Bug 841937

Created attachment 599417 [details]
abrt output from exception

Please do not compress text attachments.  Bugzilla cannot search through them when you do this and it makes it harder for us to read them.

Created attachment 599434 [details]
abrt file uncompressed

anaconda 13.21.176 exception report
Traceback (most recent call first):
  File "/usr/lib/python2.6/site-packages/yum/sqlutils.py", line 168, in executeSQLQmark
    return cursor.execute(query, params)
  File "/usr/lib/python2.6/site-packages/yum/history.py", line 845, in trans_data_pid_beg
    VALUES (?, ?, ?)""", (self._tid, pid, state))
  File "/usr/lib/python2.6/site-packages/yum/history.py", line 935, in beg
    self.trans_data_pid_beg(pid, state)
  File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 1477, in runTransaction
    self.skipped_packages, rpmdb_problems, cmdline)
  File "/usr/lib/anaconda/yuminstall.py", line 1101, in _run
    self.runTransaction(cb=cb)
  File "/usr/lib/anaconda/yuminstall.py", line 1083, in run
    if self._run(instLog, cb, intf) == DISPATCH_BACK:
  File "/usr/lib/anaconda/yuminstall.py", line 1865, in doInstall
    rc = self.ayum.run(self.instLog, cb, anaconda.intf, anaconda.id)
  File "/usr/lib/anaconda/backend.py", line 242, in doInstall
    return anaconda.backend.doInstall(anaconda)
  File "/usr/lib/anaconda/dispatch.py", line 210, in moveStep
    rc = stepFunc(self.anaconda)
  File "/usr/lib/anaconda/dispatch.py", line 126, in gotoNext
    self.moveStep()
  File "/usr/lib/anaconda/gui.py", line 1388, in nextClicked
    self.anaconda.dispatch.gotoNext()
  File "/usr/lib/anaconda/iw/progress_gui.py", line 79, in renderCallback
    self.intf.icw.nextClicked()
  File "/usr/lib/anaconda/gui.py", line 1409, in handleRenderCallback
    self.currentWindow.renderCallback()
IntegrityError: trans_data_pkgs.tid may not be NULL

Local variables in innermost frame:
cursor: <sqlite3.Cursor object at 0x7ff014735348>
query: INSERT INTO trans_data_pkgs
                         (tid, pkgtupid, state)
                         VALUES (?, ?, ?)
params: (None, None, 'Install')

...I don't see how this can be related to Turkish, this is just generic history code which takes the current transaction ID and the current "history pkg id" and creates a new entry for them. But both Ids are None ... which should be impossible (but if it happens should happen for every transaction).
 Moving to anaconda as I hope they have a clue what could be different.

The bug report gives no indication whether or not language is the determining factor in this error appearing.  If you install in some other language, is it reproducible?

Indeed when you select the Turkish language, the error occurs consistently. AFAIK it's the determining factor, but I haven't tested many other languages.

It obviously does *not* happen with English, so language selection at least is a determinig factor.

This request was not resolved in time for the current release.
Red Hat invites you to ask your support representative to
propose this request, if still desired, for consideration in
the next release of Red Hat Enterprise Linux.

This request was evaluated by Red Hat Product Management for
inclusion in a Red Hat Enterprise Linux release.  Product
Management has requested further review of this request by
Red Hat Engineering, for potential inclusion in a Red Hat
Enterprise Linux release for currently deployed products.
This request is not yet committed for inclusion in a release.

I don't think the problem is within anaconda. None of the variables that are being passed off to yum functions contain bad values.

I'm not entirely sure what the problem is exactly, but running an English and Turkish install side by side, I found where the problem most likely lies.

The traceback notes that the sql query being run in the "trans_data_pid_beg" function is trying to pass null values: self._tid, pid, state

So "self._tid" and "pid" are None when Turkish is selected, for some reason.

In yum/history.py:
def beg(self, rpmdb_version, using_pkgs, txmbrs, skip_packages=[],
            rpmdb_problems=[], cmdline=None):
        cur = self._get_cursor()
        if cur is None:
            return
        res = executeSQL(cur,
                         """INSERT INTO trans_beg
                            (timestamp, rpmdb_version, loginuid)
                            VALUES (?, ?, ?)""", (int(time.time()),
                                                    str(rpmdb_version),
                                                    yum.misc.getloginuid()))
        self._tid = cur.lastrowid
self._tid ends up with a value of 1 for English, and None for Turkish, so there is the first variable being set to a null value.

Then a few lines further in the same function:
        ....
        for txmbr in txmbrs:
            pid   = self.pkg2pid(txmbr.po)
            state = self.txmbr2state(txmbr)
            self.trans_data_pid_beg(pid, state)

pid is getting set to None for Turkish and receives int values for English.

It's a python bug.. http://bugs.python.org/issue13099

Created attachment 783251 [details]
backport of upstream fix

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHSA-2013-1582.html