No destination error, when destinations ARE defined

  • ResolvedModerator Bet Hannon

    (@bethannon1)


    7 years, 9 months ago

    About 25 of our almost 150 sites that are running BWPup have begun throwing an odd error (they worked just fine for months/years before):

    [INFO] Backup file is: /home/sitename/public_html/wp-content/uploads/backwpup-eebeca-temp/backwpup_eebeca01_site_daily_2017-06-26.zip
[26-Jun-2017 05:00:04] ERROR: No destination correctly defined for backup! Please correct job settings.
[26-Jun-2017 05:00:04] One old log deleted
[26-Jun-2017 05:00:04] ERROR: Job has ended with errors in 0 seconds. You must resolve the errors for correct execution.

    I’m getting this even though a destination for S3 IS properly configured! I am NOT getting this when I manually run the jobs, only when they trigger as scheduled. Since this is only happening on scheduled jobs, it makes me wonder if it’s connected to another issue we recently had on our VPS server.

    We had an issue on our VPS server about a month ago, that caused a PHP module to disappear (sorry, can’t remember off top of my head which one), which caused all scheduled tasks to not run. We do have wp-cron disabled on these sites, and are running a server cron to trigger wp-cron every 30 min.

    It took 3 weeks before a client reported that their scheduled posts were not publishing, and in troubleshooting we found this missing PHP module. That was restored last week, and the next day I started receiving these error messages for backup jobs.

    Here is what I’ve tried:

    • Simply resaving the S3 destination tab settings
    • Creating a new scheduled job from scratch</li >
    • re-enabling wp-cron
    • Setting the logging level to debug (nothing additional appears in logs)
    • Checking the Settings – information: curl, self-connect, etc. all seem to be fine

      • With all of these, I still get the errors below, but only on scheduled jobs — manually triggered jobs work fine. Scheduled posts work fine, (and the crons are triggering), so cron seems ok. It’s only happening on about 15% of our sites. Weird.

      If you have seen this sort of error before, can you point me in the right direction for additional troubleshooting?

      Does this point to perhaps an additional PHP module that might be missing? Here is what is currently on our VPS (from the Settings-Info tab):
      Core, PDO, Phar, Reflection, SPL, SimpleXML, bcmath, bz2, calendar, cgi-fcgi, ctype, curl, date, dom, exif, fileinfo, filter, ftp, gd, gettext, hash, iconv, imap, intl, ionCube Loader, json, libxml, mbstring, mcrypt, mysqli, mysqlnd, openssl, pcre, pdo_mysql, pdo_pgsql, pdo_sqlite, pgsql, posix, pspell, session, soap, sockets, sqlite3, standard, tidy, tokenizer, xml, xmlreader, xmlrpc, xmlwriter, xsl, zip, zlib

    • This topic was modified 7 years, 9 months ago by Bet Hannon.
Viewing 7 replies - 1 through 7 (of 7 total)

  • Hi @bethannon1, I’ll be honest, I have never seen that error.

    Could you enable WP_DEBUG to allow errors to be shown, then try to run the cron manually from SSH, assuming you know how to do this? Let me know if any errors appear, either in the shell or in the log.

    Moderator Bet Hannon

    (@bethannon1)

    So I enabled debug and (since I’m not sure about how to trigger cron from SSH) I set a test job to run at a close future time, and just waited for it, since the server cron is triggering wp-cron every 5 min.

    Nothing shows up in the WP error log. Still the same errors in the job.

    I’ve contacted my host, and we confirmed that the cron is running. They triggered wp-cron from SSH, and there were no errors there. We thought there might be a mod_security issue, but that turned out not to be the case.

    The backups still DO complete without error manually, but give the same “No destination correctly defined” errors for scheduled jobs.

    Any other ideas for our troubleshooting?

    Moderator Bet Hannon

    (@bethannon1)

    I have worked with my hosting company support team, and they aren’t seeing anything unusual.

    Since the backup jobs do complete when run manually (including the send it to Amazon S3), but not when triggered by cron (either server cron or regular wp-cron), any other troubleshooting suggestions for trying to figure out why it would be getting those “no destination properly configured” errors on scheduled backups?

    That just seems really weird – that all the scripts and functions work fine when manually triggered, but not when triggered by a cron.

    Moderator Bet Hannon

    (@bethannon1)

    Just found a new clue:

    All the sites that are having this issue seem to have an .htaccess file in the uploads folder from WordFence:

    # BEGIN Wordfence code execution protection
<IfModule mod_php5.c>
php_flag engine 0
</IfModule>
<IfModule mod_php7.c>
php_flag engine 0
</IfModule>

AddHandler cgi-script .php .phtml .php3 .pl .py .jsp .asp .htm .shtml .sh .cgi
Options -ExecCGI
# END Wordfence code execution protection

    The files look like they’ve been there for a while though, so I’m not sure how they are causing this error. Will be looking for the WF option that created them.

    Moderator Bet Hannon

    (@bethannon1)

    Just an update: it appears that this issue is being caused by a setting in WordFence. When the option for “Disable Code Execution for Uploads directory” is checked, it puts an .htaccess file in the uploads directory. I have confirmed that this is the cause of this issue.

    I have turned off this option for now, and given my testing, I think my backups will be completing as scheduled. I have been in touch with the WordFence team, and they are working on this.

    Thanks for your help!

    Moderator Bet Hannon

    (@bethannon1)

    Ugh! So it looks like I was overly optimistic that this issue is solved. A couple of scheduled backups ran in the last 15 minutes, and we are getting the same errors of “no destination correctly defined”.

    Part of the troubleshooting above was realizing that I was getting errors like:
    client denied by server configuration: /home/sitename/public_html/wp-content/uploads/backwpup-eebeca-temp/backwpup_eebeca01_site_daily_2017-06-26.zip
    basically, WordFence was not letting BWPup write to the uploads folder. I’m no longer getting THAT error.

    In fact, I’m not getting any errors at all.

    Any further troubleshooting tips would be welcome!

    Moderator Bet Hannon

    (@bethannon1)

    update: FINALLY have this issue resolved!

    I thought I had checked this before, but in working with my hosting support team, they found that PHP 7 was missing both the zip and curl extensions. 0_0

    Originally, this backup job was set when the PHP version of the site was 5.6. About a month ago, as we were trying to deal with some high CPU usages, we tried to get most of our sites upgraded to PHP 7. Nothing seemed to go wrong at that time, but one of the missing extensions in PHP 7 was one that was triggering crons, which meant that no backups were happening, which meant no error messages were being received!

    I think everything is back to normal. Thanks for your help along the way!

Viewing 7 replies - 1 through 7 (of 7 total)
  • The topic ‘No destination error, when destinations ARE defined’ is closed to new replies.