Illegal character in authority while trying to do External Restore

**stsimb** · 11-11-2013, 09:35 PM

Our zimbra installation is as follows:
Zimbra 8.0.4 OSE, ZeXtras 1.8.11, CentOS 6 x86_64, 17 Zimbra servers of which 9 are mailbox servers. We use Multi-Master Replication in our LDAPs.

Trying to move a mailbox between two of our servers, ie Zx Backup Moving a mailbox between two mailboxd servers - ZeXtras Suite Wiki the doExternalRestore operation fails.

[zimbra@zm-mbox-02 ~]$ zxsuite core getNotification id 78e07e46-d8cd-4a27-b96b-cf0ebf6b4276

Subject: ZeXtras ZxBackup Notification, External Restore failed.
Date: 11/11/2013 23:22:44
Operation id: 6f2ed028-759f-4262-a99b-5981d2e3958c
Group: BACKUP
Level: Error
Ack: false
Server: zm-mbox-02.cloud.forthnet.prv
Notification id: 78e07e46-d8cd-4a27-b96b-cf0ebf6b4276
Text:
This is an automated notification from ZxBackup about External Restore.

Operation External Restore failed.

Operation Id: 6f2ed028-759f-4262-a99b-5981d2e3958c
Operation Host: zm-mbox-02.cloud.forthnet.prv

Operation requested by: zimbra

ZeXtras Version: 1.8.11
ZeXtras commit: a7c892d274217bc312945152e49013aadc8c2b35
Zimbra version: 8.0.4_GA_5737 20130524120036 20130524-1204 FOSS

- exception -
Invalid server config backup
at com.zextras.backup.ZEBackupServer.getLdapMasterNam e ( ZEBackupServer.java:1193 )
at com.zextras.backup.ZEBackupServer.restoreCOS ( ZEBackupServer.java:800 )
at com.zextras.backup.ZEBackupServer.restorePostAccou nt ( ZEBackupServer.java:674 )
at com.zextras.op.backup.ZEExternalRestoreOperation.d oOperation ( ZEExternalRestoreOperation.java:726 )
at com.zextras.op.ZEOperation.exec ( ZEOperation.java:652 )
at com.zextras.backup.ZEBackupThread.execOperation ( ZEBackupThread.java:225 )
at com.zextras.backup.ZEBackupThread.run ( ZEBackupThread.java:114 )
Caused by: java.net.URISyntaxException: Illegal character in authority at index 7: ldap://zm-ldap-01.cloud.forthnet.prv:389 ldap://zm-ldap-02.cloud.forthnet.prv:389
at java.net.URI$Parser.fail ( URI.java:2829 )
at java.net.URI$Parser.parseAuthority ( URI.java:3167 )
at java.net.URI$Parser.parseHierarchical ( URI.java:3078 )
at java.net.URI$Parser.parse ( URI.java:3034 )
at java.net.URI.<init> ( URI.java:595 )
at com.zextras.backup.ZEBackupServer.getLdapMasterNam e ( ZEBackupServer.java:1188 )
at com.zextras.backup.ZEBackupServer.restoreCOS ( ZEBackupServer.java:800 )
at com.zextras.backup.ZEBackupServer.restorePostAccou nt ( ZEBackupServer.java:674 )
at com.zextras.op.backup.ZEExternalRestoreOperation.d oOperation ( ZEExternalRestoreOperation.java:726 )
at com.zextras.op.ZEOperation.exec ( ZEOperation.java:652 )
at com.zextras.backup.ZEBackupThread.execOperation ( ZEBackupThread.java:225 )
at com.zextras.backup.ZEBackupThread.run ( ZEBackupThread.java:114 )

- debug info -
-- parameters --
onlyProvisioning = false
restoreChatBuddies = true
filterDeleteAccount = true
obeyHSM = false
humanDate = last available
OpStartTime = 1384204961426
origin = ZxLink
skipCos = true
date = null
skipAliases = true
domainNameList = forthnet.gr
skipDistList = true
dataPath = /mail-backup/zm-mbox-01.cloud.forthnet.prv/
notificationeMails =
accountNameList = zimbra1r@forthnet.gr
skipSystemAccount = true
-- context --
numAccountsRestoredPhase1 = 1
extMap = com.zextras.backup.ZEBackupExtMap@393cea4e
ignoredLdapAttributes = zimbraId, zimbraCreateTimestamp
currentAccount = zimbra1r@forthnet.gr
failed = 1
usedCOSList =
backup_version = 8.0.4
completed = 1
numTotalAccounts = 1
Last progress info:
--
Operation log path: /opt/zimbra/log/zextras_ZxBackup_ExternalRestore_6f2ed028-759f-4262-a99b-5981d2e3958c.log
Warnings/Errors: 0/0
Restored Accounts(only attributes): 1/1
Restored Accounts(common items): 0/1
Restored Accounts(only shared & datasources): 0/1
Current Account: zimbra1r@forthnet.gr
Restored/Skipped Items: 0/0
Unrestored Items: 0

**Cine** · 11-12-2013, 09:50 AM

Hello stsimb!

After investigating the issue you report, we found out that there is a bug affecting ZeXtras Backup when dealing with multi-master infrastructures.
The bug will be fixed in ZeXtras Suite 1.8.12, but there is a quick and easy workaround:

- Locate the latest backup_{DD_MM_YY#hh_mm_ss} file in the "server/" subdirectory Backup Path of the "source" server.
- Open the file with a text editor.
- Search for the following string: "ldap_master_url"
- Edit this part
from

Code:

"ldap_master_url":"ldap://zm-ldap-01.cloud.forthnet.prv:389 ldap://zm-ldap-02.cloud.forthnet.prv:389"

to

Code:

"ldap_master_url":"ldap://zm-ldap-01.cloud.forthnet.prv:389"

This won't affect the backup in any way but will allow the restore to be completed.
Mind that a new backup_{...} file will be created at each *Scan operation, so remember to edit the appropriate file before running a restore...

Have a nice day,
Cine
the ZeXtras Team