task queue not running after upgrading from 11.(B?) (alpha-11) to A12.D (head)

[omega8cc]

task queue not running after upgrading from 11.(B?) (alpha-11) to A12.D (head)

[omega8cc]

When you upgraded main Aegir install (Barracuda), the new root dir of hostmaster is located in /var/aegir/host_master/001/. Please check how your cron entry looks like for aegir user in /var/spool/cron/crontabs/aegir. You can also try this:

$ su -s /bin/bash - aegir

$ drush '@hostmaster' hosting-dispatch -d

And pastebin your output.

[erutan]

I rolled back to alpha 11 with the provisioning patches, so my output will be for that (not the a12.d that wasn't running the task queue). note: edited out actual domain.

drush '@hostmaster' hosting-dispatch -d
Drush bootstrap phase : _drush_bootstrap_drupal_root() [0.08 sec, [bootstrap]
6.96 MB]
Loading drushrc "/var/aegir/hostmaster-0.4-alpha11/drushrc.php" into [bootstrap]
"drupal" scope. [0.08 sec, 6.96 MB]
Initialized Drupal 6.19 root directory at [notice]
/var/aegir/hostmaster-0.4-alpha11 [0.09 sec, 8.19 MB]
Drush bootstrap phase : _drush_bootstrap_drupal_site() [0.1 sec, 8.19[bootstrap]
MB]
Initialized Drupal site aegir.example.edu at [notice]
sites/aegir.example.edu [0.1 sec, 8.19 MB]
Loading drushrc [bootstrap]
"/var/aegir/hostmaster-0.4-alpha11/sites/aegir.example.edu/drushrc.php"
into "site" scope. [0.1 sec, 8.19 MB]
Drush bootstrap phase : _drush_bootstrap_drupal_configuration() [0.12[bootstrap]
sec, 8.19 MB]
Use of undefined constant HTTP_HOST - assumed 'HTTP_HOST' [notice]
global.inc:6 [0.12 sec, 8.24 MB]
Drush bootstrap phase : _drush_bootstrap_drupal_database() [0.14 sec,[bootstrap]
8.21 MB]
Successfully connected to the Drupal database. [0.14 sec, 8.21 MB] [bootstrap]
Drush bootstrap phase : _drush_bootstrap_drupal_full() [0.16 sec, [bootstrap]
8.59 MB]
Found command: hosting-dispatch (commandfile=hosting) [0.27 sec, 18.2[bootstrap]
MB]
Initializing drush commandfile: drush_make [0.27 sec, 18.21 MB] [bootstrap]
Initializing drush commandfile: drush_make_d_o [0.28 sec, 18.21 MB] [bootstrap]
Initializing drush commandfile: provision [0.28 sec, 18.21 MB] [bootstrap]
Load alias @platform_hostmaster [0.28 sec, 18.23 MB] [notice]
Load alias @server_master [0.28 sec, 18.24 MB] [notice]
Loading mysql driver for the db service [0.28 sec, 18.26 MB] [notice]
Loading nginx driver for the http service [0.28 sec, 18.27 MB] [notice]
Including profiles/hostmaster/modules/hosting/dispatch.hosting.inc [bootstrap]
[0.28 sec, 18.28 MB]
dispatching queues [0.29 sec, 18.33 MB] [notice]
too early for queue cron [0.3 sec, 19.92 MB] [notice]
too early for queue tasks [0.3 sec, 19.92 MB] [notice]
Command dispatch complete [0.3 sec, 19.87 MB] [notice]
Timer Cum (sec) Count Avg (msec)
page 0.184 1 183.86

Peak memory usage was 20.1 MB [0.3 sec, 19.87 MB] [memory]

[omega8cc]

This "too early for queue cron" suggest that you have still cron semaphore in the variables table not released for some reason (broken task, timeout etc). The solution is to delete the semaphore using in the hostmaster site: drush vdel hosting_queue_cron_running. If that will not help, check if you have correct system date/time.

[erutan]

I've run ntpdate on the server before - aside from right after restoring from a snapshot it was never more than a few seconds off.

Trying the latest barracuda update seemed to have worked - I've got an error free upgrade and the queue is firing off.

[erutan]

I'm pretty sure this is just related to my firewall issues. I'll close this for now.

[rovo79]

@omega8cc

I am running into this same issue.

Cron Queues Last Run 6 days ago

Aegir Hosting System v.A14.B.dev

Output:

Drush bootstrap phase : _drush_bootstrap_drupal_root() [0.5 sec, 7.61 MB] [bootstrap]
Loading drushrc "/var/aegir/host_master/002/drushrc.php" into "drupal" scope. [0.53 sec, [bootstrap]
7.61 MB]
Initialized Drupal 6.19 root directory at /var/aegir/host_master/002 [0.62 sec, 8.85 MB] [notice]
Drush bootstrap phase : _drush_bootstrap_drupal_site() [0.63 sec, 8.84 MB] [bootstrap]
Initialized Drupal site 184-106-227-137.static.cloud-ips.com at [notice]
sites/184-106-227-137.static.cloud-ips.com [0.65 sec, 8.84 MB]
Loading drushrc [bootstrap]
"/var/aegir/host_master/002/sites/184-106-227-137.static.cloud-ips.com/drushrc.php" into
"site" scope. [0.65 sec, 8.84 MB]
Drush bootstrap phase : _drush_bootstrap_drupal_configuration() [0.66 sec, 8.85 MB] [bootstrap]
Drush bootstrap phase : _drush_bootstrap_drupal_database() [0.8 sec, 8.86 MB] [bootstrap]
Successfully connected to the Drupal database. [0.8 sec, 8.86 MB] [bootstrap]
Drush bootstrap phase : _drush_bootstrap_drupal_full() [0.92 sec, 9.24 MB] [bootstrap]
Found command: hosting-dispatch (commandfile=hosting) [1.49 sec, 18.2 MB] [bootstrap]
Initializing drush commandfile: drush_make [1.49 sec, 18.21 MB] [bootstrap]
Initializing drush commandfile: drush_make_d_o [1.49 sec, 18.21 MB] [bootstrap]
Initializing drush commandfile: provision [1.49 sec, 18.21 MB] [bootstrap]
Load alias @platform_002 [1.49 sec, 18.23 MB] [notice]
Load alias @server_master [1.5 sec, 18.24 MB] [notice]
Loading mysql driver for the db service [1.51 sec, 18.27 MB] [notice]
Loading nginx driver for the http service [1.51 sec, 18.28 MB] [notice]
Including profiles/hostmaster/modules/hosting/dispatch.hosting.inc [1.54 sec, 18.29 MB] [bootstrap]
dispatching queues [1.59 sec, 18.34 MB] [notice]
found 0 running tasks, starting 5 out of 5 items [1.65 sec, 19.88 MB] [notice]
Forking : (/usr/local/bin/php /var/aegir/drush/drush.php --php=/usr/local/bin/php [notice]
--items='5' --quiet --root='/var/aegir/host_master/002'
--uri='184-106-227-137.static.cloud-ips.com' hosting-tasks --backend &) > /dev/null [1.66
sec, 19.88 MB]
Command dispatch complete [1.75 sec, 19.83 MB] [notice]
Timer Cum (sec) Count Avg (msec)
page 1.103 1 1102.56

Peak memory usage was 20.06 MB [1.82 sec, 19.83 MB] [memory]

[omega8cc]

@rovo79 your output says there is everything correct, no hosting_queue_cron_running blocking cron for sites.

Anyway, Aegir should really monitor it and display alert or simply release too old hosting_queue_cron_running semaphore.

[rovo79]

So everything shows as being correct?

Is there anything I can do to force the release of old hosting_queue_cron_running ?

[omega8cc]

You don't have hosting_queue_cron_running variable when you see this output.