Re: Improving tracking/processing of buildfarm test failures

Hello hackers,

Please take a look at the June report on buildfarm failures:
# SELECT br, COUNT(*) FROM failures WHERE dt >= '2025-06-01' AND
  dt < '2025-07-01' GROUP BY br;
REL_13_STABLE: 4
REL_14_STABLE: 3
REL_15_STABLE: 4
REL_16_STABLE: 6
REL_17_STABLE: 4
REL_18_STABLE: 1
master: 54
-- Total: 76
(Counting test failures only, excluding indent-check, Configure, Build
errors.)

# SELECT COUNT(*) FROM (SELECT DISTINCT issue_link FROM failures WHERE
  dt >= '2025-06-01' AND dt < '2025-07-01');
12

# SELECT issue_link, COUNT(*) FROM failures WHERE dt >= '2025-06-01' AND
  dt < '2025-07-01' GROUP BY issue_link ORDER BY 2 DESC LIMIT 5;
https://www.postgresql.org/message-id/fbc5d94e-6fbd-4a64-85d4-c9e284a58eb2%40gmail.com : 22
-- Fixed

https://www.postgresql.org/message-id/CAAKRu_ZCOzQpEumLFgG_%2Biw3FTa%2BhJ4SRpxzaQBYxxM_ZAzWcA%40mail.gmail.com : 17
-- Fixed

https://www.postgresql.org/message-id/16098.1745079444%40sss.pgh.pa.us : 5
-- Fixed

https://www.postgresql.org/message-id/dbf561f7-465e-4086-adfa-733b9b9a34b3@gmail.com : 3
-- Fixed

https://www.postgresql.org/message-id/Slot's : 3
-- Fixed

# SELECT COUNT(*) FROM failures WHERE dt >= '2025-06-01' AND
  dt < '2025-07-01' AND issue_link IS NULL; -- Unsorted/unhelpful failures
18

Short-lived failures:
5

Best regards,
Alexander