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

plesk 8.0 crashes, empty pages on admin panel

Discussion in 'Plesk for Linux - 8.x and Older' started by djnono17, Jun 20, 2006.

  1. djnono17

    djnono17 Guest

    0
     
    Okay, big problem,
    since this morning, I have no longer access to psa, just an empty page.

    I restarted all services, reboot the server, nothing....

    So, I added logging to the httpsd.conf file :
    ErrorLog /var/log/psa_error_log

    I restart the server, nothing in my log, but when I load a page :
    [Tue Jun 20 16:38:12 2006] [notice] Apache/1.3.33 (Unix) mod_ssl/2.8.22 OpenSSL/0.9.7e PHP/5.0.5 configured -- resuming normal operations
    [Tue Jun 20 16:38:12 2006] [notice] Accept mutex: sysvsem (Default: sysvsem)
    [6786] EACCELERATOR: PHP crashed on opline 57 of strncmp() at /usr/local/psa/admin/auto_prepend/auth.php3:40

    [Tue Jun 20 16:38:16 2006] [notice] child pid 6786 exit signal Segmentation fault (11)


    I'm running on fedora fc4, and no modifications on the server have been done since at least two weeks.

    any idea ?
     
  2. alphamega

    alphamega Guest

    0
     
    /tmp/eaccelerator does not exist

    Hello Colleagues,

    We are running
    -Red Hat Enterprise Linux ES release 4 (Nahant Update 4)
    -Plesk 8.0.1 RedHat el4 80061018.10
    -Plesk CP runs on PHP 5.0.5 (did come with Plesk)
    -Plesk CP runs eAccelerator 0.9.3 (did come with Plesk)

    I was seeing also errors like
    Code:
    [Thu Dec  7 08:28:42 2006] [notice] child pid 20516 exit signal Segmentation fault (11)
    [20524] EACCELERATOR: PHP crashed on opline 2 of main() at /usr/local/psa/admin/plib/class.Form.php:11
    in the file /usr/local/psa/admin/logs/httpsd_error_log

    I discovered using
    Code:
    /usr/local/psa/admin/bin/php -i |grep -i eaccel
    that eAccelerator is configured to use /tmp/eaccelerator as cache_dir.
    That directory did not exist on our Plesk 8.0.1 server.

    So I created it using:
    Code:
    mkdir /tmp/eaccelerator
    chmod 700 /tmp/eaccelerator
    chown psaadm:psaadm /tmp/eaccelerator
    And to make sure the dir does not get deleted by tmpwatch after a period of inactivity I opened the file /etc/cron.daily/tmpwatch and added the following line to the beginning of the file:
    Code:
    touch /tmp/eaccelerator
    After I restarted Plesk and I accessed the Plesk controlpanel several times, I see eAccelerator is truly using the directory to store cache files.

    I don't know for sure it this is the solution to the errors, I didn't discover how to reproduce the errors yet. But it seems clear to me that is better to have the /tmp/eaccelerator dir existing.
     
Loading...