It looks like PHP crashes very early.
If you access your PHP error log (ask your hosting company where it is, if you don’t know – it varies), then that should have some useful information in there.
David
Thread Starter
Squazz
(@squazz)
Edit: My bad, that’s not the right one. Will try to find the right one
(Support is closed, so will contact you tomorrow when I know more)
Thread Starter
Squazz
(@squazz)
http://pastebin.com/g5w0gRce
I found the correct PHP error log
Thread Starter
Squazz
(@squazz)
Yeah, it displayed a notice, and I changed .htaccess as it wanted me to. After that the notice went away.
Tricky. There’s nothing of interest in the error log – the error log starts 2 days later; is it possible it got truncated/rotated, and the portion of the error log from the backup date is no longer there?
It seems to crash very early…. but so far, there’s no information on why/exactly which line.
If you upload this to your website as anything.php, in the same directory as WordPress …
<?php
define('WP_DEBUG', true);
define('UPDRAFTPLUS_BROWSERLOG', true);
define('DOING_CRON', true);
require_once('wp-load.php');
do_action('updraft_backup_all');
… and then visit it in a browser, then what is the output?
(The only change here is that with the WP_DEBUG line, it has more change of catching the PHP error output).
Thread Starter
Squazz
(@squazz)
Sorry for the late answer, got a lot to look at for the time being.
I get this:
Opened log file at time: Thu, 05 Feb 2015 22:02:35 +0000 on http://aa-festival.dk
UpdraftPlus WordPress backup plugin (http://updraftplus.com): 1.9.51 WP: 4.1 PHP: 5.3.29 (Linux wsb12.surf-town.net 2.6.32-504.3.3.el6.x86_64 #1 SMP Wed Dec 17 01:55:02 UTC 2014 x86_64) MySQL: 5.1.49-log Server: LiteSpeed safe_mode: 0 max_execution_time: 900 memory_limit: 512M (used: 14.3M | 14.8M) multisite: N mcrypt: Y LANG: ZipArchive::addFile: Y
Free space on disk containing Updraft's temporary directory: 474616.9 Mb
Tasks: Backup files: 1 (schedule: weekly) Backup DB: 1 (schedule: daily)
Requesting semaphore lock (fd)
Semaphore (fd) was stuck, set lock time to 2015-02-05 22:02:35
Semaphore (fd) reset to 1
Set semaphore last lock (fd) time to 2015-02-05 22:02:35
Semaphore lock (fd) complete
Backup run: resumption=0, nonce=fb2eb97fc82d, begun at=1423173755 (0s ago), job type=backup
Scheduling a resumption (1) after 300 seconds (1423174055) in case this run gets aborted
Fatal error: Cannot redeclare class UpdraftPlus_BinZip in /hsphere/local/home/aafestival/aa-festival.dk/wp-content/plugins/updraftplus/class-zip.php on line 12
Fatal error: Cannot redeclare class UpdraftPlus_BinZip in /hsphere/local/home/aafestival/aa-festival.dk/wp-content/plugins/updraftplus/class-zip.php on line 12
I think I saw something like this once before – if I recall correctly, the cause was a buggy PHP opcode cacheing module. I can’t remember which one (there are several – xcache, zend, APC, eaccelerator). Have you got one installed? If so, try deactivating it and restarting whatever’s running PHP (e.g. php-fcgi).
David
Thread Starter
Squazz
(@squazz)
I have Autoptimize & W3 Total cache 🙂
Deactivating Autoptimize gave me this:
Opened log file at time: Fri, 06 Feb 2015 00:29:01 +0000 on http://aa-festival.dk
UpdraftPlus WordPress backup plugin (http://updraftplus.com): 1.9.51 WP: 4.1 PHP: 5.3.29 (Linux wsb12.surf-town.net 2.6.32-504.3.3.el6.x86_64 #1 SMP Wed Dec 17 01:55:02 UTC 2014 x86_64) MySQL: 5.1.49-log Server: LiteSpeed safe_mode: 0 max_execution_time: 900 memory_limit: 512M (used: 14.1M | 14.8M) multisite: N mcrypt: Y LANG: ZipArchive::addFile: Y
Free space on disk containing Updraft's temporary directory: 474558.3 Mb
Tasks: Backup files: 1 (schedule: weekly) Backup DB: 1 (schedule: daily)
Requesting semaphore lock (fd)
Semaphore lock (fd) failed (line 41)
Failed to gain semaphore lock (fd) - another backup of this type is apparently already active - aborting (if this is wrong - i.e. if the other backup crashed without removing the lock, then another can be started after 3 minutes)
Hi,
Those aren’t opcode cachers – an opcode cache is at a lower level – part of the PHP engine on the webserver. If it’s not your webserver, you’ll need to ask your web hosting company what they’re using.
David
Thread Starter
Squazz
(@squazz)
Oh, ok.
Then I’m pretty shure they are using varnish cache 🙂
That’s also something different… that’s an HTTP cache; a PHP opcode cacher is one of these: https://en.wikipedia.org/wiki/PHP_accelerator – here’s a list of them: https://en.wikipedia.org/wiki/List_of_PHP_accelerators
Thread Starter
Squazz
(@squazz)
Oh, my bad 🙂
I’ll contact them then. But the support is Sloooow, it might take some time
Thread Starter
Squazz
(@squazz)
They are using Zend OPcache
Can you ask them to temporarily disable it, so that you can verify that the problem goes away when it’s turned off? As I say, the only time I’ve ever seen this before, IIRC, it was a buggy opcache that was the problem.
