Aleksey Kamensky wrote:
> And on the other hand saying "not an issue because no one else does it this
> way" is not a good approach IMO as it only stimulates copying and does not
> help those offering improvements.
Feel free to submit a patch improving the situation.
--
-- Howard Chu
CTO, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc/
Chief Architect, OpenLDAP http://www.openldap.org/project/
Full_Name: Pietro Cerutti
Version: lmdb 0.9.21
OS: FreeBSD 11.1
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (69.191.176.33)
This patch allows for a finer-grained control over which FreeBSD versions
support fdatasync. This allows us to get rid of a custom patch in our ports.
--- mdb.c.orig 2017-06-01 16:51:10 UTC
+++ mdb.c
@@ -125,6 +125,8 @@ typedef SSIZE_T ssize_t;
# define MDB_FDATASYNC fsync
#elif defined(ANDROID)
# define MDB_FDATASYNC fsync
+#elif defined(__FreeBSD_version) && __FreeBSD_version < 1101000
+# define MDB_FDATASYNC fsync
#endif
#ifndef _WIN32
--94eb2c075988d198060559c39077
Content-Type: text/plain; charset="UTF-8"
On Wed, Sep 20, 2017 at 7:28 PM, Howard Chu <hyc(a)symas.com> wrote:
> aleksey.kamensky(a)gmail.com wrote:
>
>> Full_Name: Alexey Kamenskiy
>> Version: 2.4.45
>> OS: CentOS Linux release 7.3.1611 (Core)
>> URL: ftp://ftp.openldap.org/incoming/
>> Submission from: (NULL) (220.248.35.50)
>>
>>
>> Setting up fresh 2.4.45 with MDB noticed that MDB does allow to create
>> multiple
>> databases in the same directory (noticed by mistake of creating multiple
>> MDBs in
>> default /var/lib/ldap).
>>
>> It shows no error and no any warning. After this ALL of the databases are
>> writeable and readable. On reading the results returned for all 3
>> databases (and
>> not for the DIT specified in ldapsearch -b).
>>
>
> None of the other backends protect from this either. (Nor can they, since
> any individual backend doesn't know anything about any other backend's
> configuration.) Not a bug.
You are right that other backends (I am talking here hdb/bdb) do not
explicitly check for this, but those backends will produce error on attempt
to read/write into database created in the directory where DB already
exists. LMDB just works in this case and returns bad results if try to read.
Even if not intentionally those errors serve as a reminder that one should
not create two DBs in same directory.
And on the other hand saying "not an issue because no one else does it this
way" is not a good approach IMO as it only stimulates copying and does not
help those offering improvements.
--94eb2c075988d198060559c39077
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr"><br><div class=3D"gmail_extra"><br><div class=3D"gmail_quo=
te">On Wed, Sep 20, 2017 at 7:28 PM, Howard Chu <span dir=3D"ltr"><<a hr=
ef=3D"mailto:hyc@symas.com" target=3D"_blank">hyc(a)symas.com</a>></span> =
wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bord=
er-left:1px #ccc solid;padding-left:1ex"><a href=3D"mailto:aleksey.kamensky=
@gmail.com" target=3D"_blank">aleksey.kamensky(a)gmail.com</a> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
Full_Name: Alexey Kamenskiy<br>
Version: 2.4.45<br>
OS: CentOS Linux release 7.3.1611 (Core)<br>
URL: <a href=3D"ftp://ftp.openldap.org/incoming/" rel=3D"noreferrer" target=
=3D"_blank">ftp://ftp.openldap.org/incomin<wbr>g/</a><br>
Submission from: (NULL) <a href=3D"tel:%28220.248.35.50" value=3D"+12202483=
550" target=3D"_blank">(220.248.35.50</a>)<br>
<br>
<br>
Setting up fresh 2.4.45 with MDB noticed that MDB does allow to create mult=
iple<br>
databases in the same directory (noticed by mistake of creating multiple MD=
Bs in<br>
default /var/lib/ldap).<br>
<br>
It shows no error and no any warning. After this ALL of the databases are<b=
r>
writeable and readable. On reading the results returned for all 3 databases=
(and<br>
not for the DIT specified in ldapsearch -b).<br>
</blockquote>
<br>
None of the other backends protect from this either. (Nor can they, since a=
ny individual backend doesn't know anything about any other backend'=
;s configuration.) Not a bug.</blockquote><div><br></div><div>You are right=
that other backends (I am talking here hdb/bdb) do not explicitly check fo=
r this, but those backends will produce error on attempt to read/write into=
database created in the directory where DB already exists. LMDB just works=
in this case and returns bad results if try to read.</div><div><br></div><=
div>Even if not intentionally those errors serve as a reminder that one sho=
uld not create two DBs in same directory.</div><div><br></div><div>And on t=
he other hand saying "not an issue because no one else does it this wa=
y" is not a good approach IMO as it only stimulates copying and does n=
ot help those offering improvements.</div><div>=C2=A0</div></div></div></di=
v>
--94eb2c075988d198060559c39077--
Full_Name: Dieter Kluenter
Version:
OS: OpenSUSE
URL: ftp://ftp.openldap.org/incoming/dkluenter-20-07-17-autoca.patch
Submission from: (NULL) (93.214.247.185)
test066-autoca aborts with
adding new entry "cn=module,cn=config"
adding new entry "olcOverlay=autoca,olcDatabase={1}mdb,cn=config"
dn: cn=localhost,ou=Servers,dc=example,dc=com
userCertificate;binary:
userPrivateKey;binary:
ldap_start_tls: Connect error (-11)
additional info: TLS: hostname does not match CN in peer certificate
this small patch solves this error
aleksey.kamensky(a)gmail.com wrote:
> Full_Name: Alexey Kamenskiy
> Version: 2.4.45
> OS: CentOS Linux release 7.3.1611 (Core)
> URL: ftp://ftp.openldap.org/incoming/
> Submission from: (NULL) (220.248.35.50)
>
>
> Setting up fresh 2.4.45 with MDB noticed that MDB does allow to create multiple
> databases in the same directory (noticed by mistake of creating multiple MDBs in
> default /var/lib/ldap).
>
> It shows no error and no any warning. After this ALL of the databases are
> writeable and readable. On reading the results returned for all 3 databases (and
> not for the DIT specified in ldapsearch -b).
None of the other backends protect from this either. (Nor can they, since any
individual backend doesn't know anything about any other backend's
configuration.) Not a bug.
--
-- Howard Chu
CTO, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc/
Chief Architect, OpenLDAP http://www.openldap.org/project/
Full_Name: Alexey Kamenskiy
Version: 2.4.45
OS: CentOS Linux release 7.3.1611 (Core)
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (220.248.35.50)
Setting up fresh 2.4.45 with MDB noticed that MDB does allow to create multiple
databases in the same directory (noticed by mistake of creating multiple MDBs in
default /var/lib/ldap).
It shows no error and no any warning. After this ALL of the databases are
writeable and readable. On reading the results returned for all 3 databases (and
not for the DIT specified in ldapsearch -b).
--On Friday, July 07, 2017 8:45 AM +0000 djkurtz(a)google.com wrote:
> Full_Name: Daniel Jonathan Kurtz
> Version: 2.4.45
> OS: linux
> URL: ftp://ftp.openldap.org/incoming/
> Submission from: (NULL) (2401:fa00:1:b:dcfc:ce39:b80:8f26)
Out of curiosity, what OS were you compiling on that didn't have
/dev/urandom available?
--Quanah
--
Quanah Gibson-Mount
Product Architect
Symas Corporation
Packaged, certified, and supported LDAP solutions powered by OpenLDAP:
<http://www.symas.com>
Full_Name: Quanah Gibson-Mount
Version: HEAD
OS: N/A
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (47.208.148.239)
Currently test059 shows how replication of cn=config can work for a replica,
using back-ldif and standard syncrepl, but we do not have a corresponding script
using delta-syncrepl. We should similarly have a corresponding test script for
cn=config MMR using delta-syncrepl for the cn=config db.
> IMO using recursive mutexes means your code is broken. We introduced these for
> accesslog.c but in fact we could avoid them at zero cost. Also I don't see the
> relevance of libevent to this discussion. We use our own event mechanism and
> it is more efficient than libevent.
libevent is a dependency for the load balancer that I intend to propose
for integration into the project after all the relevant dependencies
have come in.
There is a new version of this patch that provides an implementation on
each platform or defers to the existing one (as per each platform's
documentation). Untested except on POSIX and most of them seem pretty
arcane anyway.
ftp://ftp.openldap.org/incoming/Ondrej-Kuznik-20170918-ITS8638-libldap_r-re…
--
OndÅ™ej KuznÃk
Senior Software Engineer
Symas Corporation http://www.symas.com
Packaged, certified, and supported LDAP solutions powered by OpenLDAP