Skip to main content

[Case 82261] 11.40 incomplete/hung update [symptom: exim & imap services down]

Comments

8 comments

  • cPanelMichael
    re: [Case 82261] whm/cpanel 11.40.0 update problem with exim & imap services Hello :) Please run the following command and let us know the output: /scripts/check_cpanel_rpms
    Do you notice any error messages when restarting Exim or IMAP? Are there any errors during the restart in the /var/log/exim_paniclog file? Also, since you have services down, please open a support ticket so we can take a closer look: Submit A Ticket Post the ticket number here so we can update this thread with the outcome. Thank you.
    0
  • x-origin
    re: [Case 82261] whm/cpanel 11.40.0 update problem with exim & imap services cPanelMichael thank you for your reply, The ticket number is 4388071, and i quote all the informations in the ticket. Special thanks.
    0
  • cPanelMichael
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] To update, there is an internal case open to address the issue where Dovecot does not function properly after updating to cPanel version 11.40. The internal case number is 82261. While the specific cause of this issue is not yet known (possibly attributed to issues with RPMs), the current workaround to resolve the issue with Dovecot is to correct the ownership and permission values via: chown root.dovecot /usr/local/cpanel/bin/dovecot-wrap ; chmod 4750 /usr/local/cpanel/bin/dovecot-wrap
    Note: The issue reported with the RPMs on this thread was resolved by removing some duplicate RPMs and running: /usr/local/cpanel/scripts/check_cpanel_rpms Thank you.
    0
  • High Voltage
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] Thanks cPanelMichael. Your workaround got my mail server up & running again.
    0
  • sahostking
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] Thanks this may have helped
    0
  • cPanelNick
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] In the instances we have seen this problem, the update was terminated in the middle of the process (out of memory, server reboot, or killed off in most instances). The permissions are normally set on the dovecot wrapper closer to the end of the update. If the update does not fully complete you can manually fix the problem with the following: chown root:dovecot /usr/local/cpanel/bin/dovecot-wrap chmod 4750 /usr/local/cpanel/bin/dovecot-wrap yum clean all yum update /usr/local/cpanel/scripts/check_cpanel_rpms --fix
    Manually fix any problems with the rpm database as needed. /scripts/upcp --force
    If you are unable to determine and correct the reason for the update being terminated prematurely, please do open a ticket (link in my signature) so we can help you look into the matter. Its is important that the problem causing the update to be incomplete is resolved to prevent seemingly random problems from occurring. Its very likely that the next major update will not be able to complete as well if the underlying cause is not resolved now. The primary cause we see is the server running out of memory.
    0
  • tss
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] The exact same problem happened to my server with this update. Never had an update problem before. Here's what happens: update fails, no exim, no imap. I found out about this when the mail server stopped delivering and sending emails. Hundreds of notification emails followed, but because I didn't subscribe on my cell phone to the particular email address they were sent, I didn't see them until much later. Fortunately, it didn't cause any problems with Apache, PHP, MySQL, pure-ftp, or sshd. They kept running, so at least my clients' sites didn't go down. The first email had the subject of "cPanel update hanging" telling me "/usr/local/cpanel/scripts/upcp" was running for longer than 6 hours. The next email, 10 minutes later, had the subject of "Cpanel version change is blocked", it was fatal, and that my "RPM database appears unstable". About 12 hours after these messages were sent I noticed that I wasn't receiving expected emails at certain addresses. When I fired up my laptop and tried to log in to webmail on port 2096, I received a white screen instead of my expected SquirrelMail client. I tried a different email address on the same server with the same result. For some reason, only after these failed attempts to log in did I start getting 2 messages, every 5 minutes, from the cPanel Service monitor. "FAILED: imap on " and "FAILED: exim on ". The messages was fairly detailed, but didn't really tell me what the underlying problem was. From the imap message: TCP Transaction Log: << * OK [CAPABILITY IMAP4rev1 LITERAL+ SASL-IR LOGIN-REFERRALS ID ENABLE IDLE STARTTLS AUTH=PLAIN AUTH=LOGIN] Dovecot ready. >> A001 LOGIN __cpanel__service__auth__imap__ << A001 NO [UNAVAILABLE] Temporary authentication failure. [:2013-11-08 19:30:01] imap: ** [A001 NO [UNAVAILABLE] Temporary authentication failure. [:2013-11-08 19:30:01] != A001 OK] : Died at /usr/local/cpanel/Cpanel/TailWatch/ChkServd.pm line 894, <$socket_scc> line 2. The confusing part was that exim and dovecot services (at least some of them) appeared on the running processes page of WHM. The exim message was similar. A few details were: << 535 Incorrect authentication data exim: ** [535 Incorrect authentication data != 2] : Died at /usr/local/cpanel/Cpanel/TailWatch/ChkServd.pm line 894, <$socket_scc> line 11. and Starting clamd: [ OK ] Starting exim: [ OK ] 0 processes (antirelayd) sent signal 9 /usr/local/cpanel/scripts/update_sa_rules: running in background So exim and dovecot were running, but the authentication service wasn't authenticating. I finally saw that the update had failed, and hoped that was the cause of the problem, so I attempted a manual forced update. That's when I saw a virtually identical log to the one shown in x-origin's original post above. However, the solution for me came from a Russian forum ([url=http://forum.searchengines.ru/showthread.php?p=12317747]cPanel 11.40.0.19 (RELEASE) -) which I found by searching for "Test install failed: package cpanel-perl-514-Devel-NYTProf-5.05-1.cp1136.x86_64 is already installed". Thanks to google translate and someone's comment, I realized that I needed remove the RPMs that were causing the problem with the command "rpm -e". I tried 2 different manual updates; both failed on different RPMs. The first time the log had the identical message as shown above: Downloading
    0
  • cPanelNick
    re: [Case 82261] whm/cpanel 11.40.0 incomplete/hung update [symptom: exim & imap services] [quote="tss, post: 1503082"> The first email had the subject of "cPanel update hanging" telling me "/usr/local/cpanel/scripts/upcp" was running for longer than 6 hours. The next email, 10 minutes later, had the subject of "Cpanel version change is blocked", it was fatal, and that my "RPM database appears unstable".
    Please do open a ticket so we can determine why the update was hung up on your server to ensure it doesn't happen again next update.
    0

Please sign in to leave a comment.