notifications not being sent out

Issue #198 closed
Former user created an issue

Seemingly randomly, notifications won't be triggered. I have a semi-complex work schedule that covers sending notifications to different users based on the time of day (working hours vs off hours). I then have two tiers of SLA based on the priority of the issue to send the notification out at different intervals. The issues track time correctly when you view the SLA Overview tab on them. However, sometimes the email notification is sent, some times it is not. For instance, yesterday morning we had a High priority track time in the "Mornings" SLA and it sent its notification after 30 minutes. This morning, we had another issue that was High Priority and it was tracking the time in the "Mornings" SLA, but it did not notify. It does not appear to be an email issue from what I can tell, as I can see other emails in the system and Jira logs, but I see nothing for the issues that don't notify, which leads me to believe it is something internal to the SLA plugin that is missing the trigger to notify.

Comments (21)

  1. Tuncay Senturk repo owner

    Hi,

    Are you sure that
    - issue was still in progress according to SLA?
    - issue was not in paused status?

    If yes, is there anything in the log file, could you share the logs with me?
    Regards

  2. Brett Taylor

    Yes the issue was still in progress, no it was not paused. I created two other issues this morning with the same settings and neither triggered a notification. However, yesterday there were two issues with the same settings that did. I found the "debug" setting for jira email logging and confirmed that a notification was not even created this morning. I do not know where the logs for the plugin are.

  3. Brett Taylor

    the Working Hours schedule that this falls under is MTWThF 0000-1400. Critical priority did notify this morning, High priority did not. The time appears to be tracking correctly on the issues.

  4. Tuncay Senturk repo owner

    Under JIRA installation folder, there is log directory and you will see atlassian-jira.log file

  5. Tuncay Senturk repo owner

    Hi again Brett,

    Besides log file, if you have DB access I can share an SQL which will help me understand whether the addon tried to send email or not.

    Regards
    Tuncay

  6. Brett Taylor

    my atlassian-jira.log was rotated apr 4th and is at 0 bytes now, so it doesn't show anything from the past few days for some reason. I'd be happy to give the sql query a try.

  7. Tuncay Senturk repo owner

    Hi Brett,

    There should be problem with your log (regardless of this problem), 4 days without any log for JIRA is unbelievable.
    Anyway here is the query, please replace PROJ-123 with the key for the issue that you did not get notifiaction.

    select * from AO_C5D949_TTS_ISSUE_SLA where ISSUE_KEY='PROJ-123'
    
  8. Brett Taylor

    MariaDB [jira]> select * from AO_C5D949_TTS_ISSUE_SLA where ISSUE_KEY='ESC-110'; +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ | ACTUAL_TARGET_DATE | EXPECTED_TARGET_DATE | FINISHED | ID | INDICATOR | ISSUE_ID | ISSUE_KEY | LAST_STATUS_DATE | LAST_STATUS_ID | MULTIPLE_TRANSITION | NOTIFIED | ORIGIN_DATE | ORIGIN_STATUS_ID | PAUSED | PAUSED_DURATION | RESET_DATE | SLA_ID | SUM_OF_BACK_TO_ORIGIN_STATUS | TARGET_STATUS_ID | UPDATE_COUNT | WORKING_DURATION | +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ | 2016-04-07 13:00:29 | 2016-04-06 14:31:50 | 1 | 1541 | EXCEED | 10730 | ESC-110 | 2016-04-06 10:51:50 | 10100 | 0 | 0 | 2016-04-06 10:51:50 | 1 | 0 | 0 | NULL | 5 | 0 | 10100 | 1 | 21600000 | | 2016-04-07 13:00:29 | 2016-04-06 11:22:50 | 1 | 1542 | EXCEED | 10730 | ESC-110 | 2016-04-06 10:51:50 | 10100 | 0 | 0 | 2016-04-06 10:51:50 | 1 | 0 | 0 | NULL | 6 | 0 | 10100 | 1 | 58119000 | | 2016-04-07 13:00:29 | 2016-04-06 20:31:50 | 1 | 1543 | EXCEED | 10730 | ESC-110 | 2016-04-06 10:51:50 | 10100 | 0 | 0 | 2016-04-06 10:51:50 | 1 | 0 | 0 | NULL | 7 | 0 | 10100 | 1 | 14340000 | +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ 3 rows in set (0.00 sec)

    MariaDB [jira]> select * from AO_C5D949_TTS_ISSUE_SLA where ISSUE_KEY='ESC-114'; +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ | ACTUAL_TARGET_DATE | EXPECTED_TARGET_DATE | FINISHED | ID | INDICATOR | ISSUE_ID | ISSUE_KEY | LAST_STATUS_DATE | LAST_STATUS_ID | MULTIPLE_TRANSITION | NOTIFIED | ORIGIN_DATE | ORIGIN_STATUS_ID | PAUSED | PAUSED_DURATION | RESET_DATE | SLA_ID | SUM_OF_BACK_TO_ORIGIN_STATUS | TARGET_STATUS_ID | UPDATE_COUNT | WORKING_DURATION | +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ | 2016-04-07 06:37:39 | 2016-04-07 14:31:35 | 1 | 1511 | SUCCESS | 10734 | ESC-114 | 2016-04-07 05:16:35 | 10100 | 0 | 0 | 2016-04-07 05:16:35 | 1 | 0 | 0 | NULL | 5 | 0 | 10100 | 1 | 0 | | 2016-04-07 06:37:39 | 2016-04-07 05:47:35 | 1 | 1512 | EXCEED | 10734 | ESC-114 | 2016-04-07 05:16:35 | 10100 | 0 | 0 | 2016-04-07 05:16:35 | 1 | 0 | 0 | NULL | 6 | 0 | 10100 | 1 | 4864000 | | 2016-04-07 06:37:39 | 2016-04-07 20:31:35 | 1 | 1513 | SUCCESS | 10734 | ESC-114 | 2016-04-07 05:16:35 | 10100 | 0 | 0 | 2016-04-07 05:16:35 | 1 | 0 | 0 | NULL | 7 | 0 | 10100 | 1 | 0 | +---------------------+----------------------+----------+------+-----------+----------+-----------+---------------------+----------------+---------------------+----------+---------------------+------------------+--------+-----------------+------------+--------+------------------------------+------------------+--------------+------------------+ 3 rows in set (0.00 sec)

    MariaDB [jira]>

    114 did not notify, 110 did.

  9. Tuncay Senturk repo owner

    Hi Brett,

    What's the time string you entered for the SLA Notifier? And did you define Notifier for all SLA definitions. I see that SLA (with ID 5) for issue 114 was succeeded, while the other (with ID 6) not.
    Also, you mentioned that 110 did notify, however the NOTIFIED field is 0 which means it did not. I am not sure but there should be something wrong with logs.

    Anyway, we're currently implementing 'multiple' SLA notifiers feature. I believe that you won't have problem after that.

  10. Brett Taylor
    Critical    | [Waiting for Eng.]    | [In Progress] | 2m    | Morning Off Hours UTC     
    Critical    | [Waiting for Eng.]      | [In Progress]   | 2m    | Nights and Weekends UTC       
    Critical    | [Waiting for Eng.]    | [In Progress] | 2m    | 10-4 Weekdays UTC     
    High    | [Waiting for Eng.]    | [In Progress] | 31m   | Morning Off Hours UTC     
    High    | [Waiting for Eng.]    | [In Progress] | 31m   | Nights and Weekends UTC       
    High    | [Waiting for Eng.]    | [In Progress] | 31m   | 10-4 Weekdays UTC
    

    10-4 Weekdays is - MTWThF 1400-2000

    Nights and Weekends is - MTWThF 2000-2359, SaSu 0000-2359

    Mornings is - MTWThF 0000-1400

    110 definitely did notify, i can see the issue was created in PagerDuty (these all send email to PagerDuty to kick off a call schedule). and 114 definitely did not notify.

  11. Tuncay Senturk repo owner

    Notification '2 minutes' before the deadline?
    So, it is normal that you did not have a notification for SLA with SUCCESS one.
    Notifier scheduler wakes up and notifies every two minutes. So, when the job runs and tries to find those records, the issue may already be gone through EXCEEDED.
    Generally getting email two minutes before deadline does not make sense. Are these values just for test or in production?

    Regards
    Tuncay

  12. Brett Taylor

    No, the SLA is 2 minute, and the notification is 1 minute before. So the notification for a critical issue should be sent out 1 minute after it is created. I wanted them to send when the SLA is breached but you don't give that option.

  13. Tuncay Senturk repo owner

    Hmm, that's the case.
    So you want to be notified after the issue is breached.

    This is one of our topics to implement but I am not sure whether it will be ready with multiple notifier option.

    Thanks for the feedback
    Tuncay

  14. Brett Taylor

    I'm going to set the notifications to 3 minutes before expiration, and then make the SLAs further out to adjust. I'd still rather just have a "notify on exceeded" option but it sounds like this should work the same for our purposes.

  15. Log in to comment