Skip to main content

cPanel update issue

dchen
Hello, I received this error when cPanel tries to update itself. Can anyone please advice on how to fix this:
'httpupdate.cpanel.net'. [2015-09-27 00:16:26 -0400] Successfully verified signature for cpanel (key types: release). [2015-09-27 00:16:26 -0400] Maximum sync children set to 1 based on 1239M available memory. [2015-09-27 00:16:26 -0400] Staging files for x3mail [2015-09-27 00:16:27 -0400] Staging cpanel changes [2015-09-27 00:16:27 -0400] Retrieving and staging /cpanelsync/11.50.1.3/binaries/linux-c6-x86_64/.cpanelsync.bz2 [2015-09-27 00:16:27 -0400] Using mirror '208.74.125.42' for host 'httpupdate.cpanel.net'. [2015-09-27 00:16:28 -0400] Successfully verified signature for cpanel (key types: release). [2015-09-27 00:16:28 -0400] Retrieving and staging /cpanelsync/11.50.1.3/cpanel/.cpanelsync.bz2 [2015-09-27 00:16:29 -0400] Successfully verified signature for cpanel (key types: release). [2015-09-27 00:16:29 -0400] Maximum sync children set to 1 based on 1217M available memory. [2015-09-27 00:16:29 -0400] Staging files for binaries/linux-c6-x86_64 [2015-09-27 00:17:57 -0400] ***** FATAL: Failed to stage "cpanel changes" because of an error: Read failed: Input/output error at /usr/local/cpanel/scripts/updatenow.static-cpanelsync line 41205. [2015-09-27 00:17:57 -0400] The Administrator will be notified to review this output when this script completes => Log closed Sun Sep 27 00:17:57 2015 [2015-09-27 00:17:58 -0400] 17% complete [2015-09-27 00:17:58 -0400] E Running `/usr/local/cpanel/scripts/updatenow --upcp --log=/var/cpanel/updatelogs/update.1443327234.log --force` failed, exited with code 12 (signal = 0)
Thanks, -D Additional info. I have manually ran /scripts/upcp --force command and same result. This error has been persist for several days now since the WHM is set to auto update. If it is mirror error, then it still not working.

Comments

10 comments

Date Votes
  • dalem
    1. check you disk space df -h df -i cpanel should block the update if your out of disk space but I would check it just the same 2. Check your file system for corruption hence the error "Input/output error" run a manual fsck on your drive partitions
    0
  • dchen
    Thanks. I don't have problem with disk space. However, when I ran fsck /dev/sda1 or other, I got this: fsck /dev/sdb1 fsck from util-linux-ng 2.17.2 e2fsck 1.41.12 (17-May-2010) /dev/sdb1 is mounted. e2fsck: Cannot continue, aborting. Do you think this caused my problems? Thank you, -D
    0
  • Jcats
    No you can't run a FSCK on a mounted partition. You would need to run 'umount' but this can be very bad since its more than likely your root partition. If you have errors, I would check the following log files.
    # dmesg # cat /var/log/messages | less
    They may have some indication of disk errors, etc.
    0
  • dalem
    Indeed you need to umount the partition first
    0
  • cPanelMichael
    Hello :) This is typically the result of hard drive errors. Check /var/log/messages when this happens to see if you notice any particular output. You could also use "strace" on the upcp process if you are comfortable doing so to see which file is not readable:
    strace -ff -T -tt -v -s 1024 -o /root/upcp.output /scripts/upcp
    The output is then readable at /root/upcp.output for investigation. Thank you.
    0
  • dchen
    Thanks. I tried your command and got the same result as before:
    [2015-10-03 13:40:03 -0400] Staging files for binaries/linux-c6-x86_64 [2015-10-03 13:44:27 -0400] ***** FATAL: Failed to stage "cpanel changes" because of an error: Read failed: Input/output error at /usr/local/cpanel/scripts/updatenow.static-cpanelsync line 41205. [2015-10-03 13:44:27 -0400] The Administrator will be notified to review this output when this script completes => Log closed Sat Oct 3 13:44:27 2015 [2015-10-03 13:44:30 -0400] 17% complete [2015-10-03 13:44:30 -0400] E Running `/usr/local/cpanel/scripts/updatenow --upcp --log=/var/cpanel/updatelogs/update.1443893944.log` failed, exited with code 12 (signal = 0) => Log closed Sat Oct 3 13:44:39 2015
    I have asked hosting company to check for disk errors. All they did was to power cycle the server. Best, -Dong
    0
  • dchen
    I looked at the line 41205:
    sub _get_checksum_lib { my ( $self, $digest, $file, $size ) = @_; my $ctx = $digest eq 'sha' ? Digest::SHA->new(512) : Digest::MD5->new(); my $hash; if ( open( my $fh, '<', $file ) ) { if ( $ctx->addfile($fh) ) { $hash = $ctx->hexdigest(); } else { $hash = ''; $self->{'logger'}->error("Unable to read file ($file) to validate it's signature!"); $self->{'logger'}->set_need_notify(); } } else { $hash = ''; $self->{'logger'}->error("Unable to open $file to get it's checksum: $!"); $self->{'logger'}->error( "Stat: " . `stat '$file'` ); #give more debug info $self->{'logger'}->set_need_notify(); return ''; } return $hash; }
    What is wrong with $hash? It is just a variable. Best, -D
    0
  • cPanelMichael
    Feel free to open a support ticket using the link in my signature so we can take a closer look. You can post the ticket number here so we can update this thread with the outcome. Thank you.
    0
  • dchen
    Thank you. I just submitted a support request.
    0
  • cPanelMichael
    To update, it was determined the system require FSCK (File System Check) to repair some corrupted files. Thank you.
    0

Please sign in to leave a comment.

Didn't find what you were looking for?

New post