock (I'm
assuming this is the key mistake -- but I don't remember if everything
during the build is supposed to be grouped to mock or not...?), and then
finally this:
--
/etc/init.d/rdisc: line 16: /etc/sysconfig/network: No such file or
directory
Traceback (most recent call last):
File "/usr/lib/python2.6/site-packages/yum/rpmtrans.py", line 395, in
callback
File "/usr/lib/python2.6/site-packages/yum/rpmtrans.py", line 499, in
_unInstStop
File "/usr/lib/python2.6/site-packages/yum/rpmtrans.py", line 246, in
_scriptout
File "/usr/lib/python2.6/site-packages/yum/history.py", line 863, in
log_scriptlet_output
File "/usr/lib/python2.6/site-packages/yum/sqlutils.py", line 168, in
executeSQLQmark
sqlite3.OperationalError: unable to open database file
error: python callback <bound method RPMTransaction.callback of
<yum.rpmtrans.RPMTransaction instance at 0x270be18>> failed, aborting!
/var/lib/mock/epel-6-x86_64/root/var/lib/rpm: No such file or directory
--
Which is where the builds completely fail because it couldn't
locate/create a correct buildroot.
I've never had any problem with Mock before, and haven't seen a problem
like this for the last several years with it -- so I'm assuming this
isn't affecting others.
Any insights would be greatly appreciated -- this has put passable test
builds for EPEL and some side projects like LibreCAD on hold for me,
which really chafes.
|