System Status - There Are One Or More Cron Tasks That Have Been Executing For More Than 60 Minutes.

[EmptyMind]

I get this notice on the dashboard. It's just always there. 

 

Cron runs, the View Tasks list appears that everything is running correctly. 

 

There is no 'long running' cron process, or orphaned php processes running on the server. 

 

Is there a log that this is checking and something is out of sync. What information is this reading? How can I 'reset' the rogue entry correctly to remove this notice. 

 

James

 

 

[RockIsland]

Removing the offending cron log from the database and manually running the cron cleared the error from the system status widget. Clearing the cache solved the support manager pro menu issue as well. All is good again. Thanks guys!

[Tyson]

Some cron task has failed to complete recently.

 

To see which tasks are failing, you can check the logs for them:

SELECT * FROM `log_cron` WHERE `run_id` !=0 AND `end_date` IS NULL ORDER BY `start_date` DESC;

Check the `run_id` with the `cron_task_runs`.`id`, and then the `cron_task_runs`.`task_id` with `cron_tasks`.`id` to determine the name of the cron task that is actually failing.

[EmptyMind]

OK, So now that I've tracked down that its the 'provision pending services' task thats 'failing' (without any notice thereof anywhere else), How do I find out.. WHY. 

 

Especially given that, this instance is not yet 'live' and there are NO active services. :|

 

OK, so i found one 'pending' service. A test shared account which was set up prior to my enabling the cron, and the test server no longer is active, so it is continually trying to setup the account on a now non-existent server. 

 

Perhaps you should think about a quick 'status' check on the 'view cron tasks' page. If you have detected a problem, one should not require going into the mysql databases to determine what it is. Marking the 'run time' in red with a status that it did not finish correctly, for example.

 

Also, if a task is failing, perhaps having that information available.. you know.. an error entry like  'Error provisioning service id <num>: 'server not found'' or something somewhere that could be referenced? 

 

I've currently deleted the 'pending service'. DO I need to manually delete the rows in the log as referenced above to clear that error? Or will it now autocorrect?

 

James

 

Edited for clarity. 

[Tyson]

You don't need to delete the log. It will try again 6 hours from the last time it failed, and then it will be back on its normal schedule.

 

We plan to have a page that lists failed tasks, why (if available), with an option to retry them. This is part of a 2-step process, of which the first step was to at least tell you that something is not running correctly by way of the System Status plugin.

[dait]

I have received this message today. My case is about Support Manager's plugin that checks IMAP for new tickets. The odd thing is that it stopped working in the middle of the night (it is 7:20 AM here and it is reported that it failed at 3:30 AM). My understanding is that this has something to do with unhandled exeption conditions or unhandled error states so that CRON task never finishes. It would be nice to have more info from Blesta in some kind of debug log - what exactly happened and why. In other words, I agree with EmptyMind.

[JoieDeMort]

I agree. Had that error too. Also agree that going into mysql tables to figure out what's the issue is unacceptable.

[Michael]

it disappears soon we had it some time and vanished itself

[Paul]

it disappears soon we had it some time and vanished itself

 

The lock is cleared after 6 hours.. and the task will start to be executed again.

 

As Tyson mentioned, the query in this thread is meant to help identify which tasks are failing and we plan to implement this in the interface so that running the query won't be necessary. Most people don't experience this issue at all, so this tool was fairly low priority, and the lock is self cleared.

[Michael]

The lock is cleared after 6 hours.. and the task will start to be executed again.

 

As Tyson mentioned, the query in this thread is meant to help identify which tasks are failing and we plan to implement this in the interface so that running the query won't be necessary. Most people don't experience this issue at all, so this tool was fairly low priority, and the lock is self cleared.

 

Yeah it's weird haha I had it back in 3.0.0 and I think Dave said he had in b7 or some beta version. But it's nothing to be alarmed of as like you said mate it fixes it self 6 hours later

[Paul]

Yeah it's weird haha I had it back in 3.0.0 and I think Dave said he had in b7 or some beta version. But it's nothing to be alarmed of as like you said mate it fixes it self 6 hours later

It occurred during the beta for everyone when we added the check to the system status widget. Previously, there was a bug causing one cron task never to complete properly, and the subsequent release included a fix and the check in the widget. This resulted in everyone having that warning present for the first few hours after upgrading.

[Michael]

It occurred during the beta for everyone when we added the check to the system status widget. Previously, there was a bug causing one cron task never to complete properly, and the subsequent release included a fix and the check in the widget. This resulted in everyone having that warning present for the first few hours after upgrading.

Ah that would be why I had it because I've never had that warning since 3.0.0 b7

[Alex Vojacek]

I am having the same problem and it's on the same service the original poster had

 

Attempting to provision paid pending services.

 

I don't have the slightest idea as to how identify this service or remove this from cron.

 

I agree, looking into the mysql database is totally unaceptable, plus, there is not indication in the system as to what cron task is failing so, the user cannot do anything about it except to find this thread on google.

[RockIsland]

I'm also experiencing this issue with a fresh install of 3.2.0. I have checked the logs and no tasks are showing as uncompleted, yet this warning has been there for 2 of the 4 days it has been installed. Looking at this thread, this bug seems to have been around for a while. Is there a workaround for this bug?

[Michael]

I'm also experiencing this issue with a fresh install of 3.2.0. I have checked the logs and no tasks are showing as uncompleted, yet this warning has been there for 2 of the 4 days it has been installed. Looking at this thread, this bug seems to have been around for a while. Is there a workaround for this bug?

 

It's not really a bug and it just fixes itself. It pops up randomly.

[RockIsland]

Not a bug?  A warning of an error, where there isn't one, can't be an expected event, lol.

 

It doesn't appear to affect any operations but it sure doesn't give one a warm and fuzzy feeling to see an error like this on the software that runs my business.

I wonder now if the issue is with the check itself or with some cron task reporting as completed but actually not. It's been said above the check clears itself in 6 hours but it's been on my dashboard for the last 2 days.

[Michael]

Not a bug?  A warning of an error, where there isn't one, can't be an expected event, lol.

 

It doesn't appear to affect any operations but it sure doesn't give one a warm and fuzzy feeling to see an error like this on the software that runs my business.

I wonder now if the issue is with the check itself or with some cron task reporting as completed but actually not. It's been said above the check clears itself in 6 hours but it's been on my dashboard for the last 2 days.

 

It's not expected however it only happens to a few people, and it depends on different things:

- Imports

- Clock times don't match server

- New installations and cron hasn't run

- Cron fails and needs to wait

- other things.

[Tyson]

You should take a look at the company automation page (Settings -> Company -> Automation) and check which task has a spin icon and possibly red text signifying it has stalled. As Licensecart mentioned, it could happen for almost any reason, but depends on the cron task. We'd have to know which one is causing the problem to see what's going on. I don't think it's simply a false positive, especially when it has been recurring.

[RockIsland]

Thanks guys.

Tyson, I did indeed find the stalled task. Not sure how I missed it before. The stalled task is the Download Tickets cron. Oddly enough, there are two identical tasks showing in the list. I disabled one and am now off to check the email configs.

[Tyson]

If you have installed the Support Manager Pro 3rd-party plugin, it may conflict with the default Support Manager, and could be why you see two identical cron tasks.

 

You could also try to remove the log for that failed cron task via the database (see second post for a query to find it), and then run the cron manually through the System Automation section, which may show an error when it attempts to run the download tickets task.

[PauloV]

Thanks guys.

Tyson, I did indeed find the stalled task. Not sure how I missed it before. The stalled task is the Download Tickets cron. Oddly enough, there are two identical tasks showing in the list. I disabled one and am now off to check the email configs.

 

Can you confirm if you have both or only one Support Manager/Support Manager Pro?

[RockIsland]

Yes, I did install support manager pro and it did conflict with support manger. I have since uninstalled support manager pro but had issues with removal as well. Even though it has been uninstalled, the menu still exists but shows mysql error when clicked. Not sure if that is related or not. 

 

I did discover some mail issues(phpmailparse) with the server when I investigated, but after running the cron manually the error still exists. I'll try removing the log from the dbase next.

[Michael]

Yes, I did install support manager pro and it did conflict with support manger. I have since uninstalled support manager pro but had issues with removal as well. Even though it has been uninstalled, the menu still exists but shows mysql error when clicked. Not sure if that is related or not. 

 

I did discover some mail issues(phpmailparse) with the server when I investigated, but after running the cron manually the error still exists. I'll try removing the log from the dbase next.

 

Menu you need to clear your cache, go to your blesta install see the cache folder inside that, remove the 1 folder. 

 

Or you can re-save your staff group.

[PauloV]

Yes, I did install support manager pro and it did conflict with support manger. I have since uninstalled support manager pro but had issues with removal as well. Even though it has been uninstalled, the menu still exists but shows mysql error when clicked. Not sure if that is related or not. 

 

I did discover some mail issues(phpmailparse) with the server when I investigated, but after running the cron manually the error still exists. I'll try removing the log from the dbase next.

 

Its you Blesta Cache

 

Please go tru FTP and delete in your blesta the folder with the name "1" on /cache/ directory

 

In the next release we will add the auto remove cache code, and also a warning to POP3/IMAP import conflit with existent Support Manager. You can still use the PIPE without any conflits with both Support Manager and Support Manager Pro.

[PauloV]

Removing the offending cron log from the database and manually running the cron cleared the error from the system status widget. Clearing the cache solved the support manager pro menu issue as well. All is good again. Thanks guys!

 

what was in the cron log? just to trie to debug on our side to fix issues with Support Manager Pro?

[RockIsland]

I'm not really sure as I simply deleted the first 50 entries from the log_cron table. If it happens again I'll check through the entries and see if it gives up any info.