IT:Queues/en: Perbedaan revisi
(Replaced content with "==emailouput==") |
(Importing a new version from external source) |
||
| (4 revisi antara oleh 2 pengguna tidak ditampilkan) | |||
| Baris 3: | Baris 3: | ||
The following queues are used within iSIKHNAS to transfer tasks between different elastic servers. | The following queues are used within iSIKHNAS to transfer tasks between different elastic servers. | ||
| − | ==emailinput== | + | ==emailinput== |
| + | Messages are placed on the queue by the email parsing script on the messaging server, and collected by the processing daemon on the processing server. Messages include the '''To''' and '''From''' addresses and a reference to the location (in S3) of any attached Excel file. | ||
==emailouput== | ==emailouput== | ||
| + | Messages are placed on this queue by different servers wanting to send an email. They are collected by the email sending daemon on the messaging server. Messages include the '''To''' address, subject, message content, and a reference to any attachments in S3. | ||
==reportrequest== | ==reportrequest== | ||
The report creation daemon listens to this queue and generates output reports. | The report creation daemon listens to this queue and generates output reports. | ||
| + | |||
| + | ==updatesidecar== | ||
| + | Changes made on the live system trigger a function which places the new data in this queue. A daemon on the training server inserts or updates the new data into the Sidecar database | ||
==deadletter== | ==deadletter== | ||
This is a special queue for messages which could not be processed. Normally, un-handled messages are placed back on the queue so another server can process them. However, after 20 attempts, if the message has not been successfully processed, there is either a problem with the message format or with the server. In this case the message is placed in the deadletter queue. Messages from the deadletter queue are reported regularly to administrators for debugging purposes. | This is a special queue for messages which could not be processed. Normally, un-handled messages are placed back on the queue so another server can process them. However, after 20 attempts, if the message has not been successfully processed, there is either a problem with the message format or with the server. In this case the message is placed in the deadletter queue. Messages from the deadletter queue are reported regularly to administrators for debugging purposes. | ||
Revisi terkini pada 7 November 2014 21.36
Daftar isi
Queues
The following queues are used within iSIKHNAS to transfer tasks between different elastic servers.
emailinput
Messages are placed on the queue by the email parsing script on the messaging server, and collected by the processing daemon on the processing server. Messages include the To and From addresses and a reference to the location (in S3) of any attached Excel file.
emailouput
Messages are placed on this queue by different servers wanting to send an email. They are collected by the email sending daemon on the messaging server. Messages include the To address, subject, message content, and a reference to any attachments in S3.
reportrequest
The report creation daemon listens to this queue and generates output reports.
updatesidecar
Changes made on the live system trigger a function which places the new data in this queue. A daemon on the training server inserts or updates the new data into the Sidecar database
deadletter
This is a special queue for messages which could not be processed. Normally, un-handled messages are placed back on the queue so another server can process them. However, after 20 attempts, if the message has not been successfully processed, there is either a problem with the message format or with the server. In this case the message is placed in the deadletter queue. Messages from the deadletter queue are reported regularly to administrators for debugging purposes.