Question

Email (IMAP) stop working after 7 days


Badge

We load in FME Server 2017 backup from FME Server 2014. Specified Email (IMAP) publications was validated and it was ok. After 7 days FME Server stop getting mails and we have to push validate again. Is there any configuration for this or how to make it work without stoping?


12 replies

Badge +2
Hi @martin

 

 

Do you notice anything amiss in the email log? If you navigate in the Resources to Logs > Core > Current > Publishers > email.log

 

Badge +11
Hi @martin

 

 

Do you notice anything amiss in the email log? If you navigate in the Resources to Logs > Core > Current > Publishers > email.log

 

Also @martin, Confirming the IMAP service is precisely the same and was not changed to say Office365? I've had another issue where this service stops working after some days.

 

 

Please share the log that jlutherthomas asked about.

 

Badge +1

Hi,

I'm working on this issue also. email.log doesn't give anything much (picture below). Also, IMAP service remains unchanged.

Cheers,

Zelimir

Badge +9

Hi,

I'm working on this issue also. email.log doesn't give anything much (picture below). Also, IMAP service remains unchanged.

Cheers,

Zelimir

Hi @zzupljanin,

 

Would you mind sending the imap.log? It's in the same folder as the email.log.

 

Thanks,

 

Sienna

 

Badge +1
Hi @zzupljanin,

 

Would you mind sending the imap.log? It's in the same folder as the email.log.

 

Thanks,

 

Sienna

 

Hi, Sienna

 

 

log is provided szgdds3-imap.txt

 

I've hidden e-mail address for security reasons.

 

 

 

Badge +9
Hi, Sienna

 

 

log is provided szgdds3-imap.txt

 

I've hidden e-mail address for security reasons.

 

 

 

Hello @zzupljanin,

 

 

It could potentially be occurring on the side of the email provider. Could you talk to whoever manages the email accounts and see if they can find anything on the email side which may say if/why they're disallowing the connection?

 

 

How often are you polling the imap service? If you make it less frequent will it still stop after 7 days?

 

 

Thank you for sending the log. It looks normal, would you mind looking through the old logs when the configuration stopped working and see if there is more info on that day? Were there any errors in the log?

 

 

Thanks,

 

Sienna
Badge +1
Hello @zzupljanin,

 

 

It could potentially be occurring on the side of the email provider. Could you talk to whoever manages the email accounts and see if they can find anything on the email side which may say if/why they're disallowing the connection?

 

 

How often are you polling the imap service? If you make it less frequent will it still stop after 7 days?

 

 

Thank you for sending the log. It looks normal, would you mind looking through the old logs when the configuration stopped working and see if there is more info on that day? Were there any errors in the log?

 

 

Thanks,

 

Sienna

Yeah, sorry for not replying. I couldn't find anything wrong in logs from days later. It seems that this problem randomly occurs. It took some time (few months) since last occurence.

 

We've contacted e-mail administrator. They didn't touch anything on their side for few years.

 

Service is polling every 1 minute. We can't go lower than that.

 

Just an idea: If we delete this IMAP service and create new one (with same parameters) from scratch... can it work? Gonna try that and see what happens next few weeks.

I'd appreciate any new idea

 

Cheers.

 

Badge +12

I can't answer the "why" in relation to the Imap issue, but this is similar to the issue we had with a network directory watching at a client, that was caused by a network disconect due to something (e.g. backup, etc). Again, it was fixed by validating. So we validate Notifications (via schedule) on all our FME Server and our clients FME Servers at least daily. And we remove any websocket connections created by topic monitoring. Attached is the FME Workspace...with the end removed, so you can decide how you notify yourself (we utilise SMS).

We also monitor our abiity to send and receive emails from FME Server, but that is a far more complex process, and not one that I can share.

notifications.zip

Badge +9

Yeah, sorry for not replying. I couldn't find anything wrong in logs from days later. It seems that this problem randomly occurs. It took some time (few months) since last occurence.

 

We've contacted e-mail administrator. They didn't touch anything on their side for few years.

 

Service is polling every 1 minute. We can't go lower than that.

 

Just an idea: If we delete this IMAP service and create new one (with same parameters) from scratch... can it work? Gonna try that and see what happens next few weeks.

I'd appreciate any new idea

 

Cheers.

 

Hello @zzupljanin,

I'm sorry you are experiencing this issue again. I would try recreating the IMAP service and trying again. Or you could try what Todd suggested above. If you are still encountering an issue, could you please open a support case? https://www.safe.com/support/report-a-problem/

Kind Regards,

Sienna

Badge

I'm having a similar issue too with FME Server 2018.1.1 in which it will stop polling an IMAP e-mail address. This is a huge issue and I hope it can be fixed soon as I have filed a report with Safe about it.

Badge +1

I can't answer the "why" in relation to the Imap issue, but this is similar to the issue we had with a network directory watching at a client, that was caused by a network disconect due to something (e.g. backup, etc). Again, it was fixed by validating. So we validate Notifications (via schedule) on all our FME Server and our clients FME Servers at least daily. And we remove any websocket connections created by topic monitoring. Attached is the FME Workspace...with the end removed, so you can decide how you notify yourself (we utilise SMS).

We also monitor our abiity to send and receive emails from FME Server, but that is a far more complex process, and not one that I can share.

notifications.zip

Hi, this is really nice.

 

Another same problem occurred recently. We're considering applying your workaround.

 

One question though:

 

On last step, I see you're testing if http response is 204. If it is, all is ok. If it's not then you send SMS.

Why are you testing 204 and not 200?

 

I'm only getting 200 response, which does seem that is ok, but your tester sends this to send SMS portion.

 

Regards,

Želimir

Badge +12

Hi, this is really nice.

 

Another same problem occurred recently. We're considering applying your workaround.

 

One question though:

 

On last step, I see you're testing if http response is 204. If it is, all is ok. If it's not then you send SMS.

Why are you testing 204 and not 200?

 

I'm only getting 200 response, which does seem that is ok, but your tester sends this to send SMS portion.

 

Regards,

Želimir

Hi @zzupljanin,

I'm on holiday at the moment, so no don't have FME software close at hand, but I looked at the process in a texteditor. I believe the tester that references an http status 204 follows httpcaller_7. That httpcaller is a "Put" and appears to be working on updating a subscriber.

From the documentation a 204 is a success for this call...https://demos-safe-software.fmecloud.com/fmerest/apidoc/v3/#!/notifications/put_put_28

Cheers,

Todd

Reply