1. Please take a little time for this simple survey! Thank you for participating!
    Dismiss Notice
  2. Dear Pleskians, please read this carefully! New attachments and other rules Thank you!
    Dismiss Notice
  3. Dear Pleskians, I really hope that you will share your opinion in this Special topic for chatter about Plesk in the Clouds. Thank you!
    Dismiss Notice

Synchronization between Plesk- and Mailserver failed

Discussion in 'Plesk Expand 2.3 Troubleshooting Issues' started by AndreasK, Oct 28, 2008.

  1. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    Hello everybody,

    we monitored a bug regarding the synchronization between Plesk and Mailserver. When we delete a domain on the Pleskserver, the domain still exists on the Mailserver. The error occured since the update on Expand 2.3.1.

    Best regards
    Andreas
     
  2. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    You need to update the Expand to 2.3.2.
     
  3. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    Hey gold,

    thank you for the very fast help, the bug is corrected with the new version.

    But we have a new error now:

    In the Expand login both mailservers now show a "Problem" in the "Sync status" column, but when we create a client the server syncs it with the mailserver and there are no problem accounts. Does someone has a soluten for the "Problem"-Status?

    Thank you!
    Andreas
     
  4. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    You need to go to 'Server > System Health > Operations' page and check the status of 'Synchronize mail servers' operation.
     
  5. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    Hello gold,

    we did what you told us but we still get the following errors:


    <errtxt>Action didn't handle a signal: Terminated</errtxt>

    <errtext>Can`t connect to MySQL server. Please check its state and retry action.</errtext>

    <errtext>[Operator] Mail server traffic synchronizing has errors. A number of errors happened while syncing traffic:
    *) (2108) [Transport] PleskAgent answer validation failed. Invalid reply from Plesk server #9 (xxxx.xxxx.xxxx.xxxx). The message does not conform to protocol 1.4.2.0 version. Make sure your Plesk and Plesk Expand servers have all available patches and hotfixes applied, if that does not help, then report a problem to Parallels. XML validation error: Error at file 'xmlPacket', line 2, column 697: Unknown element 'set_traffic'.

    *) (2108) [Transport] PleskAgent answer validation failed. Invalid reply from Plesk server #9
    The message does not conform to protocol 1.4.2.0 version. Make sure your Plesk and Plesk Expand servers have all available patches and hotfixes applied, if that does not help, then report a problem to Parallels. XML validation error: Error at file 'xmlPacket', line 2, column 1071: Unknown element 'set_traffic'.</errtext>

    <errtext>[Operator] Error gathering mail statistics. Errors occured during statistics gathering:
    [Operator] PleskAgent error. Error mail server answer for object #2370: (1013) [PleskAgent - GENERAL] Object does not exists. domain does not exist.</errtext>
    <id>5</id>

    I think this problem has something to do with the error that was created with version 2.3.1. There are some domains on the mailserver which have not been deleted by the expand. Do you agree with that?

    Thank you & best regards
    Andreas
     
  6. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    No. I think that the problem lies in the failed 'Synchronize mail servers' operations. You post errors for 'Synchronize mail traffic & disk usage stats for centralized mail' operation. Please look the status of 'Synchronize mail servers' operation.
     
  7. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    Hi gold,

    thank you for the fast help. These are the errors:

    <?xml version="1.0" encoding="UTF-8" standalone="no"?>
    <packet version="0.1.1.0">
    ****<result>
    ********<status>error</status>
    ********<errtxt>Action didn't handle a signal: Terminated</errtxt>
    ****</result>
    </packet>
    ____________________________________________________________________

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="1948269" version="2.3.2.17">
    ****<sync>
    ********<system>
    ************<status>error</status>
    ************<errcode>4000</errcode>
    ************<errtext>Can`t connect to MySQL server. Please check its state and retry action.</errtext>
    ********</system>
    ****</sync>
    </packet>

    These are two different servers.

    Best regards
    Andreas
     
  8. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    What is the general status (green, yellow, red) of 'Synchronize mail servers' operations?
     
  9. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    The status says (yellow):

    OK (2 failures the last 7 days)
     
  10. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    This is normal.
     
  11. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    OK, but it is confusing when both mailservers report a problem in the "centralized mail" overview. Is it just a matter of time till the server reports no errors?
     
  12. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    We need for additional info. Please find the last 'Synchronize mail servers' action on 'Server > Action Log' page and post result of following command:
    Code:
    cat /usr/local/expand/var/log/operator.log | grep LAST_ACTION_ID
    
    where LAST_ACTION_ID - ID of action found.
     
  13. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    the ID seems to be missing in the log. Is there a solution how to resync the mailservers in the expand?
     
  14. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    We found something in the logs, these seems to be the errors:

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="1997330" version="2.3.2.17">
    ****<sync>
    ********<system>
    ************<status>error</status>
    ************<errcode>4000</errcode>
    ************<errtext>Can`t connect to MySQL server. Please check its state and retry action.</errtext>
    ********</system>
    ****</sync>
    </packet>


    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="1987509" version="2.3.2.17">
    ****<sync>
    ********<system>
    ************<status>error</status>
    ************<errcode>4000</errcode>
    ************<errtext>Can`t connect to MySQL server. Please check its state and retry action.</errtext>
    ********</system>
    ****</sync>
    </packet>

    We checked our SQL databases, everything is ok there.

    # zcat /usr/local/expand/var/log/operator.log.1.gz |grep 1997330
    Nov 03 2008 04:19:36 [1997330:13184] INFO: Request to operator 'plesk_mail_server'
    Nov 03 2008 04:19:36 [1997330:13184] INFO: ... action_id=1997330
    Nov 03 2008 04:19:36 MailSrvOp[1997330:13184] NOTICE: Command 'sync' requested
    Nov 03 2008 04:19:36 MailSrvOp[1997330:13184] INFO: Sync command called
    Nov 03 2008 04:19:36 MailSrvOp[1997330:13184] INFO: Synching all mail servers
    Nov 03 2008 04:19:36 MailSrvOp[1997330:13184] INFO: Synchronization of mail server #5
    Nov 03 2008 04:19:36 [1997330:13184] INFO: Locking object(s) #5 in the 'exp_mail_server' table - state 'flush', action #1997330.
    Nov 03 2008 04:19:36 [1997330:13184] INFO: Object(s) #5 have been successfully locked in state 'flush' by action #1997330 - 1 row(s) affected.
    Nov 03 2008 04:20:00 [1997330:13184] ERR: Failed to sync updated CMail object #7709 in reverse: (13208) [Operator] DNS zone not found. Domain alias #1155 DNS zone.
    Nov 03 2008 04:20:00 [1997330:13184] ERR: Synching of object #7709 failed: [Operator] DNS zone not found. Domain alias #1155 DNS zone.
    Nov 03 2008 04:20:00 [1997330:13184] ERR: Failed to sync updated CMail object #5261 in reverse: (13208) [Operator] DNS zone not found. Domain #2333 DNS zone.
    Nov 03 2008 04:20:00 [1997330:13184] ERR: Synching of object #5261 failed: [Operator] DNS zone not found. Domain #2333 DNS zone.
    Nov 03 2008 04:20:05 [1997330:13184] INFO: Unlocking object(s) #5 in the 'exp_mail_server' table.
    Nov 03 2008 04:20:05 MailSrvOp[1997330:13184] ERR: General error: Can`t connect to MySQL server. Please check its state and retry action.
    Nov 03 2008 04:20:05 [1997330:13184] INFO: Result code: 40
    Nov 03 2008 04:28:16 [1997342:27416] WARNING: Object #5 is locked in state 'flush' by action #1997330, but this action is in state 'failed' - not 'running'. Marking for forced unlocking.

    ___________________________________________________________________________________________

    # zcat /usr/local/expand/var/log/operator.log.2.gz |grep 1987509
    Nov 02 2008 04:19:28 [1987509:13184] INFO: Request to operator 'plesk_mail_server'
    Nov 02 2008 04:19:28 [1987509:13184] INFO: ... action_id=1987509
    Nov 02 2008 04:19:28 MailSrvOp[1987509:13184] NOTICE: Command 'sync' requested
    Nov 02 2008 04:19:28 MailSrvOp[1987509:13184] INFO: Sync command called
    Nov 02 2008 04:19:28 MailSrvOp[1987509:13184] INFO: Synching all mail servers
    Nov 02 2008 04:19:28 MailSrvOp[1987509:13184] INFO: Synchronization of mail server #5
    Nov 02 2008 04:19:28 [1987509:13184] INFO: Locking object(s) #5 in the 'exp_mail_server' table - state 'flush', action #1987509.
    Nov 02 2008 04:19:28 [1987509:13184] INFO: Object(s) #5 have been successfully locked in state 'flush' by action #1987509 - 1 row(s) affected.
    Nov 02 2008 04:19:54 [1987509:13184] ERR: Failed to sync updated CMail object #7709 in reverse: (13208) [Operator] DNS zone not found. Domain alias #1155 DNS zone.
    Nov 02 2008 04:19:54 [1987509:13184] ERR: Synching of object #7709 failed: [Operator] DNS zone not found. Domain alias #1155 DNS zone.
    Nov 02 2008 04:19:55 [1987509:13184] ERR: Failed to sync updated CMail object #5261 in reverse: (13208) [Operator] DNS zone not found. Domain #2333 DNS zone.
    Nov 02 2008 04:19:55 [1987509:13184] ERR: Synching of object #5261 failed: [Operator] DNS zone not found. Domain #2333 DNS zone.
    Nov 02 2008 04:19:57 [1987509:13184] INFO: Unlocking object(s) #5 in the 'exp_mail_server' table.
    Nov 02 2008 04:19:57 [1987509:13184] INFO: Object(s) #5 have been unlocked - 1 row(s) affected.
    Nov 02 2008 04:19:58 MailSrvOp[1987509:13184] INFO: Synchronization of mail server #7
    Nov 02 2008 04:19:58 [1987509:13184] INFO: Locking object(s) #7 in the 'exp_mail_server' table - state 'flush', action #1987509.
    Nov 02 2008 04:19:58 [1987509:13184] INFO: Object(s) #7 have been successfully locked in state 'flush' by action #1987509 - 1 row(s) affected.
    Nov 02 2008 04:20:00 [1987509:13184] INFO: Unlocking object(s) #7 in the 'exp_mail_server' table.
    Nov 02 2008 04:20:00 [1987509:13184] INFO: Object(s) #7 have been unlocked - 1 row(s) affected.
    Nov 02 2008 04:20:00 MailSrvOp[1987509:13184] INFO: Synchronization of mail server #8
    Nov 02 2008 04:20:00 [1987509:13184] INFO: Locking object(s) #8 in the 'exp_mail_server' table - state 'flush', action #1987509.
    Nov 02 2008 04:20:00 [1987509:13184] INFO: Object(s) #8 have been successfully locked in state 'flush' by action #1987509 - 1 row(s) affected.
    Nov 02 2008 04:20:05 [1987509:13184] INFO: Unlocking object(s) #8 in the 'exp_mail_server' table.
    Nov 02 2008 04:20:05 MailSrvOp[1987509:13184] ERR: General error: Can`t connect to MySQL server. Please check its state and retry action.
    Nov 02 2008 04:20:05 [1987509:13184] INFO: Result code: 40
    Nov 02 2008 04:32:12 [1987521:13626] WARNING: Object #8 is locked in state 'flush' by action #1987509, but this action is in state 'failed' - not 'running'. Marking for forced unlocking.
     
  15. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    You need
    1. Restart the Expand (/etc/init.d/expand restart)
    2. Reload DNS records on the 'Plesk Servers > Reload Servers' page for the Plesk servers having failed DNS zones (Domain alias #1155 DNS zone, Domain #2333 DNS zone)

    What OS and MySQL sever versions have you?
     
  16. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    On the web and mailservers we use

    OS: Debian etch 4.0
    MySQL: 5.0.32

    each 64 Bit systems. Our expand is a 32 Bit Debian etch 4.0.
     
  17. gold

    gold Regular Pleskian

    25
    57%
    Joined:
    Jan 8, 2008
    Messages:
    307
    Likes Received:
    0
    How many such actions have you?
    What MySQL server version use you with Expand?
     
  18. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    We use the same SQL version like the other servers!
     
  19. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    So we

    1. Restart the Expand
    2. Reload DNS records - that completed with errors like:

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="2020811" version="2.3.2.17">
    <refresh>
    <result>
    <status>error</status>
    <errcode>4003</errcode>
    <errtext>[Operator] PleskAgent error. (1015) [PleskAgent - GENERAL] Object owner is not found. Domain does not exist.</errtext>

    AND

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="2020810" version="2.3.2.17">
    <refresh>
    <result>
    <status>error</status>
    <errcode>2003</errcode>
    <errtext>[Transport] Bad HTTP answer code. Plesk server is unreachable now: HTTP answer code is 500 instead of 200(OK).
    Connection log
    ----
    CURLINFO_TEXT: About to connect() to xxxx.xxxx.xxxx.xxxx port xxxx
    CURLINFO_TEXT: Trying xxxx.xxxx.xxxx.xxxx... CURLINFO_TEXT: connected
    CURLINFO_TEXT: Connected to xxxx.xxxx.xxxx.xxxx (xxxx.xxxx.xxxx.xxxx) port xxxx
    CURLINFO_TEXT: successfully set certificate verify locations:
    CURLINFO_TEXT: CAfile: /etc/ssl/certs/ca-certificates.crt
    CApath: none
    CURLINFO_TEXT: SSLv2, Client hello (1):
    CURLINFO_SSL_DATA_OUT: CURLINFO_TEXT: SSLv3, TLS handshake, Server hello (2):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSLv3, TLS handshake, CERT (11):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSLv3, TLS handshake, Server key exchange (12):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSLv3, TLS handshake, Server finished (14):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSLv3, TLS handshake, Client key exchange (16):
    CURLINFO_SSL_DATA_OUT: CURLINFO_TEXT: SSLv3, TLS change cipher, Client hello (1):
    CURLINFO_SSL_DATA_OUT: CURLINFO_TEXT: SSLv3, TLS handshake, Finished (20):
    CURLINFO_SSL_DATA_OUT: CURLINFO_TEXT: SSLv3, TLS change cipher, Client hello (1):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSLv3, TLS handshake, Finished (20):
    CURLINFO_SSL_DATA_IN: CURLINFO_TEXT: SSL connection using DHE-RSA-AES256-SHA
    CURLINFO_TEXT: Server certificate:
    CURLINFO_TEXT: subject: /C=DE/O=*.domain/OU=GT50713369/OU=See www.rapidssl.com/resources/cps (c)08/OU=Domain Control Validated - RapidSSL(R)/CN=*.domain
    CURLINFO_TEXT: start date: 2008-09-17 14:57:16 GMT
    CURLINFO_TEXT: expire date: 2009-09-18 14:57:16 GMT
    CURLINFO_TEXT: issuer: /C=US/O=Equifax Secure Inc./CN=Equifax Secure Global eBusiness CA-1
    CURLINFO_TEXT: SSL certificate verify ok.
    CURLINFO_HEADER_OUT: POST /enterprise/control/agent.php HTTP/1.1
    Host: xxxx.xxxx.xxxx.xxxx
    Accept: */*
    HTTP_AUTH_PASSWD: x
    Content-Type: text/xml
    Content-Length: 157

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?><packet version="1.5.1.0"><domain><get><filter/><dataset><gen_info/></dataset></get></domain></packet>CURLINFO_HEADER_IN: HTTP/1.1 500 Internal Server Error
    CURLINFO_HEADER_IN: Date: Wed, 05 Nov 2008 15:01:18 GMT
    CURLINFO_HEADER_IN: Server: Apache
    CURLINFO_HEADER_IN: X-Powered-By: PHP/5.2.6
    CURLINFO_HEADER_IN: Expires: Thu, 19 Nov 1981 08:52:00 GMT
    CURLINFO_HEADER_IN: Last-Modified: Wed, 05 Nov 2008 15:01:18 GMT
    CURLINFO_HEADER_IN: Cache-Control: no-store, no-cache, must-revalidate, post-check=0, pre-check=0
    CURLINFO_HEADER_IN: Pragma: no-cache
    CURLINFO_HEADER_IN: P3P: CP="NON COR CURa ADMa OUR NOR UNI COM NAV STA"
    CURLINFO_HEADER_IN: Set-Cookie: PHPSESSID=x; path=/
    CURLINFO_HEADER_IN: Set-Cookie: locale=en-US; expires=Thu, 05-Nov-2009 15:01:18 GMT; path=/
    CURLINFO_HEADER_IN: Set-Cookie: locale=en-US; expires=Thu, 05-Nov-2009 15:01:18 GMT; path=/
    CURLINFO_HEADER_IN: Connection: close
    CURLINFO_HEADER_IN: Transfer-Encoding: chunked
    CURLINFO_HEADER_IN: Content-Type: text/xml
    CURLINFO_DATA_IN: CURLINFO_TEXT: Closing connection #0
    CURLINFO_TEXT: SSLv3, TLS alert, Client hello (1):
    CURLINFO_SSL_DATA_OUT:
    The server status changed to 'problem'.</errtext>
    <server_id>9</server_id>
    </result>
    </refresh>
    </packet>
     
  20. AndreasK

    AndreasK Regular Pleskian

    26
    23%
    Joined:
    Sep 6, 2006
    Messages:
    173
    Likes Received:
    0
    And we got another error:

    <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
    <packet action_id="2019099" version="2.3.2.17">
    ****<refresh>
    ********<system>
    ************<status>error</status>
    ************<errcode>4000</errcode>
    ************<errtext>Duplicate entry 'domain-4' for key 2. Error query: INSERT INTO plesk_dns_zone (server_id, name, display_name, soa_email, soa_ttl, soa_refresh, soa_retry, soa_expire, soa_minimum, type, status) VALUES (4, 'domain', 'domain', 'domain@domain', 86400, 10800, 3600, 604800, 10800, 'master', 'enabled')</errtext>
    ********</system>
    ****</refresh>
    </packet>

    In the expand the domain is a normal domain AND an alias so it is a duplicate. I deleted the domain without any problems. So we try to migrate the error clients now, probably we can identify them by using the domain IDs.
     
Loading...