| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
With unique indexes, any rename of a record that has an attribute that
is uniquely indexed needs to be done as a delete followed by an add,
otherwse you'll get an error that the attribute value already exists.
|
|
When a attribute is marked unique we know that if we find a match
it will be the only possible match. This means that in a list of
subtrees connected by an &, it is best to first load the index values
for the unique entries, as if they find something then we know we
won't have to look any further.
This helps with searches like this:
(&(objectclass=user)(samaccountname=tridge))
the old code would first have loaded the very large index for the
objectclass=user attribute, and then loaded the single entry for
samaccountname=tridge. Now we load the samaccountname=tridge entry
first, notice that it gives us a single result, and stop, thereby
skipping the load of the objectclass=user index record completely.
|
|
When a attribute is marked as LDB_ATTR_FLAG_UNIQUE_INDEX then attempts
to add a 2nd record that has the same attribute value for this
attribute as another record will fail.
This provides a much more efficient mechanism for ensuring that
attributes like objectGUID are unique
|
|
When looking at performance problems with ldb it can be useful to see
which searches causes unindexed full searches. This makes it easy to
enable that.
|
|
one-level indexing was not always effective due to some broken logic
in the indexing code. This change means that if normal indexing fails,
we can still fall back on one-level indexing.
This reduces the number of full unindexed searches in s4 quite a lot
|
|
|
|
This op will be used by the partition module to give us good
transaction semantics across the 4 partitions that sam.ldb uses.
|
|
Extended operations return was not going thorugh the same patch as others
leaving the ctx around. Plus we were neutralizing the spy too early so that it
didn't set the request_terminated flag as it should have.
This should hopefully fix the failures in the build farm.
|
|
In some code paths ltdb_context was still referenced even after we were returned
an error by one of the callbacks. Because the interface assumes that once an
error is returned the ldb_request may be freed, and because the ltdb_context was
allocated as a child of the request, this might cause access to freed memory.
Allocate the ltdb_context on ldb, and keep track of what's going on with the
request by adding a spy children on it. This way even if the request is freed
before the ltdb_callback is called, we will safely free the ctx and just quietly
return.
|
|
do not reference it from ldb.h
|
|
behavior anyway, and given we can only have one transaction active per
ldb context this is the only sane model we can support.
Fix ldb_tdb transactions, we could return back with an error with neither
committing nor canceling the actual tdb transaction in some error paths
within the ltdb commit and cancel transaction paths.
Added also some debugging to trace what was going on.
|
|
Separate again the public from the private headers.
Add a new header specific for modules.
Also add service function for modules as now ldb_context and ldb_module are
opaque structures for them.
|
|
metze
|
|
list=""
list="$list event_context:tevent_context"
list="$list fd_event:tevent_fd"
list="$list timed_event:tevent_timer"
for s in $list; do
o=`echo $s | cut -d ':' -f1`
n=`echo $s | cut -d ':' -f2`
r=`git grep "struct $o" |cut -d ':' -f1 |sort -u`
files=`echo "$r" | grep -v source3 | grep -v nsswitch | grep -v packaging4`
for f in $files; do
cat $f | sed -e "s/struct $o/struct $n/g" > $f.tmp
mv $f.tmp $f
done
done
metze
|
|
|
|
It changes some "return 0" in "return LDB_SUCCESS"
|
|
This also asserts that it is used only for index records.
Andrew Bartlett
|
|
Signed-off-by: Stefan Metzmacher <metze@samba.org>
|
|
ldb indexing can cause huge files, and huge memory usage. This
experiment allows us to keep indexes in memory during a transaction,
then to write the indexes to disk when the transaction completes. The
result is that the db is much smaller (we have seen improvements of
about 100x in file size) and memory usage during large transactions is
also greatly reduced
Note that this patch uses the unusual strategy of putting pointers
into a ldb (and thus into a tdb). This works because the pointers are
only there during a transaction, so the pointers are not exposed to
any other users of the database. The pointers allow us to avoid some
really bad allocation problems with tdb record allocation during the
re-indexing.
|
|
re-indexing in ldb is triggered on any modification to the @ATTRIBUTES
or @INDEXLIST records. This happens to produce a worst-case
fragmentation of the database, as all @INDEX records are deleted then
re-created. By repacking after re-indexing we ensure that the database
ends up without extreme fragmentation.
|
|
(Even if the callback takes some time, this isn't a ldb_tdb timeout
any more)
Andrew Bartlett
|
|
remove some unused functions.
|
|
it should always have been. Make it also async so that it is not a special case.
|
|
|
|
|
|
This commit applies some cosmetic corrections for the LDB backend modules.
|
|
|
|
(This used to be commit 3c058f50cc3b91d540feb51fb698d90565b2b7c9)
|
|
Remove trailing spaces and try to fit 80 columns where possible
(This used to be commit edf6b77a1314d8f91839836855ae049393f73aca)
|
|
(This used to be commit d62f2bcc85c13605c133db250e0a86d2d6ccc481)
|
|
(This used to be commit b1a7810f3e70f9a831d9b8e85d531e448072adaf)
|
|
Andrew Bartlett
(This used to be commit 0016231edd514e8db620bafc44ce877fcac19ed9)
|
|
tmp_ctx,
and also makes code more readable and debuggable.
Eliminate silly parenthesys.
Simo.
(This used to be commit 166195b487ffa51933f772a56f47f7f0d4c867eb)
|
|
(This used to be commit a71e40ab5dc7f0a5f6d2d1eb930f378cd143b186)
|
|
(This used to be commit 8625cd403ba3a7d2b1b1fccfeb5efd7e21de0135)
|
|
bug 5090 by Matthias Dieter Wallnöfer <mwallnoefer@yahoo.de>
Andrew Bartlett
(This used to be commit cc2d0c9f15a9c687d212df14d8ffb6c60ad15242)
|
|
Andrew Bartlett
(This used to be commit d7e65da56454bc7721083e0aa7fa2e9c47f2b79d)
|
|
Andrew Bartlett
(This used to be commit 05cc2a7d966a10f1f111d7bae3261e1087fdffe6)
|
|
Use the checkBaseOnSearch attribute to control if we should check the
base DN on search requests.
Also ensure we honour any errors in searching, not just errors in the
supplied 'done' callback.
Andrew Bartlett
(This used to be commit deaac92f439ef001bfe052df170d6e34e8ba5845)
|
|
Jeremy.
(This used to be commit 52b26645b04a9c5fb70e7b869b60c9157f821d50)
|
|
transaction. When we are in a transaction then we could be in a top level modify operation (such as rename), so we must use a writeable traverse so that the async callbacks can do the modifies while the search is progressing.
- don't do the lockall operation on the tdb during a ldb search if in
a transaction, as this would prevent modifies by callbacks as well
(This used to be commit aa9ab431e071882f42ebc882e809ae1d4b8778d4)
|
|
Subclass support was designed to avoid needing to spell out the full
list of objectClasses that an entry was in. However, Samba4 now
enforces this restriction in the objectClass module, and the way
subclass matching was handled was complex and counter-intuitive in my
opinion (and did not match LDAP).
Andrew Bartlett
(This used to be commit f5ce04b904e14445a2a7e7f92e7e1f64b645c6f2)
|
|
metze
(This used to be commit 8f2db3c130ce85d38f805836a7df039822ede066)
|
|
rename of ldb entries for a case change (only).
I've modified the testsuite to verify this.
Andrew Bartlett
(This used to be commit 9cccd00dac44dd9152ec03cecf5ffac24f918445)
|
|
(This used to be commit 917bd737cb07817664d9088860588d47525f5ff8)
|
|
(This used to be commit 08bb1ef643ab906f1645cf6f32763dc73b1884e4)
|
|
(This used to be commit 40c0919aaa9c1b14bbaebb95ecce53eb0380fdbb)
|
|
(This used to be commit 84b468b2f8f2dffda89593f816e8bc6a8b6d42ac)
|
|
test to prove the behaviour of LDAP renames etc.
Fix LDB to return correct error code when failing to rename one DN
onto another.
Andrew Bartlett
(This used to be commit 3f3da9c4710b7752ed97f55c2fc3d32a63d352af)
|
|
We can't ever allow duplicates, even if the client sends them
Andrew Bartlett
(This used to be commit 10277f27246b9e16ed36fb72eb4c318b43cb9395)
|
