michalantkowicz commented on issue #13752:
URL: https://github.com/apache/iceberg/issues/13752#issuecomment-3163892707

   @nferrario I tested it carefeuly (hope it's last update 😓 ) and it looks 
that with 1 pod I don't observe data loss - while with two of them data gap 
remained after the restart now it's indeed reprocessed:
   
   
   ```
   -- pod is unhealthy, "java.lang.IllegalStateException: Connection pool shut 
down" logs are appearing:
   
   bucket_start           event_count_kafka     event_count_iceberg  diff    
diff_percent
   2025-08-07 10:50:00    21265                 21265                0       
100.0
   2025-08-07 10:52:00    18392                 18392                0       
100.0
   2025-08-07 10:54:00    21523                 21523                0       
100.0
   2025-08-07 10:56:00    19667                 19667                0       
100.0
   2025-08-07 10:58:00    18558                 18558                0       
100.0
   2025-08-07 11:00:00    23561                 23561                0       
100.0
   2025-08-07 11:02:00    19303                 19303                0       
100.0
   2025-08-07 11:04:00    20151                 20151                0       
100.0
   2025-08-07 11:06:00    19417                 18819                598     96
   2025-08-07 11:08:00    18888                 9613                 9275    50
   2025-08-07 11:10:00    20246                 13564                6682    66
   2025-08-07 11:12:00    19219                 12904                6315    67
   2025-08-07 11:14:00    29560                 19855                9705    67
   2025-08-07 11:16:00    18490                 12406                6084    67
   2025-08-07 11:18:00    18522                 12427                6095    67
   2025-08-07 11:20:00    21311                 14295                7016    67
   2025-08-07 11:22:00    18620                 12493                6127    67
   2025-08-07 11:24:00    18740                 10300                8440    54 
   ```
   
   then after restart:
   ```
   -- pod is healthy, no error logs
   
   bucket_start           event_count_kafka     event_count_iceberg  diff    
diff_percent
   2025-08-07 10:50:00    21265                 21265                0       100
   2025-08-07 10:52:00    18392                 18392                0       100
   2025-08-07 10:54:00    21523                 21523                0       100
   2025-08-07 10:56:00    19667                 19667                0       100
   2025-08-07 10:58:00    18558                 18558                0       100
   2025-08-07 11:00:00    23561                 23561                0       100
   2025-08-07 11:02:00    19303                 19303                0       100
   2025-08-07 11:04:00    20151                 20151                0       100
   2025-08-07 11:06:00    19417                 19417                0       100
   2025-08-07 11:08:00    18888                 18888                0       100
   2025-08-07 11:10:00    20246                 20246                0       100
   2025-08-07 11:12:00    19219                 19219                0       100
   2025-08-07 11:14:00    29560                 29560                0       100
   2025-08-07 11:16:00    18490                 18490                0       100
   2025-08-07 11:18:00    18522                 18522                0       100
   2025-08-07 11:20:00    21311                 21311                0       100
   2025-08-07 11:22:00    18620                 18620                0       100
   2025-08-07 11:24:00    21105                 21105                0       100
   2025-08-07 11:26:00    19228                 19228                0       100
   2025-08-07 11:28:00    17477                 17477                0       100
   2025-08-07 11:30:00    22943                 22943                0       100
   ```
   
   so as for now this issue looks much much closer to the issue you've linked 🙌 
   
   interesting why I was losing data with two pods but I may investigate is 
later 
   
   do you know what is the release plan for Iceberg and when we can expect 
version 1.9.3? I saw that releases are usually happening ~once per month I 
wonder is there any other chance to have it in Aug...


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to