Re: 389-Directory/1.3.1.6 cannot setup replica

[Noriko Hosoi <nhosoi@xxxxxxxxxx>]

Thahk you for your update.

Others should be ok, but this looks suspicious.  We recommend to put it back to on.
set nsslapd-pluginbetxn off

--noriko

On 11/05/2014 12:43 PM, 陳含林 wrote:

Thanks for your replay!

After change  nsDS5ReplicaTransportInfo from TLS to LDAP,  total update finished in 5 sec  and succeeded. This should be a workaround for my case.

While nsDS5ReplicaTransportInfo set to TLS, i have tried 

  set nsslapd-maxbersize to 2147483647

  set nsslapd-cachememsize to 2147483648

  set nsslapd-pluginbetxn off

  set export LDAPTLS_REQCERT=never  in /etc/sysconfig/dirsrv

all of above not work.

2014-11-06 4:26 GMT+08:00 Noriko Hosoi <nhosoi@xxxxxxxxxx>:

Hello Andrey,

On 11/05/2014 11:53 AM, Ivanov Andrey (M.) wrote:

Hi,

i'm having the same problem. I'm in the process of migration from our 389DS v1.2.10.25/CentOS5 to 389DS on CentOS 7. Everything is working fine on standalone servers but the replication (especially online initialization). It stucks _each time_ during the online initialization  with SSL/TLS (and sometimes without SSL/TLS). And exactly with the same error messages as you describe.

The network problems in my case are excluded - i used both virtual machines on the same ESXi in the same network and/or physical servers, the results are the same.

I've tried compiling all the branches latest available (tags 1.3.2.24, 1.3.1.22, 1.3.3.5). In all the cases the result was the same. The server pushing the updates just gets stuck at some random number of entries sent to consumer (we have ~30000 entries, it gets stuck at random somewhere from 1200 to 25000 entries, the entries stuck have nothing particular in size, it's completely random).

When you tried the versions "(tags 1.3.2.24, 1.3.1.22, 1.3.3.5)", it does not report any extra errors in the error log? E.g.,

Incoming BER Element was @@@ bytes, max allowable is ### bytes. Change the nsslapd-maxbersize attribute in cn=config to increase.

Next time it happens, could it be possible to get the stacktraces from the hung server?

http://www.port389.org/docs/389ds/FAQ/faq.html#debugging-hangs

1.2.10.24 compiled with compat-db-4.7 on CentOS 7 has the least of these problems (and the initial replication is 10 times faster - it takes 8 seconds instead of 80 for 1.3.x!). I've been using 1.2.10.24 on CentOS5 compiled with mozilla ldap labraries and 1.2.10.23 obn CentOS 7 compiled with opendlap librarires. The first one had no problems at all to push the initial replication, the second one had intermittent problems, but much less than v1.3.x

I've noticed that this problem is getting worse  (or simply appears) if :

* the replica is be of type 3 (multi-master), with replication agreements in both directions

* our schema has several additional attribuites, it may be also important

We improved the schema replication at some time, which should be in 1.3.2.24 and 1.3.3.5.
https://fedorahosted.org/389/ticket/47721

* if the virtual machine has only one CPU. Adding a second CPU increases the number of transferred entries before the initialization gets stuck. So it may me some thread/transaction  contention or deadlock.

* if the replication agreement uses SSL(port 636) or TLS(port389). Using port 389 with LDAP protocol instead of TLS/SSL increases the number of transferred entries before the initialization gets stuck. Sometimes the initialization even ends successfully in this case.

* decreasing nsslapd-db-checkpoint-interval (say, to 5 seconds) also gets the problem worse

This indicates the issue is in the BDB?

Thanks,
--noriko

When the on-line intialization is finished (if it finishes), there are no problems. I think it is related to the volume of data transferred, so small incremental updates do not generate any problem....

If necessary, i will make any debugs/tests - it is a critical element of our infrastructure, so i'd like this problem to be resolved...

Regards,

Andrey IVANOV

---

De: "陳含 林" <laneovcc@xxxxxxxxx>

À: 389-users@xxxxxxxxxxxxxxxxxxxxxxx
Envoyé: Mercredi 5 Novembre 2014 18:01:37
Objet: 389-Directory/1.3.1.6 cannot setup replica

hello all,

I have setup a IDM/freeipa  master  using  CentOS7 ,  and import about 5000  hosts.

then i try to setup a  IDM/freeipa replication server by using  ipa-replica-install.

It seems the total update on replication server   hangs after about 1000+ entries imported.

I try to trigger  a  total update by setting  nsds5beginreplicarefresh, but the result was the same.

Any one help me ?  Thanks!

idm1 is the master, idm2 is the replication server.

master server  logs:

[06/Nov/2014:00:21:48 +0800] - 389-Directory/1.3.1.6 B2014.219.1825 starting up

[06/Nov/2014:00:21:48 +0800] schema-compat-plugin - warning: no entries set up under cn=computers, cn=compat,dc=idc

[06/Nov/2014:00:21:51 +0800] - Skipping CoS Definition cn=Password Policy,cn=accounts,dc=idc--no CoS Templates found, which should be added before the CoS Definition.

[06/Nov/2014:00:21:51 +0800] - Skipping CoS Definition cn=Password Policy,cn=accounts,dc=idc--no CoS Templates found, which should be added before the CoS Definition.

[06/Nov/2014:00:21:51 +0800] - slapd started.  Listening on All Interfaces port 389 for LDAP requests

[06/Nov/2014:00:21:51 +0800] - Listening on All Interfaces port 636 for LDAPS requests

[06/Nov/2014:00:21:51 +0800] - Listening on /var/run/slapd-IDC.socket for LDAPI requests

[06/Nov/2014:00:21:51 +0800] - Entry "uid=admin,ou=people,o=ipaca" -- attribute "krbExtraData" not allowed

[06/Nov/2014:00:40:26 +0800] NSMMReplicationPlugin - agmt="cn=meToidm2.ra.cn.idc" (idm2:389): The remote replica has a different database generation ID than the local database.  You may have to reinitialize the remote replica, or the local replica.

[06/Nov/2014:00:40:26 +0800] NSMMReplicationPlugin - Beginning total update of replica "agmt="cn=meToidm2.ra.cn.idc" (idm2:389)".

replication server  logs:

[06/Nov/2014:00:40:18 +0800] - 389-Directory/1.3.1.6 B2014.219.1825 starting up

[06/Nov/2014:00:40:18 +0800] ipalockout_get_global_config - [file ipa_lockout.c, line 185]: Failed to get default realm (-1765328160)

[06/Nov/2014:00:40:18 +0800] ipaenrollment_start - [file ipa_enrollment.c, line 393]: Failed to get default realm?!

[06/Nov/2014:00:40:18 +0800] - slapd started.  Listening on All Interfaces port 389 for LDAP requests

[06/Nov/2014:00:40:18 +0800] - Listening on All Interfaces port 636 for LDAPS requests

[06/Nov/2014:00:40:18 +0800] - Listening on /var/run/slapd-IDC.socket for LDAPI requests

[06/Nov/2014:00:40:19 +0800] NSMMReplicationPlugin - agmt="cn=meToidm1.ra.cn.idc" (idm1:389): The remote replica has a different database generation ID than the local database.  You may have to reinitialize the remote replica, or the local replica.

[06/Nov/2014:00:40:20 +0800] NSMMReplicationPlugin - multimaster_be_state_change: replica dc=idc is going offline; disabling replication

[06/Nov/2014:00:40:20 +0800] - WARNING: Import is running with nsslapd-db-private-import-mem on; No other process is allowed to access the database

[06/Nov/2014:00:40:40 +0800] - import userRoot: Processed 1581 entries -- average rate 79.0/sec, recent rate 79.0/sec, hit ratio 0%

[06/Nov/2014:00:41:00 +0800] - import userRoot: Processed 1581 entries -- average rate 39.5/sec, recent rate 39.5/sec, hit ratio 0%

[06/Nov/2014:00:41:20 +0800] - import userRoot: Processed 1581 entries -- average rate 26.4/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:41:41 +0800] - import userRoot: Processed 1581 entries -- average rate 19.8/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:42:01 +0800] - import userRoot: Processed 1581 entries -- average rate 15.8/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:42:21 +0800] - import userRoot: Processed 1581 entries -- average rate 13.2/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:42:41 +0800] - import userRoot: Processed 1581 entries -- average rate 11.3/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:43:01 +0800] - import userRoot: Processed 1581 entries -- average rate 9.9/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:43:21 +0800] - import userRoot: Processed 1581 entries -- average rate 8.8/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:43:41 +0800] - import userRoot: Processed 1581 entries -- average rate 7.9/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:44:01 +0800] - import userRoot: Processed 1581 entries -- average rate 7.2/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:44:21 +0800] - import userRoot: Processed 1581 entries -- average rate 6.6/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:44:41 +0800] - import userRoot: Processed 1581 entries -- average rate 6.1/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:45:01 +0800] - import userRoot: Processed 1581 entries -- average rate 5.6/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:45:21 +0800] - import userRoot: Processed 1581 entries -- average rate 5.3/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:45:21 +0800] - import userRoot: Decided to end this pass because the progress rate has dropped below the 50% threshold.

[06/Nov/2014:00:45:21 +0800] - import userRoot: Ending pass number 1 ...

[06/Nov/2014:00:45:22 +0800] - import userRoot: Foreman is done; waiting for workers to finish...

[06/Nov/2014:00:45:22 +0800] - import userRoot: Workers finished; cleaning up...

[06/Nov/2014:00:45:22 +0800] - import userRoot: Workers cleaned up.

[06/Nov/2014:00:45:22 +0800] - import userRoot: Sweeping files for merging later...

[06/Nov/2014:00:45:22 +0800] - WARNING: Import is running with nsslapd-db-private-import-mem on; No other process is allowed to access the database

[06/Nov/2014:00:45:22 +0800] - import userRoot: Sweep done.

[06/Nov/2014:00:45:22 +0800] - import userRoot: Beginning pass number 2

[06/Nov/2014:00:45:42 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 214748285.8/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:46:02 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 107374142.9/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:46:22 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 71582761.9/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:46:42 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 53687071.4/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:47:02 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 42949657.1/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:47:22 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 35791381.0/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:47:43 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 30678326.5/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:48:03 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 26843535.7/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:48:23 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 23860920.6/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:48:43 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 21474828.6/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:49:03 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 19522571.4/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:49:23 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 17895690.5/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:49:43 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 16519098.9/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:50:03 +0800] - import userRoot: Processed 1581 entries (pass 2) -- average rate 15339163.3/sec, recent rate 0.0/sec, hit ratio 0%

[06/Nov/2014:00:50:23 +0800] - import userRoot: Processed 1722 entries (pass 2) -- average rate 0.5/sec, recent rate 43.0/sec, hit ratio 0%

[06/Nov/2014:00:50:26 +0800] - import userRoot: Workers finished; cleaning up...

[06/Nov/2014:00:50:26 +0800] - import userRoot: Workers cleaned up.

[06/Nov/2014:00:50:26 +0800] - import userRoot: Sweeping files for merging later...

[06/Nov/2014:00:50:26 +0800] - WARNING: Import is running with nsslapd-db-private-import-mem on; No other process is allowed to access the database

[06/Nov/2014:00:50:26 +0800] - import userRoot: Sweep done.

[06/Nov/2014:00:50:26 +0800] - import userRoot: Beginning 2-way merge of up to 44 files...

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "aci".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "automountkey".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "carLicense".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "cn".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "displayname".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "entryusn".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "fqdn".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "gidnumber".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "givenName".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "ipakrbprincipalalias".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "ipasudorunas".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "ipasudorunasgroup".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "ipauniqueid".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "krbPrincipalName".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "macAddress".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "mail".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "mailAlternateAddress".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "mailHost".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "managedby".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "manager".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "member".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "memberallowcmd".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "memberdenycmd".

[06/Nov/2014:00:50:26 +0800] - import userRoot: No files to merge for "memberHost".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "memberOf".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "memberservice".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "memberUser".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "nscpEntryDN".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "nsds5ReplConflict".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "nsuniqueid".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "ntUniqueId".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "ntUserDomainId".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "objectclass".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "ou".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "owner".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "secretary".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "seeAlso".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "sn".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "sourcehost".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "telephoneNumber".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "title".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "uid".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "uidnumber".

[06/Nov/2014:00:50:27 +0800] - import userRoot: No files to merge for "uniquemember".

[06/Nov/2014:00:50:27 +0800] - import userRoot: Merging completed in 0 seconds.

[06/Nov/2014:00:50:27 +0800] - import userRoot: Indexing complete.  Post-processing...

[06/Nov/2014:00:50:27 +0800] - import userRoot: Generating numSubordinates complete.

[06/Nov/2014:00:50:27 +0800] - import userRoot: Flushing caches...

[06/Nov/2014:00:50:27 +0800] - import userRoot: Closing files...

[06/Nov/2014:00:50:27 +0800] - import userRoot: Import complete.  Processed 1722 entries in 607 seconds. (2.84 entries/sec)

[06/Nov/2014:00:50:27 +0800] NSMMReplicationPlugin - multimaster_be_state_change: replica dc=idc is coming online; enabling replication

[06/Nov/2014:00:50:27 +0800] - Skipping CoS Definition cn=Password Policy,cn=accounts,dc=idc--no CoS Templates found, which should be added before the CoS Definition.

mapping tree

# extended LDIF

#

# LDAPv3

# base <cn=mapping tree,cn=config> with scope subtree

# filter: (objectclass=*)

# requesting: ALL

#

# mapping tree, config

dn: cn=mapping tree,cn=config

objectClass: top

objectClass: extensibleObject

cn: mapping tree

# dc\3Didc, mapping tree, config

dn: cn=dc\3Didc,cn=mapping tree,cn=config

objectClass: top

objectClass: extensibleObject

objectClass: nsMappingTree

cn: dc=idc

cn: "dc=idc"

nsslapd-state: backend

nsslapd-backend: userRoot

nsslapd-referral: ldap://idm1.ra.cn.idc:389/dc%3Didc

# replica, dc\3Didc, mapping tree, config

dn: cn=replica,cn=dc\3Didc,cn=mapping tree,cn=config

cn: replica

nsDS5Flags: 1

objectClass: top

objectClass: nsds5replica

objectClass: extensibleobject

nsDS5ReplicaType: 3

nsDS5ReplicaRoot: dc=idc

nsds5ReplicaLegacyConsumer: off

nsDS5ReplicaId: 3

nsDS5ReplicaBindDN: cn=replication manager,cn=config

nsState:: AwAAAAAAAADzUlpUAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAAAA==

nsDS5ReplicaName: 6cf48589-650a11e4-90ecc46a-d11f6771

nsds5ReplicaChangeCount: 0

nsds5replicareapactive: 0

# meToidm1.ra.cn.idc, replica, dc\3Didc, mapping tree, config

dn: cn=meToidm1.ra.cn.idc,cn=replica,cn=dc\3Didc,cn=mapping tree,cn=config

cn: meToidm1.ra.cn.idc

objectClass: nsds5replicationagreement

objectClass: top

nsDS5ReplicaTransportInfo: TLS

description: me to idm1.ra.cn.idc

nsDS5ReplicaRoot: dc=idc

nsDS5ReplicaHost: idm1.ra.cn.idc

nsds5replicaTimeout: 600

nsDS5ReplicaBindDN: cn=replication manager,cn=config

nsDS5ReplicaPort: 389

nsDS5ReplicaBindMethod: simple

nsDS5ReplicatedAttributeList: (objectclass=*) $ EXCLUDE memberof idnssoaserial

  entryusn krblastsuccessfulauth krblastfailedauth krbloginfailedcount

nsds5ReplicaStripAttrs: modifiersName modifyTimestamp internalModifiersName in

 ternalModifyTimestamp

nsDS5ReplicaCredentials: {DES} 

nsDS5ReplicatedAttributeListTotal: (objectclass=*) $ EXCLUDE entryusn krblasts

 uccessfulauth krblastfailedauth krbloginfailedcount

nsds5replicareapactive: 0

nsds5replicaLastUpdateStart: 0

nsds5replicaLastUpdateEnd: 0

nsds5replicaChangesSentSinceStartup:

nsds5replicaLastUpdateStatus: 402 Replication error acquiring replica: unknown

  error - Replica has different database generation ID, remote replica may nee

 d to be initialized

nsds5replicaUpdateInProgress: FALSE

nsds5replicaLastInitStart: 0

nsds5replicaLastInitEnd: 0

# search result

search: 2

result: 0 Success

# numResponses: 5

# numEntries: 4

--
389 users mailing list
389-users@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-users

--
389 users mailing list
389-users@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-users

--
389 users mailing list
389-users@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-users

--
389 users mailing list
389-users@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-users

--
389 users mailing list
389-users@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-users