samba.git/python/samba/dbchecker.py, branch talloc-2.1.10

dbcheck: Improve dbcheck to find (and may fix) dangling msDS-RevealedUsers

2017-03-13T04:10:12+00:00

We cannot add missing backlinks because of the duplicate checking. There
seems to be no trivial way to add the bypass.

Signed-off-by: Garming Sam 
Reviewed-by: Andrew Bartlett

dbchecker: Stop ignoring linked cases where both objects are alive

2017-02-23T22:58:21+00:00

Previously, this did nothing and the code was both untested and unused.

Removes the knownfail entry for dbcheck.

Signed-off-by: Garming Sam 
Reviewed-by: Andrew Bartlett 
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12600

dbcheck: Do not regard old one-way-links as errors

2017-02-13T02:39:23+00:00

Samba does not maintain one way links when the target is deleted or renamed
so do not fail dbcheck because of such links, but allow them to be updated.

This matters because administrators and make test expect that normal Samba
operation do NOT cause the database to become corrupt, and any error from
dbcheck tends to trigger alarms (or test failures).

If an object pointed at by a one way link is renamed or deleted in normal
operations (such as intersiteTopologyGenerator pointing at a demoted DC),
or make test, then this could trigger.

Signed-off-by: Andrew Bartlett 
Reviewed-by: Garming Sam 
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12577

dbcheck: Correct message for orphaned backlinks

2016-11-22T01:10:16+00:00

The backlink name is in attrname, not in link_name

Signed-off-by: Andrew Bartlett 
Reviewed-by: Garming Sam 
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12297

dbcheck: Be more careful with link checks

2016-11-22T01:10:16+00:00

Here we are more careful when checking links, flagging errors only
when a non-deleted forward link appears incorrect.  In particular, we
trust the GUID more than we trust the name, as otherwise we can get
caught out if there is a swap of names, (the link should follow the
swap, staying on the same target GUID).

Signed-off-by: Andrew Bartlett 
Reviewed-by: Garming Sam 
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12297

dbcheck: Correctly initialise keep_transaction in missing_parent test

2016-11-04T03:41:19+00:00

Otherwise there is no point to this variable, we are trying to work out
if the subsequent modify succeded

Signed-off-by: Andrew Bartlett 
Reviewed-by: Garming Sam 

BUG: https://bugzilla.samba.org/show_bug.cgi?id=9954

dbcheck: confirm RID Set presence and consistency

2016-11-04T03:41:19+00:00

Signed-off-by: Andrew Bartlett 
Reviewed-by: Garming Sam 

BUG: https://bugzilla.samba.org/show_bug.cgi?id=9954

dbcheck: Abandon dbcheck if we get an error during a transaction

2016-08-29T10:46:21+00:00

Otherwise, anything that the transaction has already done to the DB will be left in the DB
even despite the failure.  For example, if a fix wrote to the DB, but then failed a post-write
check, then the fix will not be unrolled.

This is because we do not have nested transactions in TDB.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=12178

Signed-off-by: Andrew Bartlett 
Reviewed-by: Stefan Metzmacher 

Autobuild-User(master): Stefan Metzmacher 
Autobuild-Date(master): Mon Aug 29 12:46:21 CEST 2016 on sn-devel-144

dbcheck: Add a rule regarding replica locations

2016-07-21T04:37:08+00:00

This fixes any RW DCs with repsFrom without the corresponding link. On
any RODC, this just reports an error (and doesn't fix it).

(the knownfail entry is also now removed)

BUG: https://bugzilla.samba.org/show_bug.cgi?id=9200

Signed-off-by: Garming Sam 
Reviewed-by: Andrew Bartlett

dbcheck: Split out valid stale DN links and invalid ones

2016-07-15T08:01:30+00:00

Signed-off-by: Garming Sam 
Reviewed-by: Andrew Bartlett