Wrong date in Scheduler and "no IMAP connection"

[rickcrites]

When I check the scheduler for checking inbound emails the detail view shows that the "last successful completion" was 12/31/1969 07:00PM. The lower part of the screen shows hundreds of attempts, all with the same wrong time.

And, every time it runs, my sugarcrm.log file gets filled with error messages for EACH of my 100+ users, saying that the system is failing the job with:

Tue Nov 7 04:40:14 2006,009 [17956] FATAL SugarCRM - SCHEDULERS: could not get an IMAP connection resource for ID [ aa0dfb0f-c5b0-2381-73d6-4548d76f2509 ]. Skipping mailbox [ Avery Parker ].

EDIT: By the way, NONE of my email accounts are set to IMAP. Every one is set to POP3, no SSL.

Any work-around for either of these?

Any plans to fix them?

Any hope we will have a working, robust email system in Sugar any time soon?

Thanks,

[rickcrites]

I found the cause of the unavailable IMAP connection. So, emails are being successfully harvested for my users.

But, I still have the wrong date reporting in the "Last Successful Completion" part of the "check inbound emails" scheduler item.

In the sugarcrm.log file I am also getting the following:

Wed Nov 8 20:46:01 2006,032 [26132] FATAL SugarCRM - -----> Scheduler got a bad execute time: [ 1970-01-01 00:00:00 GMT-0 ]

I have checked to be sure that the server has the correct time. It does. My local computer does as well.

What could be going on here? Any ideas?

Thanks.

[clint]

Hi Rick -

Trying to reproduce the error with wrong dates being displayed in the scheduler detail view. Haven't been able to reproduce it yet.

What are your date and time display formats?

Clint

[rickcrites]

Hi Rick -

Trying to reproduce the error with wrong dates being displayed in the scheduler detail view. Haven't been able to reproduce it yet.

What are your date and time display formats?

Clint

In config.php, the default date format and datef are both m/d/Y. Default time format and timef are both h:iA.

Thanks for bringing this to my attention. I just checked and found that the "check email" scheduler routine has apparently not been running / working for about three days. It apparently was working for about 2 weeks since the 9th of November, and then suddenly quit about last Friday.

To try to debug it, I checked my cron settings on the server. That was fine. I opened that scheduler job in edit view to see if I could see anything unusual. There were no job logs showing at all. I saved and closed that screen and went back to typing this message. Now I find that there is a scheduler job for checking email "In Progress" according to the job log. And, some but not all of my email has been harvested into Sugar. Now, several minutes later, refreshing the screen shows the job still "In Progress" and several email still not POP'ed into Sugar. I'm thinking the scheduler routine is getting "stuck" somewhere, somehow.

The execute time on the job that is "In Progress" is 12/31/1969. The DetailView screen also shows Last Successful Run 12/31/1969 07:00PM .

Now, a few minutes later, after opening the email harvester job to EditView again, I how have two jobs in the job log, both of which say "In Progress." But, there are still un-harvested emails on the mail server.

All of this is quite puzzling to me. I hope it gives you enough data to get an idea what might be wrong.

Here are some error messages from the sugarcrm.log file:

Mon Nov 20 11:15:59 2006,773 [30433] FATAL SugarCRM - If you see a whole string of these, there is a chance someone is attacking your system.
Mon Nov 20 14:55:15 2006,630 [17643] FATAL SugarCRM - MySQL error 1153: Got a packet bigger than 'max_allowed_packet' bytes
Mon Nov 20 14:56:01 2006,185 [19788] FATAL SugarCRM - -----> Scheduler got a bad execute time: [ 1970-01-01 00:00:00 GMT-0 ]

I am getting a long string of the Scheduler bad execute time message for the past several days.

I don't care so much what the job log says, or what data shows up in the Detail view of the scheduler job, as long as it is POPing my email as it should. But, on the other hand, I want to help you get it figured out and fixed. It certainly doesn't seem to be stable and robust at the moment.

Any ideas?

Best,

[rickcrites]

Several people have asked me privately how I solved the FATAL ERROR - "no IMAP connection" problem described above.

I think I discovered that my settings in the [mbstring] section of php.ini on the server were wrong. The passwords in the user files were being mis-handled as a result, and the scheduler showed an error on every user.

There is an entry in the WIKI here that describes how the server should be set up. Read and follow this carefully.

That might or might not be your problem. But, that was what fixed this for me.

Hope it helps.

Best,

[Skipidar]

I've configured mbstring but still have the problem with FATAL in log... Maybe there is another way to fix that?
I got version 4.5.0f

[rickcrites]

I was able to solve the problem of the wrong date when I discovered that there was no campaign "end date" in the advanced settings in the scheduler item. Putting a date there caused the bad date to disappear and the current correct date to show up. Perhaps SugarCRM could put a test in the code to be sure there is a date there, so that this doesn't occur with others.

Still intermittently having the "no IMAP connection" problem. My hosting company suggests that I use "localhost" as the SMTP server address, instead of mail.[mysugardomain].com. I am going to try this out, but don't want to go back through 100+ accounts to change that setting. I will report anything I find on this.

Best,

[sadek]

Rick,

Please keep me posted on your findings. I will file bug(s) as necessary to prevent this from happening again.

Thanks

[sadek]

Hey Rick,

I just wanted to follow up. Any additional info for us?

[rickcrites]

Hey Rick,

I just wanted to follow up. Any additional info for us?

Thanks for checking.

I am still getting an occasional one of these:

Sun Dec 17 16:50:11 2006,985 [32305] FATAL SugarCRM - SCHEDULERS: could not get an IMAP connection resource for ID [ 1dca0cfc-8703-d88a-3350-44df5c5c222f ]. Skipping mailbox [ Cheryl Acton ].

But there doesn't seem to be a pattern to it. Maybe one or two per day, and not any one particular user.

The wrong date issue was solved as noted above.

Thanks for your help.